Whole-flow control method and device for sewage treatment

Abstract

The invention provides a sewage treatment whole-process control method and device, and relates to the technical field of sewage treatment, and the method comprises the steps: collecting multi-dimensional parameter data in a sewage treatment process through a real-time sensor; preprocessing the multi-dimensional parameter data to generate multi-scale feature data; adjusting the original language model based on the sewage treatment historical corpus data and general data in the fields of environmental science and chemistry to obtain a sewage treatment control parameter prediction model; inputting the multi-scale characteristic data into the sewage treatment control parameter prediction model to generate a sewage treatment control parameter prediction result; and controlling the sewage treatment equipment based on the sewage treatment control parameter prediction result. According to the invention, complex scene data characteristics can be fully combined and sewage treatment control parameters can be optimized in real time, so that the sewage treatment operation efficiency, the resource recovery capability and the carbon emission control effect are improved.

Description

Background Art

[0002] Current sewage treatment systems usually rely on fixed rule control and empirical parameter settings, and manually adjust the equipment operating status to meet different treatment needs. However, the sewage treatment process itself involves highly nonlinear, multivariable coupled dynamic changes. This complexity makes it difficult for traditional methods to respond flexibly when facing multiple types of water quality, such as domestic sewage and medical sewage, which can easily lead to unstable treatment efficiency or unsatisfactory results.

[0003] Looking further, with the national "dual carbon" goal, reducing carbon emissions from sewage treatment plants has become an important task. However, the existing control system lacks the ability to deeply analyze carbon emission characteristics, and can usually only achieve limited optimization through extensive adjustments in a single link. This limitation makes it impossible for the system to effectively reduce greenhouse gas emissions as a whole. At the same time, renewable resources such as residual heat energy and chemical energy contained in the sewage treatment process have not been fully utilized, resulting in widespread energy waste and failing to provide strong support for promoting carbon neutrality goals.

[0004] Moreover, the operation of sewage treatment involves complex data with multiple dimensions and strong heterogeneity, including a large number of variables including physical, chemical and biological parameters, which show significant dynamic changes over time. However, the existing technology is relatively weak in processing and analyzing these data, and can usually only make simple predictions and adjustments to single parameters or local characteristics. Especially in the face of complex scenarios such as fluctuations in influent water quality or high-load operation, the system lacks a deep understanding of the relationship between multiple variables, which not only reduces the accuracy of the prediction, but also makes it difficult to coordinate the overall control strategy, ultimately affecting the effect and stability of sewage treatment.

[0005] In summary, existing sewage treatment technologies have limitations in adaptability, real-time and optimization capabilities. These problems restrict the operating efficiency, resource recovery capacity and carbon emission control effect of sewage treatment systems. There is an urgent need for a technical solution that can fully combine the characteristics of complex scenario data and optimize control parameters in real time. Summary of the invention

[0006] In order to overcome the problems existing in the related art, the present invention provides a method and device for controlling the entire process of sewage treatment.

[0007] According to a first aspect of an embodiment of the present invention, a method for controlling the entire process of sewage treatment is provided, the method comprising: Use real-time sensors to collect multi-dimensional parameter data during sewage treatment; Preprocessing the multi-dimensional parameter data to generate multi-scale feature data; The original language model was adjusted based on the historical corpus data of sewage treatment and the common data in the fields of environmental science and chemistry to obtain the sewage treatment control parameter prediction model; Inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate sewage treatment control parameter prediction results; Based on the sewage treatment control parameter prediction result, the sewage treatment equipment is controlled.

[0008] In some example embodiments of the present invention, based on the above scheme, preprocessing the multi-dimensional parameter data to generate multi-scale feature data includes: Performing a data cleaning operation on the multidimensional parameter data to remove outliers and fill in missing data to obtain a data sequence of the multidimensional parameter data; Dividing the data sequence into data segments of fixed length according to a time window mechanism to obtain segmented data; Performing standardization processing on the segmented data to obtain a standardized feature sequence; The time series difference mechanism is used to perform first-order difference and high-order difference calculations on the standardized feature sequence to generate the multi-scale feature data.

[0009] In some example embodiments of the present invention, based on the above scheme, before inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate the sewage treatment control parameter prediction result, the sewage treatment full process control method further includes: The multi-scale feature data is subjected to feature conversion processing to generate an input feature vector of the sewage treatment control parameter prediction model.

[0010] In some example embodiments of the present invention, based on the above scheme, performing feature conversion processing on the multi-scale feature data to generate an input feature vector of the sewage treatment control parameter prediction model includes: Performing feature grouping processing on the multi-scale feature data; Assign weights to each feature group and generate weighted feature vectors using a weighted summation mechanism; Performing a nonlinear transformation on the weighted eigenvector value to obtain a nonlinear transformation result; The nonlinear transformation result is processed using a B-spline basis function to obtain an input feature vector.

[0011] In some exemplary embodiments of the present invention, based on the above scheme, based on the prediction result of the sewage treatment control parameter, controlling the sewage treatment equipment includes: Analyzing the real-time state variables of the operation of the sewage treatment equipment based on the sewage treatment control parameter prediction results; Mapping the real-time state variable and the input feature vector to generate an operation instruction suitable for the sewage treatment equipment; Send operating instructions to sewage treatment equipment.

[0012] In some exemplary embodiments of the present invention, based on the above scheme, the sewage treatment full process control method further includes: Acquiring real-time operation data of the sewage treatment equipment when the sewage treatment equipment is controlled using the sewage treatment control parameter prediction result; Calculating the difference between the input feature vector and the real-time operation data; Adjusting the weight of the input feature vector according to the difference to obtain an updated input feature vector; The updated input feature vector is used to optimize the sewage treatment control parameter prediction model.

[0013] In some exemplary embodiments of the present invention, based on the above scheme, the loss function of the sewage treatment control parameter prediction model is for: in, Indicates i The true predicted value of samples, represents the mean of the prediction results, Represents the overall offset, represents the sewage vector weight, Represents the input sewage vector data, Indicates the correction offset, Indicates the total amount of sample data.

[0014] According to a second aspect of an embodiment of the present invention, a control device based on the above-mentioned sewage treatment full-process control method is provided, comprising: A data acquisition module, which is used to collect multi-dimensional parameter data in the sewage treatment process using real-time sensors; A preprocessing module, used for preprocessing the multi-dimensional parameter data to generate multi-scale feature data; A model generation module is used to adjust the original language model based on the historical corpus data of sewage treatment and the common data in the fields of environmental science and chemistry to obtain a sewage treatment control parameter prediction model; A prediction result generating module, used for inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate a sewage treatment control parameter prediction result; The control sending module is used to control the sewage treatment equipment based on the sewage treatment control parameter prediction result.

[0015] According to a third aspect of an embodiment of the present invention, there is provided an electronic device, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the full-process control method for sewage treatment in the first aspect is implemented.

[0016] According to a fourth aspect of an embodiment of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the full-process control method for sewage treatment in the first aspect.

[0017] The technical solution provided by the embodiments of the present invention may have the following beneficial effects: In the embodiment of the present invention, by real-time collection and preprocessing of multidimensional parameter data in the sewage treatment process, the key characteristics of dynamic changes in the sewage treatment process can be effectively captured; combined with the sewage treatment control parameter prediction model adjusted based on historical corpus data, the ability to understand the dynamic process of sewage treatment can be significantly improved, and the problem of insufficient adaptation of traditional control systems to data heterogeneity and high-dimensional dynamic characteristics can be overcome, thereby achieving in-depth mining and utilization of multivariate synergistic effects and improving adaptability to diversified sewage treatment scenarios; by controlling the sewage treatment equipment through the generated control parameter prediction results, the operating status of the equipment can be adjusted in real time, and the key parameter settings in the treatment process can be optimized, so that it can flexibly respond to complex situations of water quality fluctuations and load changes.

[0018] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0020] Figure 1 A schematic diagram showing a system architecture of an exemplary application environment of a sewage treatment full-process control method and device to which an embodiment of the present invention can be applied; Figure 2 Schematically showing a flow chart of a sewage treatment full-process control method according to some embodiments of the present invention; Figure 3 Schematically showing a schematic diagram of a sewage treatment full-process control device according to some embodiments of the present invention; Figure 4 A schematic diagram schematically shows a structure of a computer system of an electronic device according to some embodiments of the present invention; Figure 5 A schematic diagram of a computer-readable storage medium according to some embodiments of the present invention is schematically shown. DETAILED DESCRIPTION

[0021] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0022] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0023] It should be understood that although the terms first, second, third, etc. may be used in the present invention to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0024] Figure 1 A schematic diagram of the system architecture of an exemplary application environment of a sewage treatment full-process control method and device to which an embodiment of the present invention can be applied is shown.

[0025] like Figure 1 As shown, the system architecture 100 may include one or more terminal devices such as a desktop computer 101, a portable computer 102, a smart phone 103, a network 104 and a server 105. The network 104 is used to provide a medium for a communication link between the terminal device and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or optical fiber cables, etc. The terminal device may be various electronic devices with data processing functions, which have a display screen, and the display screen is used to display sewage treatment process information and treatment results to the user, including but not limited to the above-mentioned desktop computers, portable computers, smart phones, etc. It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. According to the implementation requirements, there may be any number of terminal devices, networks and servers. For example, the server 105 may be a sub-server cluster composed of multiple sub-servers.

[0026] The sewage treatment process control method provided in the embodiment of the present invention can generally be executed by a terminal device, and accordingly, the sewage treatment process control device is generally arranged in the terminal device. However, it is easy for a person skilled in the art to understand that the sewage treatment process control method provided in the embodiment of the present invention can also be executed by the server 105, and accordingly, the sewage treatment process control device can also be arranged in the server 105, which is not particularly limited in this exemplary embodiment.

[0027] In addition, it should be understood that the sewage treatment full process control method of the embodiment of the present invention can be configured as a software module. In some implementation scenarios, the sewage treatment full process control scheme of the present invention can be deployed separately to generate control schemes with different control requirements. In other implementation scenarios, the sewage treatment full process control scheme of the present invention can be deployed in other software as a functional module of the software, such as deployed in sewage treatment analysis software. The present invention does not place any particular restrictions on the application of the sewage treatment full process control method.

[0028] Next, the embodiments of the present invention are described in detail.

[0029] like Figure 2 As shown, Figure 2 The present invention is a flowchart of a sewage treatment full process control method according to an exemplary embodiment, comprising the following steps: S210: Collect multi-dimensional parameter data during sewage treatment using real-time sensors; S220: preprocessing the multi-dimensional parameter data to generate multi-scale feature data; S230: adjusting the original language model based on the historical corpus data of sewage treatment and the general data in the fields of environmental science and chemistry to obtain a sewage treatment control parameter prediction model; S240: inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate a sewage treatment control parameter prediction result; S250: Controlling sewage treatment equipment based on the sewage treatment control parameter prediction result.

[0030] In the embodiment of the present invention, by real-time collection and preprocessing of multidimensional parameter data in the sewage treatment process, the key characteristics of dynamic changes in the sewage treatment process can be effectively captured; combined with the sewage treatment control parameter prediction model adjusted based on historical corpus data, the ability to understand the dynamic process of sewage treatment can be significantly improved, and the problem of insufficient adaptation of traditional control systems to data heterogeneity and high-dimensional dynamic characteristics can be overcome, thereby achieving in-depth mining and utilization of multivariate synergistic effects and improving adaptability to diversified sewage treatment scenarios; by controlling the sewage treatment equipment through the generated control parameter prediction results, the operating status of the equipment can be adjusted in real time, and the key parameter settings in the treatment process can be optimized, so that it can flexibly respond to complex situations of water quality fluctuations and load changes.

[0031] In S210, real-time sensors are used to collect multi-dimensional parameter data in the sewage treatment process.

[0032] Here, the multidimensional parameter data include but are not limited to pH value, chemical oxygen demand (COD), biological oxygen demand (BOD), B / C ratio (ratio of biological oxygen demand to chemical oxygen demand), hydraulic retention time (HRT), sludge concentration (including the mass concentration of sludge suspended solids (MLSS) and the mass concentration of volatile suspended solids (MLVSS) in sludge), sludge settling ratio (SV30), sludge index (SVI), food-to-microorganism ratio (F / M), sludge age SRT, aerobic tank (Dissolved Oxygen, DO), anoxic tank OPR, the weight ratio of organic carbon and organic nitrogen in organic matter (CN ratio), reflow ratio and other parameters.

[0033] Real-time sensors can be installed at key locations such as water inlets, treatment tanks, and water outlets to perform real-time detection of sewage characteristics and equipment operating status.

[0034] During the collection process, real-time sensors periodically record the changes in various parameters to form a time-related dynamic data sequence, which is transmitted by the sensor to the control device of the sewage treatment full-process control method via wired or wireless means, and a series of subsequent processing operations after data reception are executed in the control device of the sewage treatment full-process control method.

[0035] To further improve the accuracy and comprehensiveness of data collection, real-time sensors can be set as a number of sensor groups with noise filtering and signal enhancement functions, including water quality parameter sensors (PH sensors, dissolved oxygen sensors, redox potential sensors, etc.), organic matter and pollutant sensors (chemical oxygen demand sensors, biological oxygen demand sensors, suspended solids sensors, etc.), physical parameter sensors (water quality flow sensors, temperature sensors, etc.), sludge parameter sensors (sludge concentration sensors, sludge settling ratio sensors, etc.) and gas and volatile sensors (ammonia nitrogen sensors, nitrate nitrogen sensors, hydrogen sulfide sensors, etc.), so as to be able to stably output high-quality data reflecting the actual state of sewage in a complex environment.

[0036] In S220, the multi-dimensional parameter data is preprocessed to generate multi-scale feature data.

[0037] For the multidimensional parameter data collected by real-time sensors, a data cleaning process is first performed to identify and remove outliers. The judgment of outliers is based on preset rules, for example, numerical fluctuations or invalid readings that exceed a reasonable range are defined as outliers. On this basis, the missing data contained in the outliers are filled with an interpolation algorithm to ensure the integrity and continuity of the data sequence of the multidimensional parameter data. The interpolation algorithm can be a combination of one or more of a linear interpolation algorithm, a polynomial interpolation algorithm, and a spline interpolation algorithm, and the present invention does not make specific restrictions.

[0038] The data sequence of the multidimensional parameter data is segmented according to the time window mechanism to divide the data sequence of the entire multidimensional parameter data into multiple time segments of fixed length. Here, the length of the time window is determined according to the dynamic change frequency of the sewage treatment process, which ensures that the segmented data is sufficiently representative and avoids excessive aggregation leading to loss of details. The type of time window can be a fixed time window, a sliding time window or a dynamic time window, and the present invention is also not limited. The segmented data is stored in a structured format to facilitate standardization and feature processing.

[0039] Next, the Z-score normalization operation is performed on each parameter value in the segmented data to eliminate the influence caused by the dimension or unit difference between different parameters. After normalization, the value of each parameter in the obtained standardized feature sequence is evenly distributed within a uniform scale range, which can enhance the comparability and consistency of the data.

[0040] For the standardized feature sequence, the time series difference mechanism is used to calculate the first-order difference and high-order difference respectively to extract the dynamic change characteristics of the data. The first-order difference is used to capture the instantaneous change trend of the parameter value over time, while the high-order difference reflects a more complex change pattern, providing richer dynamic information for subsequent feature extraction.

[0041] In addition, in some embodiments, the preprocessing of the multidimensional parameter data may also include data denoising, such as denoising by using Kalman filtering, wavelet denoising or low-pass filtering; or data smoothing operations, which can be set by technical personnel in this field according to actual conditions.

[0042] In S230, the original language model is adjusted based on the historical corpus data of sewage treatment and the common data in the fields of environmental science and chemistry to obtain a sewage treatment control parameter prediction model.

[0043] Here, the sewage treatment historical corpus data includes key treatment parameters of different types of sewage (such as domestic sewage, medical sewage, and industrial sewage) and their changing patterns. The parameters covered include but are not limited to the contents of the above-mentioned multidimensional parameter data.

[0044] Before using the sewage treatment historical corpus data, the sewage treatment historical corpus data can also be cleaned and normalized to eliminate redundant information and dimensional differences, and representative core features can be extracted through feature grouping.

[0045] General data in the field of environmental science and chemistry, including but not limited to standard wastewater treatment technology literature, parameter variation patterns, and environmental data related to carbon neutrality goals.

[0046] Before adjusting the original language model based on the historical corpus data of sewage treatment and the general data in the fields of environmental science and chemistry, the general data in the fields of environmental science and chemistry can also be fused with the historical corpus data to unify the feature dimensions and representation methods to ensure the consistency of the fused data in the model. When fusion of features, the weight distribution mechanism can be used to give the historical corpus data a higher priority while retaining the global constraint capability of the general data in the form of reference data.

[0047] In addition, the knowledge of experts in fields such as environment and dual carbon can be introduced to design time series characteristics, periodic characteristics, etc.

[0048] These data can be used as reference data and together with general data, or can be used together with sewage treatment historical corpus data as primary data, and the present invention does not impose any specific limitation thereto.

[0049] The original language model can be set to a GPT model (full name Generative Pre-Trained Transformer), an LLAMA model (full name Large Language Model Meta AI), a CLIP model (full name Contrastive Language–Image Pretraining), a BERT model (full name Bidirectional Encoder Representations from Transformers), etc., and those skilled in the art can set it according to actual conditions.

[0050] In an embodiment of the present invention, the LLAMA3-8B model is selected as the original language model. Compared with the GPT-4 model, the LLAMA3-8B model has lower resource requirements and is suitable for deployment in industrial scenarios. At the same time, it has higher customization efficiency and can quickly adjust the model according to demand. Compared with the CLIP model, the LLAMA3-8B model focuses more on language and structured data scenarios, and does not need to allocate excess computing resources for visual or other modality adaptation. The DistilBERT model is a lightweight version of the BERT model. Although it has the advantage of low resource usage, it has insufficient performance when processing complex multivariate prediction tasks (such as multidimensional sewage parameter control). LLAMA3-8B can achieve a better balance between high efficiency and complex scenario performance.

[0051] In the process of fine-tuning the original language model, historical corpus data is used as the main input, and general data in the fields of environmental science and chemistry are added. The model parameters are adjusted through supervised learning so that it can more accurately adapt to the sewage treatment scenarios. A hybrid fine-tuning strategy is used to improve the model's ability to understand knowledge outside of specific fields, thereby enhancing its generalization and reasoning capabilities.

[0052] In S240, the multi-scale feature data is input into the sewage treatment control parameter prediction model to generate a sewage treatment control parameter prediction result.

[0053] Since the present invention has described the contents of the sewage treatment control parameter prediction model in detail above, therefore, inputting the multi-scale feature data into the sewage treatment control parameter prediction model and generating the sewage treatment control parameter prediction results means calculating the multi-scale feature data through the sewage treatment control parameter prediction model, and using the feature weight allocation mechanism and parameter prediction algorithm constructed inside the model to mine the correlation and dynamic change law between different parameters in the multi-scale feature data. Specifically, the sewage treatment control parameter prediction model extracts the key features that affect sewage treatment layer by layer according to the distribution of the multi-dimensional input feature vector, and uses the optimal weights and adjustment strategies obtained during the training process to predict and calculate the control parameters in the sewage treatment process.

[0054] The sewage treatment control parameter prediction result includes multi-dimensional parameter data at the next moment. Since the multi-dimensional parameter data has been stated in S210, the present invention will not repeat them again.

[0055] In some embodiments, before inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate the sewage treatment control parameter prediction result, the multi-scale feature data may be subjected to feature conversion processing to generate an input feature vector of the sewage treatment control parameter prediction model. Further, this process may include: Performing feature grouping processing on the multi-scale feature data; Assign weights to each feature group and generate weighted feature vectors using a weighted summation mechanism; Performing a nonlinear transformation on the weighted eigenvector value to obtain a nonlinear transformation result; The nonlinear transformation result is processed using a B-spline basis function to obtain an input feature vector.

[0056] When performing feature grouping processing, the parameters can be grouped according to their physical properties, process links and dynamic characteristics. For example, pH value and dissolved oxygen (DO) can be grouped into the biochemical characteristic group; chemical oxygen demand (COD) and total nitrogen (Total Nitrogen, TN) can be grouped into the chemical characteristic group; hydraulic retention time (HRT) and flow rate can be grouped into the process dynamic characteristic group. Based on this, technicians in this field can further group all the contents of the above multidimensional parameter data to clarify the scope of influence of each type of feature in the sewage treatment process and provide a basis for subsequent weight allocation.

[0057] When assigning weights to each feature group, the basis for determining the weights may include historical data statistical analysis results or domain expert experience. Then, the features in the group are weighted and summed according to the weights to generate the weighted feature vector corresponding to each feature group. The formula for the weighted summation operation is: in, For the The weighted feature vector of the group feature group, and Respectively The weights and values ​​of the features, Indicates Group Features The total number of features in the group.

[0058] Nonlinear transformation of weighted feature vector values ​​can enhance the complexity and adaptability of feature representation. Nonlinear transformation can be implemented using activation functions, such as ReLU function (Rectified Linear Unit): in, Represents the weighted eigenvector value.

[0059] In other implementations, the nonlinear transformation may also be implemented using a Sigmoid function: in, Represents the weighted eigenvector value.

[0060] On this basis, the nonlinear transformation results are further processed using the B-spline basis function to generate the input feature vector. The B-spline basis function fits the data through a piecewise polynomial, so that the input feature vector retains the global trend information and captures the local characteristic changes. The calculation formula is: in, is a B-spline basis function defined on the interval, indicating that Starting from nodes, the order is The B-spline basis function value of ; represents a sequence of nodes, Represents the result of nonlinear transformation; Indicates that the current node Go to the next impact range node The segmental contribution value of Indicates that the next node The starting range of influence is to higher nodes The segment contribution value of .

[0061] The input feature vectors obtained after processing with the B-spline basis function retain the multi-scale information and dynamic change characteristics of the features. These input feature vectors are passed to the input layer of the sewage treatment control parameter prediction model, which can further enhance the learning ability of the sewage treatment control parameter prediction model for complex multivariate relationships.

[0062] In S250, the sewage treatment equipment is controlled based on the sewage treatment control parameter prediction result.

[0063] In some embodiments, based on the sewage treatment control parameter prediction result, controlling the sewage treatment equipment includes: Analyzing the real-time state variables of the operation of the sewage treatment equipment based on the sewage treatment control parameter prediction results; Mapping the real-time state variable and the input feature vector to generate an operation instruction suitable for the sewage treatment equipment; Send operating instructions to sewage treatment equipment.

[0064] The real-time state variables of sewage treatment equipment operation include key parameters that need to be adjusted during the operation of sewage treatment equipment, which are directly related to the process control and equipment operation of sewage treatment. For example, aeration equipment parameters (including the operating frequency and ventilation volume of the blower or aeration head, etc.), sludge treatment equipment parameters (including the flow rate of the sludge return pump or the operating time of the concentration and dehydration equipment, etc.), water flow control equipment parameters (including adjusting the opening of the water inlet valve and the water outlet valve, adjusting the operating speed of the water pump, etc.), chemical dosing equipment parameters (including optimizing the dosing system, such as the amount of oxidant or coagulant added), reflux control parameters (including the setting of the sludge return pump, such as start and stop time and flow adjustment, etc.), temperature control equipment parameters (including start and stop control that can be used for heating equipment or cooling equipment) and effluent control equipment parameters (operating status of the filter or deep treatment device, etc.).

[0065] Mapping the real-time state variable and the input feature vector to generate an operation instruction suitable for the sewage treatment equipment includes: According to the equipment type and operation logic, the real-time state variables are associated with the input feature vector, and each feature in the feature vector (such as DO, COD, MLSS, etc.) is mapped to parameter adjustment values ​​that can be executed by the equipment, such as the air supply of the aeration equipment, the flow rate of the reflux pump, etc.

[0066] The mapping model of equipment control is used to convert the parameter change value in the input feature vector into the specific operation value of the sewage treatment equipment. In some embodiments, the mapping model can be expressed as: in, For sewage treatment equipment The specific operation value of is the mapping weight, are the relevant parameter values ​​in the input feature vector, It is the operating benchmark value of sewage treatment equipment.

[0067] After that, the specific operation values ​​of the sewage treatment equipment are converted into control instructions, such as adjusting the blower frequency to 60Hz or setting the return pump flow rate to 100L / min. After these control instructions are sent to the sewage treatment equipment, the sewage treatment equipment is driven to dynamically adjust the operating status to ensure the optimization and stability of the treatment process. This process realizes a closed loop of prediction, analysis, control and verification, providing intelligent support for the whole process of sewage treatment.

[0068] In another embodiment of the present invention, when the sewage treatment equipment receives the control instruction and adjusts the operating state, the real-time operating data of the sewage treatment equipment is collected through real-time sensors. The real-time operating data here includes all the contents in the above-mentioned sewage treatment control parameter prediction results, which are not elaborated in the present invention.

[0069] Then, the difference between the input feature vector and the real-time running data is calculated using the formula: in, Indicates The difference of the features, Indicates the first eigenvalues, Indicates the corresponding actual value in real-time running data.

[0070] On this basis, the weight of each feature in the input feature vector is adjusted based on the difference size, and the weight of the feature with large deviation is increased to enhance the model's attention to it. The weight adjustment formula is: in, Indicates The updated weights of the features, represents the original weight value, represents the adjustment factor, Indicates The difference of features.

[0071] Recompute the input feature vector according to the adjusted weights: in, Indicates The updated feature value of features, Indicates The original feature value of each feature.

[0072] Finally, the updated input feature vector is used as new training data and re-input into the sewage treatment parameter prediction model. The model adjusts internal parameters through adaptive learning to adapt to real-time data changes.

[0073] Loss function of the sewage treatment control parameter prediction model of the present invention for: in, Indicates i The true predicted value of samples, represents the mean of the prediction results, Represents the overall offset, represents the sewage vector weight, Represents the input sewage vector data, Indicates the correction offset, Indicates the total amount of sample data. is obtained by calculating the average value of the training data, the overall offset and corrected offset Is the experience value, offset Generally take 0.5 to correct the offset Generally, it is taken as 0.08. The initial weight of the sewage vector weight is determined based on more than 100 types of data connected to the system. The present invention sets it as follows: the initial weight of the status class is 0.1, and the initial weight of the alarm class is 0.3; however, the value of this vector weight continues to change during the training process, and the final sewage vector weight is given by calculation.

[0074] By adaptively adjusting the model parameters and dynamically controlling the sewage treatment equipment in real time, it is not only possible to reduce operating losses, reduce electricity consumption, and reduce carbon emissions; it is also possible to recover a portion of chemical resources from sewage or residual sludge, generate certain economic benefits, and reduce operating costs. In addition, due to the chemical energy contained in the sewage system and the huge waste heat energy of sewage, the carbon neutrality of the sewage treatment system can be achieved by adjusting the parameters of photovoltaic power generation and water source heat pumps through a large model, and energy can be exported to the outside to help achieve carbon neutrality in urban water systems.

[0075] According to a second aspect of an embodiment of the present invention, a control device for a whole-process control method for sewage treatment is also provided. Figure 3 As shown, the control device 300 of the sewage treatment full process control method includes: The data acquisition module 310 is used to collect multi-dimensional parameter data in the sewage treatment process using real-time sensors; A preprocessing module 320, used to preprocess the multi-dimensional parameter data to generate multi-scale feature data; A model generation module 330 is used to adjust the original language model based on the historical corpus data of sewage treatment and the general data in the fields of environmental science and chemistry to obtain a sewage treatment control parameter prediction model; A prediction result generating module 340 is used to input the multi-scale feature data into the sewage treatment control parameter prediction model to generate a sewage treatment control parameter prediction result; The control sending module 350 is used to control the sewage treatment equipment based on the sewage treatment control parameter prediction result.

[0076] It should be noted that, although several modules of the sewage treatment full process control device are mentioned in the above detailed description, this division is not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules or submodules described above can be concretized in one module or unit. Conversely, the features and functions of one module or submodule described above can be further divided into multiple modules or submodules to be concretized.

[0077] In addition, in an exemplary embodiment of the present invention, an electronic device capable of implementing a full-process control method for sewage treatment is also provided.

[0078] Those skilled in the art will appreciate that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware embodiments, complete software embodiments (including firmware, microcode, etc.), or embodiments combining hardware and software aspects, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0079] Refer to the following Figure 4 An electronic device 400 according to such an embodiment of the present invention is described. Figure 4 The electronic device 400 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0080] like Figure 4 As shown, the electronic device 400 is in the form of a general computing device. The components of the electronic device 400 may include but are not limited to: the at least one processing unit 410, the at least one storage unit 420, a bus 430 connecting different system components (including the storage unit 420 and the processing unit 410), and a display unit 440.

[0081] The storage unit stores program codes, which can be executed by the processing unit 410, so that the processing unit 410 performs the steps of various exemplary embodiments of the present invention described in the above “Exemplary Method” section. For example, the processing unit 410 can perform the following steps: Figure 2S210 shown in: using real-time sensors to collect multi-dimensional parameter data in the sewage treatment process; S220: pre-processing the multi-dimensional parameter data to generate multi-scale feature data; S230: adjusting the original language model based on the historical corpus data of sewage treatment, and general data in the fields of environmental science and chemistry to obtain a sewage treatment control parameter prediction model; S240: inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate a sewage treatment control parameter prediction result; S250: controlling the sewage treatment equipment based on the sewage treatment control parameter prediction result.

[0082] The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 421 and / or a cache memory unit 422 , and may further include a read-only memory unit (ROM) 423 .

[0083] The storage unit 420 may also include a program / utility 424 having a set (at least one) of program modules 425, such program modules 425 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0084] Bus 430 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0085] The electronic device 400 may also communicate with one or more external devices 470 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 400, and / or communicate with any device that enables the electronic device 400 to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication may be performed through an input / output (I / O) interface 450. In addition, the electronic device 400 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 460. As shown, the network adapter 460 communicates with other modules of the electronic device 400 through a bus 430. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0086] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the exemplary embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present invention.

[0087] In an exemplary embodiment of the present invention, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present invention is stored. In some possible embodiments, various aspects of the present invention can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to perform the steps of various exemplary embodiments of the present invention described in the above "Exemplary Method" section of the present invention.

[0088] refer to Figure 5 As shown, a program product 500 for implementing the above-mentioned sewage treatment full process control method according to an embodiment of the present invention is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In the present invention, the readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0089] The program product may use any combination of one or more readable storage media. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0090] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0091] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0092] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the exemplary embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiment of the present invention.

[0093] Other embodiments of the invention will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. The invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the claims.

[0094] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A sewage treatment full process control method, characterized in that: include: Use real-time sensors to collect multi-dimensional parameter data during sewage treatment; Preprocessing the multi-dimensional parameter data to generate multi-scale feature data; The original language model was adjusted based on the historical corpus data of sewage treatment and the common data in the fields of environmental science and chemistry to obtain the sewage treatment control parameter prediction model; Inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate sewage treatment control parameter prediction results; Based on the sewage treatment control parameter prediction result, the sewage treatment equipment is controlled.

2. The whole process control method for sewage treatment according to claim 1 is characterized in that: Preprocessing the multi-dimensional parameter data to generate multi-scale feature data includes: Performing a data cleaning operation on the multidimensional parameter data to remove outliers and fill in missing data to obtain a data sequence of the multidimensional parameter data; Dividing the data sequence into data segments of fixed length according to a time window mechanism to obtain segmented data; Performing standardization processing on the segmented data to obtain a standardized feature sequence; The time series difference mechanism is used to perform first-order difference and high-order difference calculations on the standardized feature sequence to generate the multi-scale feature data.

3. The whole process control method for sewage treatment according to claim 1 is characterized in that: Before inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate the sewage treatment control parameter prediction result, the sewage treatment full process control method further includes: The multi-scale feature data is subjected to feature conversion processing to generate an input feature vector of the sewage treatment control parameter prediction model.

4. The whole process control method for sewage treatment according to claim 3 is characterized in that: Performing feature conversion processing on the multi-scale feature data to generate an input feature vector of the sewage treatment control parameter prediction model includes: Performing feature grouping processing on the multi-scale feature data; Assign weights to each feature group and generate weighted feature vectors using a weighted summation mechanism; Performing a nonlinear transformation on the weighted eigenvector value to obtain a nonlinear transformation result; The nonlinear transformation result is processed using a B-spline basis function to obtain an input feature vector.

5. The whole process control method of sewage treatment according to claim 3 is characterized in that: Based on the prediction result of the sewage treatment control parameter, controlling the sewage treatment equipment includes: Analyzing the real-time state variables of the operation of the sewage treatment equipment based on the sewage treatment control parameter prediction results; Mapping the real-time state variable and the input feature vector to generate an operation instruction suitable for the sewage treatment equipment; Send operating instructions to sewage treatment equipment.

6. The whole process control method for sewage treatment according to claim 3 is characterized in that: The sewage treatment full process control method also includes: Acquiring real-time operation data of the sewage treatment equipment when the sewage treatment equipment is controlled using the sewage treatment control parameter prediction result; Calculating the difference between the input feature vector and the real-time operation data; Adjusting the weight of the input feature vector according to the difference to obtain an updated input feature vector; The updated input feature vector is used to optimize the sewage treatment control parameter prediction model.

7. The whole process control method for sewage treatment according to any one of claims 1 to 6, characterized in that: The loss function of the sewage treatment control parameter prediction model for: in, Indicates i The true predicted value of samples, represents the mean of the prediction results, Represents the overall offset, represents the sewage vector weight, Represents the input sewage vector data, Indicates the correction offset, Indicates the total amount of sample data.

8. A control device based on the sewage treatment full-process control method according to any one of claims 1 to 7, characterized in that: include: A data acquisition module, which is used to collect multi-dimensional parameter data in the sewage treatment process using real-time sensors; A preprocessing module, used for preprocessing the multi-dimensional parameter data to generate multi-scale feature data; A model generation module is used to adjust the original language model based on the historical corpus data of sewage treatment and the common data in the fields of environmental science and chemistry to obtain a sewage treatment control parameter prediction model; A prediction result generating module, used for inputting the multi-scale feature data into the sewage treatment control parameter prediction model to generate a sewage treatment control parameter prediction result; The control sending module is used to control the sewage treatment equipment based on the sewage treatment control parameter prediction result.

9. An electronic device, characterized in that: include: processor; as well as A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by the processor, implement the full-process control method for sewage treatment as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the whole-process control method for sewage treatment as described in any one of claims 1 to 7 is implemented.

Images