Sewage treatment method and device, electronic equipment and program product

Abstract

The invention relates to a sewage treatment method and device, electronic equipment and a program product, and the method comprises the steps: collecting a water inlet data group of to-be-treated sewage before treatment, inputting the water inlet data group into a pre-trained water outlet prediction model, and obtaining water outlet data of the to-be-treated sewage after treatment and a corresponding prediction energy loss value; determining the effluent quality according to the effluent data, and determining a plurality of sewage regulation strategies according to the effluent quality, the target effluent condition and the determined predicted energy loss value; respectively determining energy loss values and effluent quality corresponding to the plurality of sewage regulation strategies, and determining a target sewage treatment strategy from the plurality of sewage regulation strategies according to the corresponding energy loss values and effluent quality; and treating the to-be-treated sewage according to the target sewage treatment strategy. On the basis, when the target sewage treatment strategy is adopted to treat the sewage, the sewage treatment efficiency can be improved, the effluent quality can be ensured, and meanwhile, the energy loss can be remarkably reduced.

Description

Technical Field

[0001] The present disclosure relates to the technical field of sewage treatment, and in particular to a sewage treatment method, device, electronic equipment and program product. Background Art

[0002] Sewage treatment plants are important facilities for urban environmental protection, and they undertake the key tasks of protecting water resources and improving the ecological environment.

[0003] In the related art, the purpose of sewage treatment is achieved mainly by adjusting the aeration volume based on past experience by related technicians, and when adjusting the aeration volume, related personnel need to manually adjust the aeration volume based on experience. In this process, it is difficult for related technicians to accurately determine the required aeration volume corresponding to the optimal energy loss based on experience, and thus it is impossible to ensure that the optimal energy loss is achieved. Summary of the invention

[0004] In view of this, the embodiments of the present disclosure provide a sewage treatment method, device, electronic device and program product to solve the problem of high data storage cost existing in the related art.

[0005] According to a first aspect of an embodiment of the present disclosure, a sewage treatment method is provided, the method comprising: collecting an influent data group of sewage to be treated before treatment, and inputting the influent data group into a pre-trained effluent prediction model to obtain effluent data of the sewage to be treated and a corresponding predicted energy loss value after treatment; determining effluent quality according to the effluent data, and determining a plurality of sewage regulation strategies according to the effluent quality, target effluent conditions and the predicted energy loss value; respectively determining the energy loss values ​​and effluent quality corresponding to the plurality of sewage regulation strategies, and determining a target sewage treatment strategy from the plurality of sewage regulation strategies according to the corresponding energy loss values ​​and effluent quality; and treating the sewage to be treated according to the target sewage treatment strategy.

[0006] According to a second aspect of the embodiments of the present disclosure, a sewage treatment device is provided, which is applied to the sewage treatment method of the first aspect, and the device includes: a collection module, which is used to collect an influent data group of the sewage to be treated before treatment, and input the influent data group into a pre-trained effluent prediction model to obtain effluent data and a corresponding predicted energy loss value after the sewage to be treated; a first determination module, which is used to determine the effluent quality according to the effluent data, and determine multiple sewage adjustment strategies according to the effluent quality, target effluent conditions and the predicted energy loss value; a second determination module, which is used to respectively determine the energy loss values ​​and effluent quality corresponding to multiple sewage adjustment strategies, and determine a target sewage treatment strategy from multiple sewage adjustment strategies according to the corresponding energy loss values ​​and effluent quality; and a processing module, which is used to treat the sewage to be treated according to the target sewage treatment strategy.

[0007] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising at least one processor; and a memory for storing instructions executable by at least one processor; wherein the at least one processor is used to execute instructions to implement the steps of the above-mentioned sewage treatment method.

[0008] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the steps of the above-mentioned sewage treatment method.

[0009] According to a fifth aspect of the present disclosure, there is provided a computer program product, comprising a computer program, which implements the steps of the above-mentioned sewage treatment method when the computer program is executed by a processor.

[0010] At least one of the above technical solutions adopted in the embodiments of the present disclosure can achieve the following beneficial effects: collect the inlet data group before the treatment of the sewage to be treated, and input the inlet data group into the pre-trained effluent prediction model to obtain the effluent data and the corresponding predicted energy loss value after the treatment of the sewage to be treated; determine the effluent quality according to the effluent data, and determine multiple sewage adjustment strategies according to the effluent quality, target effluent conditions and the predicted energy loss value; respectively determine the energy loss value and effluent quality corresponding to multiple sewage adjustment strategies, and determine the target sewage treatment strategy from multiple sewage adjustment strategies according to the corresponding energy loss value and effluent quality; treat the sewage to be treated according to the target sewage treatment strategy. Based on this, when the target sewage treatment strategy is used to treat sewage, the sewage treatment efficiency can be improved and the effluent quality can be guaranteed, while the energy loss can also be significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 A schematic flow chart of a sewage treatment method provided by an exemplary embodiment of the present disclosure;

[0013] Figure 2 A schematic diagram of a flow chart of an ACO algorithm for finding an optimal solution provided by an exemplary embodiment of the present disclosure;

[0014] Figure 3 The actual output chemical oxygen demand and the predicted output of the exemplary embodiment of the present disclosure are provided <24P03621CN>

[0015] Comparison chart of chemical oxygen demand;

[0016] Figure 4 A comparison chart of the actual output of the total nitrogen in the effluent and the predicted output of the total nitrogen in the effluent provided by an exemplary embodiment of the present disclosure;

[0017] Figure 5 A comparison diagram of actual energy loss and predicted energy loss of a wind turbine provided by an exemplary embodiment of the present disclosure;

[0018] Figure 6 A schematic flow chart of another sewage treatment method provided as an exemplary embodiment of the present disclosure;

[0019] Figure 7 A schematic structural diagram of a sewage treatment device provided by an exemplary embodiment of the present disclosure;

[0020] Figure 8 A schematic diagram of the structure of an electronic device provided for an exemplary embodiment of the present disclosure;

[0021] Fig. 9 A schematic diagram of the structure of a computer system provided for an exemplary embodiment of the present disclosure;

[0022] Fig.10 A schematic diagram of a computer program product provided for an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0024] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0025] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc. mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0026] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0027] The names of messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for <24P03621CN>

[0028] It is for illustrative purposes only and is not intended to limit the scope of these messages or information.

[0029] Sewage treatment plants are important facilities for urban environmental protection, and they undertake the key tasks of protecting water resources and improving the ecological environment. At present, many sewage treatment plants still use traditional operation methods, that is, the relevant technicians adjust the aeration volume based on past experience to achieve the purpose of sewage treatment. When adjusting the aeration volume, the relevant personnel need to manually adjust the aeration volume based on experience, and in the subsequent treatment process, the aeration volume needs to be adjusted in time according to the actual situation to achieve the purpose of sewage treatment. However, in this process, it is difficult for the relevant technicians to accurately determine the required aeration volume corresponding to the optimal energy loss based on their experience, so that the system cannot achieve the optimal energy loss efficiency, and the manual treatment is inefficient and the effluent quality is unstable.

[0030] The existing technologies mainly focus on automatic control systems and traditional data processing methods. The relevant technical solutions usually rely on fixed programs and manually summarized empirical rules to adjust the aeration volume. However, this method can only mechanically adjust the aeration volume according to preset rules, and lacks the ability to deeply analyze and optimize complex data in real time. For example, some existing technologies use a simple feedback control loop to adjust the operation of the fan according to the measured value of dissolved oxygen concentration to control the aeration process. Although this method can achieve automatic control to a certain extent, it cannot adapt to dynamic changes in the process and cannot ensure optimal energy loss.

[0031] In addition, existing technologies often do not make full use of advanced data analysis technologies, such as integrating machine learning or artificial intelligence into the data processing process to predict and optimize the sewage treatment process, which makes it difficult to achieve the best results in terms of treatment efficiency, energy loss control and effluent quality stability. Although there are some studies in the existing technology that attempt to improve the aeration process through model predictive control technology, these models usually require complex mathematical modeling and a large amount of historical data, and their flexibility and accuracy in practical applications still need to be improved.

[0032] In summary, existing sewage treatment technologies have obvious deficiencies in terms of intelligence and energy saving, especially in the aeration control link. Traditional methods usually adopt timed or quantitative aeration strategies and lack the ability to respond to and process real-time data, which leads to waste of energy loss and unstable treatment efficiency, and cannot adapt to the dynamic changes of treatment load. In addition, existing methods often fail to make full use of advanced data analysis and optimization algorithms, making it difficult to maintain the effluent quality in the best state, and have poor adaptability to environmental changes.

[0033] In view of the above problems, the embodiments of the present disclosure provide a sewage treatment method, device, electronic device and program product, and propose a flexible target interval optimization algorithm, which mainly includes the K-nearest neighbor classification algorithm of principal component analysis, nonlinear autoregressive network and ant colony optimization algorithm, and realizes real-time, dynamic and precise control of the aeration process, thereby improving the sewage treatment efficiency and ensuring the effluent quality, while significantly reducing energy loss. Based on this, the problems of the inability of traditional sewage treatment technology to control aeration to adapt to the dynamic changes of treatment load, high energy loss and unstable effluent quality can be solved.

[0034] Specifically, the sewage treatment method of the disclosed embodiment can overcome the limitation of the prior art in lacking in-depth analysis and real-time optimization capabilities of complex data by combining advanced data analysis technology, machine learning models and automatic control systems, thereby improving the automation level of sewage treatment and optimizing energy loss. In addition, the algorithm of the disclosed embodiment can also automatically adjust the aeration volume according to real-time monitoring data, effectively respond to environmental changes, ensure the stability of effluent quality, and reduce dependence on manual experience.

[0035] Figure 1 The following is a schematic diagram of a sewage treatment method provided by an exemplary embodiment of the present disclosure. Figure 1 As shown, the specific steps are as follows:

[0036] S101, collecting the corresponding influent data group before the treatment of the wastewater to be treated and the historical data group generated during the historical wastewater treatment process, and using the K-nearest neighbor classification algorithm of principal component analysis to process the data in the historical data group to obtain the data processing result. It should be understood that the historical data group includes the historical influent data group and the historical effluent data group before the historical wastewater treatment.

[0037] In some embodiments, a K-nearest neighbor classification algorithm combined with principal component analysis can be used to preprocess all data included in the historical data group. Among them, the K-nearest neighbor classification algorithm combined with principal component analysis is used to classify the data. In step S101, the original data of the historical data group can be first reduced in dimension using principal component analysis (PCA), and then the reduced-dimensional data is used as the input of the K-nearest neighbor algorithm (KNN), and the KNN outputs the category corresponding to each data. It should be understood that PCA can reduce the data dimension of the data, thereby improving the efficiency and accuracy of KNN classification.

[0038] In practical applications, principal component analysis (PCA) is a data dimensionality reduction technique that is mainly used to reduce the dimension of data while retaining the features that contribute the most to the variance in the data.

[0039] For example, PCA can be used to clean the data in the historical data group, process missing values, outliers, and perform necessary data type conversion operations to reduce the dimensionality of the data, select the principal components that can retain most of the data variance, reduce the data dimension while retaining important information as much as possible, and the PCA results can provide guidance for the KNN algorithm to help further optimize the KNN settings. It should be understood that the K-nearest neighbor algorithm (KNN) is a basic classification algorithm that predicts the category of samples based on the distance between different feature values. By assigning data to the most appropriate feature samples and optimizing dissolved oxygen for each classification, it not only improves the accuracy of data classification in the sewage treatment process, but also provides valuable information for the optimization of aeration volume, thereby achieving a more efficient and energy-saving sewage treatment process.

[0040] Specifically, firstly, the data in the historical data group needs to be cleaned, missing values ​​and outliers need to be processed, and necessary data type conversion needs to be performed to ensure the data quality. Then, PCA is used to reduce the data dimension, with the aim of retaining the main variability in the data while reducing the data dimension. Finally, the KNN classification model is used to classify the reduced-dimensional data according to the preset water quality indicators and preset rules. Specifically, a suitable distance metric can be selected to evaluate the similarity between data, balance the risks of overfitting and underfitting, and thus provide a basis for classification decisions. Among them, the preset rule can be the sewage treatment time.

[0041] For example, for historical data, it is necessary to first determine the start time and end time of a sewage treatment, and then filter out the data generated by the start time and end time in the data group, and at the same time divide the data into the data group corresponding to the start time and the data group corresponding to the end time according to the time corresponding to the data, so that the data group corresponding to the start time and the data group corresponding to the end time can be determined as one category. On this basis, it is also necessary to determine the water quality indicators corresponding to each data. Among them, water quality indicators can include pH value, hardness, alkalinity, various ions and chemical oxygen demand, etc., which can be set according to actual conditions. Based on this, multiple data groups can be classified according to the processing time.

[0042] In practical applications, when using PCA to reduce data dimensions, we can first calculate the covariance matrix corresponding to each data in the historical data group, and solve the eigenvalues ​​and eigenvectors corresponding to each covariance matrix, and then select the most important principal components according to the size of the eigenvalues. These principal components can capture the key information of the data, while removing redundancy and noise, and converting the original data into a lower-dimensional space, thereby improving the efficiency of data processing. Then use the KNN algorithm to analyze the data processed by PCA to determine the category of each water quality indicator.

[0043] Specifically, classification optimization is a key link in the KNN algorithm, which involves adjusting the K value and optimizing the distance metric and classification decision rules. In practical applications, the impact of different K values ​​on classification performance can be evaluated by using cross-validation and other techniques, and a K value that maximizes the classification accuracy can be selected. Different distance metrics, such as Euclidean distance, are compared to adapt to the characteristics of the data and the requirements of the classification task. It should be understood that the goal of optimization is to improve the accuracy, robustness, and generalization ability of the classifier, and to ensure that the model can perform well on new data, that is, data that has not been involved.

[0044] S102, using the classified data to train the preset model to obtain an effluent prediction model, and inputting the corresponding influent data group before the treatment of the sewage to be treated into the pre-trained effluent prediction model to obtain the predicted effluent data after the treatment of the sewage to be treated.

[0045] In some embodiments, a neural network is a machine learning model that simulates the working mode of the human brain nervous system and is composed of multiple neurons connected to each other. Each neuron receives an input signal, processes it, and then outputs it to the next layer of neurons. Through training, the neural network can learn complex patterns in the input data and use it for tasks such as classification and prediction. Based on this, the effluent prediction model of the embodiment of the present disclosure can be a nonlinear autoregressive neural network with external inputs (NonlinearAutoregressive Network with eXogenous inputs, NARX). NARX is a time series effluent prediction model that combines the characteristics of the autoregressive model with the theory of nonlinear dynamic systems and can handle the complex relationship between input and output sequences. Among them, in the prediction of energy loss and effluent parameters of a sewage treatment plant, a NARX model is constructed based on classified data. The model can use historical data to learn the dynamic changes in the aeration process, thereby predicting future energy loss and effluent parameters.

[0046] Specifically, the NARX neural network model includes an input layer, a hidden layer, and an output layer. The input layer receives the data classified in step S101, the hidden layer processes the input data through a nonlinear activation function, and the output layer provides a prediction result. When training the model, the data classified in step S101 is used as a training set, and the network weights are optimized by a back propagation algorithm. In the model application stage, the influent data group of the sewage to be treated is input into the model, and the energy loss and effluent parameters predicted for the sewage to be treated are output, and the predicted energy loss and effluent parameters are used to guide aeration control.

[0047] For example, when building a NARX neural network model, you first need to determine the input and output of the model. Among them, the input data can be determined as influent chemical oxygen demand, influent ammonia nitrogen, influent total nitrogen, influent total phosphorus, influent pH, influent flow, dissolved oxygen through correlation analysis, and the output parameters are fan energy loss and data corresponding to water quality indicators such as effluent chemical oxygen demand, ammonia nitrogen and total nitrogen. Next, design the network structure, including determining the number of hidden layers and the number of neurons in each layer. Finally, select a suitable activation function, such as ReLU or Sigmoid, to ensure how the output of the network is correctly mapped to the prediction target.

[0048] In practical applications, appropriate weights can be assigned to different categories of data or trained separately according to the data classification results in step S101 to enhance the adaptability of the model to specific categories of data. The key to training the NARX model is to select appropriate network structure and algorithm parameters. Through optimization techniques such as back propagation algorithm and gradient descent, the weights and biases of the model will be continuously adjusted to minimize the prediction error. This process may require multiple iterations until the performance of the model reaches the expected goal.

[0049] In some embodiments, the model training process involves training the network using a historical data group. The historical data group is divided into a training set, a test set, and a prediction set in a ratio of 7:2:1, and then completed through a back propagation algorithm, where the loss function is used to measure the difference between the predicted value and the actual value, and the network weights are adjusted through an estimation algorithm to minimize the loss function. It should be understood that the loss function can be the mean square error, which is the expected value of the square of the difference between the parameter estimate and the true value of the parameter in mathematical statistics. The estimation algorithm can be the LM method, which is the full name of the Levenberg-Marquardt method, which is an estimation method for least squares estimation of regression parameters in nonlinear regression.

[0050] In practical applications, in order to ensure the generalization ability of the model, it is necessary to verify the prediction effect of the model on an independent test data set. This test set contains data that has not been seen in the training process. By comparing the prediction results of the model with the actual observations, the accuracy of the model can be evaluated.

[0051] Specifically, the mean square error and determination coefficient on the prediction set will be used to evaluate the model performance. At the same time, the cross-validation method will be used to optimize the network structure. By performing cross-validation on the main parameter architecture (such as the number of hidden layers, regularization coefficient, number of training times, and optimization algorithm), the accuracy of model prediction can be improved.

[0052] S103, using the ant colony optimization algorithm to determine the sewage treatment strategy that satisfies the preset conditions and minimizes the energy loss value based on the effluent data. It should be understood that the Ant Colony Optimization (ACO) algorithm is a heuristic search algorithm that simulates the foraging behavior of ants. It solves complex optimization problems by simulating the process of ants releasing pheromones and following the pheromone concentration to find food.

[0053] In some embodiments, the optimization goal is usually to minimize energy loss while ensuring that the effluent quality meets the discharge standard. Therefore, it is necessary to find a balance between energy loss and effluent quality. The purpose of the ACO algorithm is to find the optimal setting value by setting an optimization function and introducing a penalty mechanism, so as to ensure the lowest energy loss while meeting the effluent quality requirements. The specific optimization function is:

[0054] D=E+a[(TN Pre -TN tmin ) 2 +(TN Pre -TN tman ) 2 ]

[0055] Where: D represents the output value of the optimization function, E represents the energy loss, TN Pre Indicates the total nitrogen value of the water, TN tmin Indicates the lower limit of the target range, TN tman It represents the upper limit of the target interval, and a represents the penalty coefficient for exceeding the interval range.

[0056] The embodiment of the present disclosure aims to minimize the energy loss E while ensuring the total nitrogen concentration TN of the effluent when constructing the composite optimization objective function D. Pre If the effluent total nitrogen falls within the target range (10 mg / L to 12 mg / L), then by introducing a penalty term a, when the predicted effluent total nitrogen TN Pre When the set target range (10 mg / L to 12 mg / L) is exceeded or the energy loss E is significantly increased, a higher cost is given to such a solution. At this time, the value of a is dynamically increased. Based on the degree of exceeding the target range or the increase in energy loss, the more it deviates from the ideal state, the greater the penalty of a, thereby guiding the ACO algorithm to prefer strategies that can keep the total nitrogen in the effluent within the target range and minimize energy loss.

[0057] In some embodiments, the ACO algorithm is an optimization tool based on swarm intelligence, where each ant represents a dissolved oxygen regulation strategy and updates the pheromone concentration based on their path quality (energy loss and effluent total nitrogen control). After multiple iterations, the pheromone concentration will guide the ants to find a dissolved oxygen regulation strategy with minimal energy loss and effluent total nitrogen control within the target range. In the application of intelligent aeration control, ACO searches for the optimal dissolved oxygen set point to achieve the optimization goal.

[0058] Figure 2 A schematic diagram of a flow chart of an ACO algorithm for finding an optimal solution provided by an exemplary embodiment of the present disclosure. Figure 2 As shown, the specific steps are as follows:

[0059] S201: Initialize the position of each ant.

[0060] In some embodiments, it is necessary to ensure that the total nitrogen concentration TN PreIt falls within the target interval (10 mg / L to 12 mg / L). Based on this, multiple sewage regulation strategies can be determined according to the effluent data, specifically a dissolved oxygen regulation strategy, which randomly initializes the position of each ant (dissolved oxygen value) and the pheromone concentration on the initialization path.

[0061] S202: Select a path.

[0062] In some embodiments, when each ant selects the next dissolved oxygen value at a decision point, a probabilistic selection is made based on the pheromone concentration and distance (i.e., predicted energy consumption and total nitrogen mass of the effluent) on the current path, where paths with high pheromone concentrations will have a higher probability of exploration.

[0063] S203: Determine whether the algorithm has converged. If not, execute step S204; if converged, execute step S205.

[0064] In some embodiments, after finding the optimal solution, it is necessary to evaluate it in detail, monitor the changes in the global optimal solution (the dissolved oxygen strategy with the lowest energy consumption and meeting the total nitrogen requirements of the effluent) after each round of iteration, and observe whether it tends to be stable, so as to judge whether the algorithm has converged, and use cross-validation or independent test sets to verify the found optimal dissolved oxygen strategy, evaluate its performance in actual or simulated environments, including the stability of the total nitrogen in the effluent and the degree of reduction in energy consumption, examine the algorithm's sensitivity to parameters (such as the number of ants, pheromone volatilization rate, etc.), and evaluate the impact of parameter adjustment on the optimization results to determine the optimal parameter settings.

[0065] S204, update pheromones.

[0066] In some embodiments, when the algorithm does not converge, the pheromone can be updated, and then the optimization setting value can be updated, so that the path selection can be performed again.

[0067] S205, determining whether the stop condition is met.

[0068] In some embodiments, after each round of iteration, the pheromone on the path can be updated according to the optimization function of each ant. For the path that finds a better solution (i.e., lower energy consumption and the total nitrogen in the effluent meets the target range), the amount of pheromone on it is increased; for other paths, the pheromone gradually evaporates. Based on this, through multiple iterations, the pheromone distribution can gradually reflect a better dissolved oxygen regulation strategy until a dissolved oxygen combination that meets the termination conditions can be found in each classification case.

[0069] S206, output the optimal group, that is, output the dissolved oxygen combination that meets the termination condition.

[0070] Figure 3A comparison chart of actual output chemical oxygen demand and predicted output chemical oxygen demand provided by an exemplary embodiment of the present disclosure. Figure 4 A comparison chart of actual output of effluent total nitrogen and predicted output of effluent total nitrogen provided by an exemplary embodiment of the present disclosure. Figure 5 A comparison diagram of the actual energy loss and the predicted energy loss of a wind turbine provided by an exemplary embodiment of the present disclosure. Figure 3 , Figure 4 and Figure 5 As shown, it is calculated that the energy consumption can be reduced by up to 20.8%, indicating that ACO can achieve the goal of energy saving and consumption reduction by finding the optimal dissolved oxygen combination while stabilizing emission indicators.

[0071] The sewage treatment method provided by the embodiment of the present disclosure may be executed by a terminal or by a chip applied to the terminal.

[0072] Exemplarily, the above-mentioned terminals may include one or more of mobile phones, tablet computers, wearable devices, vehicle-mounted devices, laptop computers, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), and wearable devices based on augmented reality (AR) and / or virtual reality (VR) technology, and may also include but not be limited to remote control devices, wearable devices, street lamps, home appliances and other smart terminals, which are not specifically limited in the embodiments of the present disclosure.

[0073] Figure 6 A schematic diagram of another sewage treatment method provided by an exemplary embodiment of the present disclosure. Figure 6 As shown, the specific steps include:

[0074] S601, collecting an influent data group before the wastewater to be treated is processed, and inputting the influent data group into a pre-trained effluent prediction model to obtain effluent data after the wastewater to be treated is processed and a corresponding predicted energy loss value.

[0075] In some embodiments, existing water quality testing tools can be used to test the treated sewage to obtain a group of influent data before the treatment of the treated sewage, and then input the group of influent data into a pre-trained effluent prediction model. The model outputs the effluent data after the treatment of the treated sewage and the corresponding predicted energy loss value through corresponding algorithm analysis.

[0076] S602, determining effluent quality according to effluent data, and determining a plurality of sewage regulation strategies according to the effluent quality, target effluent conditions and predicted energy loss values.

[0077] In some embodiments, the model can only mechanically determine a sewage treatment method to treat the sewage, which may consume a lot of energy to obtain a relatively good effluent quality, but in actual applications, the effluent quality does not need to be so good, and it is only necessary to meet the target effluent conditions. Therefore, the embodiment of the present disclosure can also determine multiple sewage adjustment strategies based on the effluent quality, the target effluent conditions and the predicted energy loss value, and then select the best sewage adjustment strategy to treat the sewage to be treated. It should be understood that the target effluent condition can be a specified discharge standard.

[0078] Specifically, when the output of the effluent prediction model shows that a higher effluent quality water is obtained by using a larger energy loss, the energy loss can be appropriately reduced, and multiple sewage regulation strategies can be determined based on the reduced energy loss and target effluent conditions.

[0079] S603, respectively determining energy loss values ​​and effluent quality corresponding to a plurality of sewage regulation strategies, and determining a target sewage treatment strategy from the plurality of sewage regulation strategies according to the corresponding energy loss values ​​and effluent quality.

[0080] In some embodiments, each sewage regulation strategy corresponds to an energy loss value and effluent quality. Therefore, after determining the energy loss values ​​and effluent qualities corresponding to multiple sewage regulation strategies respectively, it is only necessary to select a sewage treatment strategy with a smaller energy loss value and effluent quality that meets the target conditions. This can not only ensure lower energy loss, but also make the effluent quality meet the target effluent conditions, thereby reducing operating costs.

[0081] S604, treating the wastewater to be treated according to the target wastewater treatment strategy.

[0082] In some embodiments, when sewage is treated using a targeted sewage treatment strategy, the sewage treatment efficiency can be improved and the effluent quality can be guaranteed, while also significantly reducing energy loss.

[0083] In some embodiments, the embodiments of the present disclosure can construct a historical database, wherein the historical database includes a group of historical data generated during the historical sewage treatment process; construct a sample set based on the historical database; input the group of influent data before the treatment of the sewage to be treated into the effluent prediction model to obtain the effluent data after the treatment of the sewage to be treated and the corresponding predicted energy loss value; wherein the preset model is trained by training samples to obtain the effluent prediction model, and the training samples are obtained based on the sample set.

[0084] The effluent prediction model of the disclosed embodiment can be a nonlinear autoregressive neural network with external inputs (NARX). NARX is a time series effluent prediction model that combines the characteristics of the autoregressive model and the theory of nonlinear dynamic systems and can handle the complex relationship between the input and output sequences. In the prediction of energy loss and effluent parameters of a sewage treatment plant, a NARX model is constructed based on classified data. The model can use historical data to learn the dynamic changes in the aeration process, thereby predicting future energy loss and effluent parameters.

[0085] Specifically, the NARX neural network model includes an input layer, a hidden layer, and an output layer. The input layer receives classified data, the hidden layer processes the input data through a nonlinear activation function, and the output layer provides prediction results. When training the model, the classified data is used as a training set, and the network weights are optimized through a back propagation algorithm. In the model application stage, the influent data group of the sewage to be treated is input into the model, and the predicted energy loss and effluent parameters for the sewage to be treated are output. The predicted energy loss and effluent parameters are used to guide aeration control.

[0086] In some embodiments, before model training, it is also necessary to classify the historical data groups generated in the historical sewage treatment process according to preset rules; and construct a sample set based on the classified data.

[0087] Specifically, the K-nearest neighbor classification algorithm combined with principal component analysis can be used to preprocess all the data included in the historical data group. Among them, the K-nearest neighbor classification algorithm combined with principal component analysis is used to classify the data. In step S101, the original data of the historical data group can be first reduced in dimension using principal component analysis (PCA), and then the reduced-dimensional data is used as the input of the K-nearest neighbor algorithm (KNN), and the KNN outputs the category corresponding to each data. It should be understood that PCA can reduce the data dimension of the data, thereby improving the efficiency and accuracy of KNN classification.

[0088] In practical applications, principal component analysis (PCA) is a data dimensionality reduction technique that is mainly used to reduce the dimension of data while retaining the features that contribute the most to the variance in the data.

[0089] For example, PCA can be used to clean the data in the historical data group, process missing values, outliers, and perform necessary data type conversion operations to reduce the dimensionality of the data, select the principal components that can retain most of the data variance, reduce the data dimension while retaining important information as much as possible, and the PCA results can provide guidance for the KNN algorithm to help further optimize the KNN settings. It should be understood that the K-nearest neighbor algorithm (KNN) is a basic classification algorithm that predicts the category of samples based on the distance between different feature values. By assigning data to the most appropriate feature samples and optimizing dissolved oxygen for each classification, it not only improves the accuracy of data classification in the sewage treatment process, but also provides valuable information for the optimization of aeration volume, thereby achieving a more efficient and energy-saving sewage treatment process.

[0090] Specifically, firstly, the data in the historical data group needs to be cleaned, missing values ​​and outliers need to be processed, and necessary data type conversion needs to be performed to ensure the data quality. Then, PCA is used to reduce the data dimension, with the aim of retaining the main variability in the data while reducing the data dimension. Finally, the KNN classification model is used to classify the reduced-dimensional data according to the preset water quality indicators and preset rules. Specifically, a suitable distance metric can be selected to evaluate the similarity between data, balance the risks of overfitting and underfitting, and thus provide a basis for classification decisions. Among them, the preset rule can be the sewage treatment time.

[0091] For example, for historical data, it is necessary to first determine the start time and end time of a sewage treatment, and then filter out the data generated by the start time and end time in the data group, and at the same time divide the data into the data group corresponding to the start time and the data group corresponding to the end time according to the time corresponding to the data, so that the data group corresponding to the start time and the data group corresponding to the end time can be determined as one category. On this basis, it is also necessary to determine the water quality indicators corresponding to each data. Among them, water quality indicators can include pH value, hardness, alkalinity, various ions and chemical oxygen demand, etc., which can be set according to actual conditions. Based on this, multiple data groups can be classified according to the processing time.

[0092] In practical applications, when using PCA to reduce data dimensions, we can first calculate the covariance matrix corresponding to each data in the historical data group, and solve the eigenvalues ​​and eigenvectors corresponding to each covariance matrix, and then select the most important principal components according to the size of the eigenvalues. These principal components can capture the key information of the data, while removing redundancy and noise, and converting the original data into a lower-dimensional space, thereby improving the efficiency of data processing. Then use the KNN algorithm to analyze the data processed by PCA to determine the category of each water quality indicator.

[0093] Specifically, classification optimization is a key link in the KNN algorithm, which involves adjusting the K value and optimizing the distance metric and classification decision rules. In practical applications, the impact of different K values ​​on classification performance can be evaluated by using cross-validation and other techniques, and a K value that maximizes the classification accuracy can be selected. Different distance metrics, such as Euclidean distance, are compared to adapt to the characteristics of the data and the requirements of the classification task. It should be understood that the goal of optimization is to improve the accuracy, robustness, and generalization ability of the classifier, and to ensure that the model can perform well on new data, that is, data that has not been involved.

[0094] In some embodiments, appropriate weights can be assigned to different categories of data or trained separately based on the data classification results to enhance the adaptability of the model to specific categories of data. The key to training the NARX model is to select appropriate network structure and algorithm parameters. Through optimization techniques such as back propagation algorithm and gradient descent, the weights and biases of the model will be continuously adjusted to minimize the prediction error. This process may require multiple iterations until the performance of the model reaches the expected goal.

[0095] In practical applications, when constructing a NARX neural network model, the input and output of the model must be determined first. The input data can be determined through correlation analysis to be the inlet chemical oxygen demand, inlet ammonia nitrogen, inlet total nitrogen, inlet total phosphorus, inlet pH, inlet flow rate, dissolved oxygen, and the output parameters are the data corresponding to the water quality indicators such as fan energy loss and outlet chemical oxygen demand, ammonia nitrogen and total nitrogen. Next, the network structure is designed, including determining the number of hidden layers and the number of neurons in each layer. Finally, a suitable activation function is selected, such as ReLU or Sigmoid, to ensure how the output of the network is correctly mapped to the prediction target.

[0096] Specifically, the model training process involves using historical data groups to train the network. The historical data groups are divided into training sets, test sets, and prediction sets in a ratio of 7:2:1, and then completed through the back propagation algorithm, where the loss function (mean square error MSE) is used to measure the difference between the predicted value and the actual value. The network weights are adjusted to minimize the loss function through the Levenberg-Marquardt optimization algorithm.

[0097] In practical applications, in order to ensure the generalization ability of the model, it is necessary to verify the prediction effect of the model on an independent test data set. This test set contains data that has not been seen in the training process. By comparing the prediction results of the model with the actual observations, the accuracy of the model can be evaluated.

[0098] Specifically, the MSE and determination coefficient on the prediction set will be used to evaluate the model performance. At the same time, the cross-validation method will be used to optimize the network structure. By performing cross-validation on the main parameter architecture (such as the number of hidden layers, regularization coefficient, number of training times, and optimization algorithm), the accuracy of model prediction can be improved.

[0099] In some embodiments, determining a target sewage treatment strategy from multiple sewage regulation strategies according to corresponding energy loss values ​​and effluent quality includes: constructing an optimization function, and using the optimization function to determine the target sewage treatment strategy from multiple sewage regulation strategies. The optimization function is:

[0100] D=E+a[(TN Pre -TN tmin ) 2 +(TN Pre -TN tman ) 2 ]

[0101] Where: D represents the output value of the optimization function, E represents energy consumption, TN Pre Indicates the total nitrogen value of the water, TN tmin Indicates the lower limit of the target range, TN tman It represents the upper limit of the target interval, and a represents the penalty coefficient for exceeding the interval range.

[0102] In some embodiments, the optimization goal is usually to minimize energy loss while ensuring that the effluent quality meets the discharge standard. Therefore, it is necessary to find a balance between energy loss and effluent quality. The purpose of the ACO algorithm is to find the optimal setting value by setting an optimization function and introducing a penalty mechanism, so as to ensure the lowest energy loss while meeting the effluent quality requirements. The specific optimization function is:

[0103] D=E+a[(TN Pre -TN tmin ) 2 +(TN Pre -TN tman ) 2 ]

[0104] Where: D represents the output value of the optimization function, E represents the energy loss, TN Pre Indicates the total nitrogen value of the water, TN tmin Indicates the lower limit of the target range, TN tman It represents the upper limit of the target interval, and a represents the penalty coefficient for exceeding the interval range.

[0105] The embodiment of the present disclosure aims to minimize the energy loss E while ensuring the total nitrogen concentration TN of the effluent when constructing the composite optimization objective function D. PreIf the effluent total nitrogen falls within the target range (10 mg / L to 12 mg / L), then by introducing a penalty term a, when the predicted effluent total nitrogen TN Pre When the set target range (10 mg / L to 12 mg / L) is exceeded or the energy loss E is significantly increased, a higher cost is given to such a solution. At this time, the value of a is dynamically increased. Based on the degree of exceeding the target range or the increase in energy loss, the more it deviates from the ideal state, the greater the penalty of a, thereby guiding the ACO algorithm to prefer strategies that can keep the total nitrogen in the effluent within the target range and minimize energy loss.

[0106] In some embodiments, the ACO algorithm is an optimization tool based on swarm intelligence, where each ant represents a dissolved oxygen regulation strategy and updates the pheromone concentration based on their path quality (energy loss and effluent total nitrogen control). After multiple iterations, the pheromone concentration will guide the ants to find a dissolved oxygen regulation strategy with minimal energy loss and effluent total nitrogen control within the target range. In the application of intelligent aeration control, ACO searches for the optimal dissolved oxygen set point to achieve the optimization goal.

[0107] In some embodiments, after the sewage to be treated is treated according to the target sewage treatment strategy, the effluent quality corresponding to the sewage to be treated meets the target effluent condition, and the energy loss value is less than any one of the multiple sewage regulation strategies.

[0108] In summary, the embodiments of the present disclosure can achieve dynamic optimization of the sewage treatment process by integrating advanced real-time data monitoring systems and intelligent algorithms. Compared with traditional static treatment methods, this intelligent adjustment mechanism significantly improves the speed and efficiency of sewage treatment, greatly reduces dependence on operators, and achieves higher operational efficiency and lower operating costs.

[0109] At the same time, the NARX neural network model is used for accurate prediction to ensure the stability and reliability of the effluent quality. By learning from historical data, the effluent prediction model can accurately predict the effluent data during the treatment process, thereby achieving fine control of the effluent data.

[0110] Moreover, in the disclosed embodiment, the application of the ant colony optimization algorithm (ACO) optimizes the energy consumption management of the aeration process. By precisely controlling the aeration volume, the waste of electrical energy is effectively avoided. The detailed energy consumption analysis results confirm that, compared with the traditional treatment method, the disclosed embodiment can achieve a maximum energy consumption reduction of about 20%. This significant energy-saving effect not only reduces operating costs, but also helps to achieve the goal of green, low-carbon and circular development of sewage treatment plants.

[0111] Based on this, the disclosed embodiment can fully consider the impact of environmental changes on the sewage treatment process, showing strong adaptability. By combining the K-nearest neighbor classification algorithm (PCA-KNN) with principal component analysis, it can maintain efficient treatment effects and stable effluent quality regardless of seasonal changes or weather fluctuations. Practical applications have proved that all indicators of effluent can reach the set values, which means that the effluent quality can still meet the emission standards while reducing power consumption.

[0112] The above mainly introduces the solution provided by the embodiment of the present disclosure from the perspective of the server. It is understandable that, in order to implement the above functions, the server includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present disclosure.

[0113] The embodiments of the present disclosure may divide the server into functional units according to the above method examples. For example, each functional module may be divided corresponding to each function, or two or more functions may be integrated into one management module. The above integrated modules may be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiments of the present disclosure is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0114] In the case of dividing each functional module according to each function, an exemplary embodiment of the present disclosure provides a sewage treatment device, which may be a server or a chip applied to a server. Figure 7 The schematic diagram of the structure of a sewage treatment device provided by an exemplary embodiment of the present disclosure is as follows. Figure 7 As shown, the sewage treatment device 700 includes:

[0115] The collection module 701 is used to collect the influent data group before the sewage to be treated is treated, and input the influent data group into the pre-trained effluent prediction model to obtain the effluent data after the sewage to be treated is treated and the corresponding predicted energy loss value;

[0116] A first determination module 702 is used to determine the effluent quality according to the effluent data, and determine a plurality of sewage regulation strategies according to the effluent quality, target effluent conditions and the determined predicted energy loss value;

[0117] The second determination module 703 is used to respectively determine the energy loss values ​​and effluent quality corresponding to the plurality of sewage regulation strategies, and determine the target sewage treatment strategy from the plurality of sewage regulation strategies according to the corresponding energy loss values ​​and effluent quality;

[0118] The processing module 704 is used to process the wastewater to be treated according to the target wastewater treatment strategy.

[0119] In an optional manner, the acquisition module 701 is also used to construct a historical database, wherein the historical database includes a group of historical data generated during historical sewage treatment processes; a sample set is constructed based on the historical database; the group of influent data before the treatment of the sewage to be treated is input into the effluent prediction model to obtain the effluent data after the treatment of the sewage to be treated and the corresponding predicted energy loss value; wherein the preset model is trained through training samples to obtain the effluent prediction model, and the training samples are obtained based on the sample set.

[0120] In an optional manner, the historical data group generated in the historical sewage treatment process is classified according to preset rules; and a sample set is constructed based on the classified data.

[0121] In an optional manner, the second determination module 703 is further used to construct an optimization function, and use the optimization function to determine a target sewage treatment strategy from the multiple sewage regulation strategies.

[0122] In an optional manner, the optimization function is:

[0123] D=E+a[(TN Pre -TN tmin ) 2 +(TN Pre -TN tman ) 2 ]

[0124] Where: D represents the output value of the optimization function, E represents energy consumption, TN Pre Indicates the total nitrogen value of the water, TN tmin Indicates the lower limit of the target range, TN tman It represents the upper limit of the target interval, and a represents the penalty coefficient for exceeding the interval range.

[0125] In an optional manner, after the sewage to be treated is treated according to the target sewage treatment strategy, the effluent quality corresponding to the sewage to be treated meets the target effluent condition, and the energy loss value is less than any one of the multiple sewage regulation strategies.

[0126] An embodiment of the present disclosure also provides an electronic device, comprising: at least one processor; a memory for storing instructions executable by at least one processor; wherein the at least one processor is used to execute instructions to implement the steps of the above method disclosed in the embodiment of the present disclosure.

[0127] Figure 8 The structure diagram of an electronic device provided by an exemplary embodiment of the present disclosure is shown in FIG. Figure 8 As shown, the electronic device 800 includes at least one processor 801 and a memory 802 coupled to the processor 801 , and the processor 801 can execute corresponding steps in the above method disclosed in the embodiment of the present disclosure.

[0128] The processor 801 may also be referred to as a central processing unit (CPU), which may be an integrated circuit chip having signal processing capabilities. Each step in the method disclosed in the embodiment of the present disclosure may be completed by an integrated logic circuit of hardware in the processor 801 or by instructions in the form of software. The processor 801 may be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present disclosure may be directly embodied as being executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in a decoding processor. The software module may be located in a memory 802, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, or other mature storage media in the art. The processor 801 reads the information in the memory 802 and completes the steps of the method in combination with its hardware.

[0129] In addition, when various operations / processes according to the present disclosure are implemented by software and / or firmware, they can be transmitted from a storage medium or a network to a computer system having a dedicated hardware structure, for example, Fig. 9 The computer system 900 shown installs the programs constituting the software, and when the various programs are installed, the computer system can execute various functions, including the functions described above. Fig. 9 A schematic diagram of the structure of a computer system provided by an exemplary embodiment of the present disclosure.

[0130] Computer system 900 is intended to represent various forms of digital electronic computer equipment, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic equipment can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to be construed in detail.

[0131] And it is not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0132] like Fig. 9 As shown, the computer system 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the computer system 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0133] A plurality of components in the computer system 900 are connected to the I / O interface 905, including: an input unit 906, an output unit 907, a storage unit 908, and a communication unit 909. The input unit 906 may be any type of device capable of inputting information to the computer system 900, and the input unit 906 may receive input digital or character information, and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 907 may be any type of device capable of presenting information, and may include, but is not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 908 may include, but is not limited to, a disk, an optical disk. The communication unit 909 allows the computer system 900 to exchange information / data with other devices over a network such as the Internet, and may include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, for example, a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0134] The computing unit 901 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 901 performs the various methods and processes described above. For example, in some embodiments, the above methods disclosed in the embodiments of the present disclosure may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 908. In some embodiments, part or all of the computer program may be loaded and / or installed on an electronic device via a ROM 902 and / or a communication unit 909. In some embodiments, the computing unit 901 may be configured to perform the above methods disclosed in the embodiments of the present disclosure in any other appropriate manner (e.g., by means of firmware).

[0135] The embodiment of the present disclosure also provides a computer-readable storage medium, wherein when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the above method disclosed in the embodiment of the present disclosure.

[0136] The computer-readable storage medium in the disclosed embodiments may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. The computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. More specifically, the computer-readable storage medium may include an electrical connection based on one or more lines, a portable computer 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 of the foregoing.

[0137] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0138] Fig.10 Schematic diagram of a computer program product provided by an exemplary embodiment of the present disclosure. Fig.10 As shown, the computer program product 1000 includes a computer program 1001, wherein the computer program 1001 implements the above method disclosed in the embodiment of the present disclosure when executed by a processor.

[0139] In embodiments of the present disclosure, computer program codes for performing the operations of the present disclosure may be written in one or more programming languages ​​or combinations thereof, including but not limited to object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer.

[0140] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a portion of a code, which contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the boxes may also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified functions or operations, or can be implemented with a dedicated <24P03621CN>

[0141] Implemented using a combination of hardware and computer instructions.

[0142] The modules, components or units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the names of the modules, components or units do not, in some cases, limit the modules, components or units themselves.

[0143] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0144] The above descriptions are only some embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, a technical solution formed by replacing the above features with the technical features with similar functions disclosed in the present disclosure (but not limited to).

[0145] Although some specific embodiments of the present disclosure have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A sewage treatment method, characterized in that: include: Collecting an influent data group before the sewage to be treated is processed, and inputting the influent data group into a pre-trained effluent prediction model to obtain effluent data after the sewage to be treated is processed and a corresponding predicted energy loss value; Determining effluent quality according to the effluent data, and determining a plurality of sewage regulation strategies according to the effluent quality, target effluent conditions and the determined predicted energy loss value; Respectively determine the energy loss values ​​and effluent quality corresponding to the plurality of sewage regulation strategies, and determine a target sewage treatment strategy from the plurality of sewage regulation strategies according to the corresponding energy loss values ​​and effluent quality; The wastewater to be treated is treated according to the target wastewater treatment strategy.

2. The method according to claim 1, characterized in that The step of inputting the inlet data group into a pre-trained effluent prediction model to obtain effluent data after the treated sewage and the corresponding predicted energy loss value comprises: Constructing a historical database, wherein the historical database includes a historical data group generated during the historical sewage treatment process; Construct sample sets based on historical databases; Inputting the influent data group of the sewage to be treated before treatment into the effluent prediction model to obtain the effluent data of the sewage to be treated after treatment and the corresponding predicted energy loss value; The water output prediction model is obtained by training the preset model through training samples, and the training samples are obtained based on the sample set.

3. The method according to claim 2, characterized in that The step of constructing a sample set according to the historical database includes: Classify the historical data groups generated during the historical sewage treatment process according to preset rules; Construct a sample set based on the classified data.

4. The method according to claim 1, characterized in that: Determining a target sewage treatment strategy from a plurality of sewage regulation strategies according to corresponding energy loss values ​​and effluent quality includes: An optimization function is constructed, and a target sewage treatment strategy is determined from the plurality of sewage regulation strategies using the optimization function.

5. The method according to claim 4, characterized in that The optimization function is: D=E+a[(TN Pre -TN tmin ) 2 +(TN Pre -TN tman ) 2 ] Where: D represents the output value of the optimization function, E represents energy consumption, TN Pre Indicates the total nitrogen value of the water, TN tmin Indicates the lower limit of the target range, TN tman represents the upper limit of the target interval, and a represents the penalty coefficient for exceeding the interval range.

6. The method according to claim 1, characterized in that After the sewage to be treated is treated according to the target sewage treatment strategy, the effluent quality corresponding to the sewage to be treated meets the target effluent condition, and the energy loss value is less than any one of the multiple sewage regulation strategies.

7. A sewage treatment device, characterized in that: include: A collection module is used to collect an influent data group before the sewage to be treated is treated, and input the influent data group into a pre-trained effluent prediction model to obtain effluent data after the sewage to be treated is treated and a corresponding predicted energy loss value; A first determination module, configured to determine effluent quality according to the effluent data, and determine a plurality of sewage regulation strategies according to the effluent quality, target effluent conditions and the determined predicted energy loss value; A second determination module is used to respectively determine the energy loss values ​​and effluent quality corresponding to the plurality of sewage regulation strategies, and determine a target sewage treatment strategy from the plurality of sewage regulation strategies according to the corresponding energy loss values ​​and effluent quality; A processing module is used to process the wastewater to be treated according to the target wastewater treatment strategy.

8. An electronic device, characterized in that: include: at least one processor; a memory for storing the at least one processor-executable instruction; The at least one processor is used to execute the instructions to implement the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the steps of the method as claimed in any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the steps of the method according to any one of claims 1 to 6 when being executed by a processor.

Images