A sewage treatment data optimization method and system based on cloud computing

Abstract

The present invention provides a sewage treatment data optimization method and system based on cloud computing, which relates to the field of data processing technology. The method includes: receiving real-time key information collected by monitoring equipment in real time, including water quality indicators, equipment operating parameters, and energy consumption data, and pre-processing the key information to obtain a processed data set; constructing a cloud database based on the processed data set, and using the cloud database to perform data analysis, including trend prediction and anomaly detection, to obtain analysis results; constructing a sewage treatment data optimization model based on the analysis results, and inputting water quality standards, equipment operating efficiency, and energy consumption cost parameters into the sewage treatment data optimization model to generate an optimization plan. The present invention collects and analyzes key information on sewage treatment in real time, constructs and optimizes the data model, realizes intelligent adjustment and optimization of the sewage treatment process, improves water quality, reduces energy consumption, and improves equipment operating efficiency and overall treatment effect.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a sewage treatment data optimization method and system based on cloud computing. Background Art

[0002] Traditional systems rely on local servers or simple automated control systems to process data. These systems may not be able to handle large amounts of complex or real-time data, resulting in data processing delays or reduced accuracy. Compared with cloud computing platforms, traditional systems may have bottlenecks in data storage, computing power, and data transmission speed, limiting the efficient use of data. Traditional sewage treatment technology relies on manually set parameters and rules to operate, lacking the ability to adapt and intelligently adjust. This means that it may not be able to respond quickly when faced with changes in water quality, equipment failures, or changes in operating conditions, resulting in unstable treatment results or increased energy consumption.

[0003] It is difficult to achieve seamless sharing and integration of data between different devices or systems, resulting in data silos. This limits the comprehensive analysis and optimal use of data, making it difficult for decision makers to make decisions based on a global perspective. More on-site operation and maintenance personnel and equipment maintenance work may be required, increasing operating costs and time costs.

[0004] At the same time, due to system complexity and slow technology updates, the maintenance and upgrade process may be more time-consuming.

[0005] Lack of real-time data analysis and optimization capabilities make it impossible to dynamically adjust treatment strategies based on current data to achieve results. At the same time, due to the lack of advanced prediction models and technical support, traditional systems are difficult to accurately predict and warn of future water quality changes or equipment status. As the scale of treatment and treatment needs increase, there may be a problem of insufficient scalability, making it difficult to meet long-term development needs. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a sewage treatment data optimization method and system based on cloud computing, which realizes comprehensive monitoring and optimized management of the sewage treatment process.

[0007] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0008] In a first aspect, a sewage treatment data optimization method based on cloud computing is provided, the method comprising:

[0009] Receive real-time key information collected by monitoring equipment, including water quality indicators, equipment operating parameters, and energy consumption data, and pre-process the key information to obtain a processed data set;

[0010] Based on the processed data set, a cloud database was built and used to perform data analysis, including trend prediction and anomaly detection, to obtain analysis results;

[0011] According to the analysis results, a sewage treatment data optimization model is constructed, and water quality standards, equipment operation efficiency, and energy consumption cost parameters are input into the sewage treatment data optimization model to generate an optimization plan;

[0012] Integrate the optimization plan into the sewage treatment process and collect real-time data on water quality and energy consumption after the optimization plan is implemented;

[0013] Based on the collected effect data, the sewage treatment data optimization model is evaluated to obtain the evaluation results, and the sewage treatment data optimization model is adjusted and optimized based on the evaluation results.

[0014] Furthermore, the real-time key information collected by the monitoring equipment is received, including water quality indicators, equipment operating parameters, and energy consumption data, and the key information is pre-processed to obtain a processed data set, including:

[0015] Divide the real-time critical information into multiple components, each containing a fixed number of samples;

[0016] Initialize the parameters of the preprocessing model, including the mean and standard deviation, missing value filling scheme, and identify and handle outliers and missing values ​​in each component data;

[0017] According to the mean and standard deviation of the current pretreatment model, the component data are standardized to make the data evenly distributed, and the key features that affect the sewage treatment process are extracted according to the water quality indicators and equipment operating parameters to obtain the standardization and feature extraction results;

[0018] According to the results of standardization and feature extraction, the parameters of the preprocessing model are updated, and the standardization, feature extraction and parameter updating process are repeated until the preset number of iterations is reached, and all the preprocessed component data are merged to obtain the processed data set.

[0019] Furthermore, a cloud database was constructed based on the processed data set, and data analysis was performed using the cloud database, including trend prediction and anomaly detection, to obtain analysis results, including:

[0020] Input the processed data set into the cloud database and determine the goals of data analysis, including trend prediction and anomaly detection;

[0021] According to the goal of data analysis, determine the initial parameters of the time series model, including weights and biases, and initialize the parameters of the optimizer, including learning rate and momentum term coefficient;

[0022] The time series model is trained using historical data in the cloud database. During the training process, the learning rate of each parameter is dynamically adjusted based on the first-order moment estimate and the second-order moment estimate of the gradient.

[0023] The prediction performance of the time series model is evaluated by symmetric percentage error to obtain the evaluation results, and the parameters of the time series model are adjusted and optimized according to the evaluation results to obtain the trained time series model;

[0024] Use the trained time series model to predict future water quality indicators and key information of equipment operating parameters and generate trend prediction results;

[0025] According to the trend prediction results, set the abnormal threshold to distinguish normal data from abnormal data, and perform anomaly detection on the real-time data in the cloud database to identify anomalies in the data and generate anomaly detection results, including abnormal data points, anomaly types, abnormal time and other information;

[0026] Based on the trend prediction results and anomaly detection results, the sewage treatment process is analyzed to identify potential problems, including equipment failure and water quality deterioration, to obtain analysis results.

[0027] Furthermore, the calculation formula of the symmetric percentage error is:

[0028] ;

[0029] in,

[0030] ;

[0031] Indicates symmetrical percentage error; represents the overall scaling factor; represents the total number of samples; Represents an index variable; represents the time decay rate parameter; Indicates the maximum timestamp among all samples; Indicates The timestamp of each sample; represents the adjustment factor; Indicates The prediction error of samples; , represents a predefined error threshold; Indicates the offset; Indicates The predicted value of samples; Indicates The actual value of samples; represents a positive number; Indicates a condition.

[0032] Furthermore, based on the analysis results, a sewage treatment data optimization model is constructed, and water quality standards, equipment operating efficiency, and energy consumption cost parameters are input into the sewage treatment data optimization model to generate an optimization plan, including:

[0033] According to the analysis results, including water quality data, equipment operation data, and energy consumption data, the objective function of the sewage treatment data optimization model and the decision variables of the sewage treatment data optimization model are determined to construct the constraint conditions of the sewage treatment data optimization model and obtain the sewage treatment data optimization model framework;

[0034] Input water quality standards, equipment operating efficiency and energy consumption cost parameters into the sewage treatment data optimization model framework, and set the initial values ​​of the decision variables of the sewage treatment data optimization model to obtain the final sewage treatment data optimization model;

[0035] According to the final sewage treatment data optimization model, set the relevant parameters of ant colony optimization, including population size, number of iterations, and mutation factor, and regard each ant as a solution;

[0036] According to the concentration of pheromones and heuristic information in the current environment, the position of the ants is updated, and the objective function value corresponding to each ant is calculated in real time;

[0037] The position of each ant is continuously updated according to the concentration of pheromones and heuristic information in the current environment, and the objective function value of each ant at the new position is calculated in real time. When the predetermined number of iterations is reached, the final solution is determined according to the objective function value of each ant.

[0038] Based on the final solution, a sewage treatment optimization plan is generated, including the setting of water quality treatment parameters, expected water quality indicators, equipment operation efficiency and energy consumption cost.

[0039] Furthermore, the calculation formula of the objective function value is:

[0040] ;

[0041] in, represents the objective function value; , , represents the weight coefficient; Indicates the total number of water quality indicators; , Represents an index variable; Indicates The weight coefficient of each water quality index; Indicates The actual value of each water quality indicator; Indicates Baseline values ​​of water quality indicators; Indicates Deviation penalty coefficient of each water quality index; Indicates the actual operating efficiency of the equipment; Indicates the maximum possible efficiency of the equipment; Indicates the total number of energy consumption types; Indicates The weight coefficient of energy consumption; Indicates The cost of energy consumption; Indicates The cost increasing coefficient of energy consumption; Indicates The benchmark cost of energy consumption; An exponent representing the efficiency penalty.

[0042] Furthermore, based on the collected effect data, the sewage treatment data optimization model is evaluated to obtain evaluation results, and the sewage treatment data optimization model is adjusted and optimized based on the evaluation results, including:

[0043] Using effect data and evaluation indicators, the effect of the sewage treatment data optimization model is quantitatively evaluated.

[0044] Calculate the actual value of each evaluation indicator, and compare the actual value of each evaluation indicator with the preset benchmark value to obtain the evaluation result, including the actual value of each evaluation indicator and the comparison result with the preset benchmark value;

[0045] Based on the evaluation results, formulate an adjustment and optimization plan for the sewage treatment data optimization model, and adjust and optimize the sewage treatment data optimization model according to the adjustment and optimization plan.

[0046] In a second aspect, a sewage treatment data optimization system based on cloud computing includes:

[0047] The acquisition module is used to receive real-time key information collected by the monitoring equipment, including water quality indicators, equipment operating parameters, and energy consumption data, and pre-process the key information to obtain a processed data set;

[0048] The analysis module is used to build a cloud database based on the processed data set and use the cloud database to perform data analysis, including trend prediction and anomaly detection, to obtain analysis results;

[0049] A construction module is used to construct a sewage treatment data optimization model based on the analysis results, and input water quality standards, equipment operation efficiency, and energy consumption cost parameters into the sewage treatment data optimization model to generate an optimization plan;

[0050] The optimization module is used to integrate the optimization scheme into the sewage treatment process and collect the water quality and energy consumption effect data after the optimization scheme is implemented in real time;

[0051] The processing module is used to evaluate the effect of the sewage treatment data optimization model based on the collected effect data, obtain the evaluation results, and adjust and optimize the sewage treatment data optimization model based on the evaluation results.

[0052] According to a third aspect, a computing device includes:

[0053] one or more processors;

[0054] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0055] In a fourth aspect, a computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to implement the method described.

[0056] The above solution of the present invention includes at least the following beneficial effects:

[0057] By receiving key information such as water quality indicators, equipment operating parameters, energy consumption data, etc. collected by monitoring equipment in real time, and relying on the powerful storage and computing capabilities of the cloud database, efficient data processing and analysis are achieved. This not only improves the speed of data processing, but also ensures the accuracy and integrity of the data. Using the cloud database for data analysis, including trend prediction and anomaly detection, can timely identify potential problems and risks in the sewage treatment process. At the same time, a sewage treatment data optimization model is constructed to generate optimization plans based on parameters such as water quality standards, equipment operating efficiency, and energy consumption costs, thus realizing intelligent management of sewage treatment.

[0058] By integrating the optimization scheme into the sewage treatment process, the effect and efficiency of sewage treatment are effectively improved by adjusting the equipment operating parameters in real time and optimizing the treatment process. At the same time, the water quality and energy consumption effect data after the implementation of the optimization scheme are collected in real time, providing a feedback mechanism for the evaluation and optimization model, ensuring the continuous optimization and improvement of the sewage treatment process. Through the guidance of data analysis and optimization models, resources can be allocated more reasonably and equipment operation strategies can be adjusted, thereby reducing energy consumption and operating costs in the sewage treatment process. This not only improves the economic benefits of the enterprise, but also helps to achieve the goal of energy conservation and emission reduction. The introduction of cloud computing technology makes the sewage treatment system more scalable and flexible. With the increase in treatment scale and treatment demand, the system can easily cope with and expand the corresponding treatment capacity; at the same time, when facing new treatment technologies or equipment, the system can also quickly adapt to and integrate these new technologies or equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flow chart of a sewage treatment data optimization method and system based on cloud computing provided by an embodiment of the present invention.

[0060] Figure 2 It is a schematic diagram of a sewage treatment data optimization system based on cloud computing provided by an embodiment of the present invention.

[0061] Description of reference numerals:

[0062] 20. Sewage treatment data optimization system based on cloud computing; 21. Acquisition module; 22. Analysis module; 23. Construction module; 24. Optimization module; 25. Processing module. DETAILED DESCRIPTION

[0063] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary 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 limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0064] like Figure 1 As shown, an embodiment of the present invention provides a sewage treatment data optimization method based on cloud computing, the method comprising the following steps:

[0065] Step 11, receiving real-time key information collected by the monitoring equipment in real time, including water quality indicators, equipment operating parameters, and energy consumption data, and preprocessing the key information to obtain a processed data set;

[0066] Step 12, building a cloud database based on the processed data set, and using the cloud database to perform data analysis, including trend prediction and anomaly detection, to obtain analysis results;

[0067] Step 13: construct a sewage treatment data optimization model based on the analysis results, and input water quality standards, equipment operation efficiency, and energy consumption cost parameters into the sewage treatment data optimization model to generate an optimization plan;

[0068] Step 14, integrating the optimization scheme into the sewage treatment process, and collecting water quality and energy consumption effect data after implementing the optimization scheme in real time;

[0069] Step 15, based on the collected effect data, the sewage treatment data optimization model is evaluated to obtain an evaluation result, and the sewage treatment data optimization model is adjusted and optimized based on the evaluation result.

[0070] In the embodiment of the present invention, by receiving key information such as water quality indicators, equipment operating parameters, energy consumption data collected by monitoring equipment in real time, and utilizing the powerful data processing capabilities of the cloud database, efficient storage, processing and analysis of data are achieved. This improves the speed and accuracy of data processing and provides strong support for the optimization of the sewage treatment process. By performing data analysis through the cloud database, including trend prediction and anomaly detection, the method can timely discover potential problems and risks in the sewage treatment process and provide managers with a scientific basis for decision-making. At the same time, the constructed sewage treatment data optimization model can generate optimization plans based on parameters such as water quality standards, equipment operating efficiency, and energy consumption costs, thus realizing intelligent management of sewage treatment.

[0071] This method effectively improves the effect and efficiency of sewage treatment by integrating the optimization scheme into the sewage treatment process, adjusting the equipment operating parameters in real time, optimizing the treatment process, etc. This not only improves the water quality compliance rate, but also reduces energy consumption and operating costs. With the guidance of data analysis and optimization models, resources can be allocated more reasonably and equipment operation strategies can be adjusted, thereby reducing energy consumption and operating costs in the sewage treatment process. The introduction of cloud computing technology has made the sewage treatment system more flexible and scalable. This method can easily cope with changes in treatment scale and treatment needs, quickly integrate new technologies or equipment, and provide a strong guarantee for the long-term development of the sewage treatment industry. By optimizing the sewage treatment process and effect, this method helps to reduce the discharge of pollutants and protect water resources, thereby promoting environmental protection and sustainable development.

[0072] In a preferred embodiment of the present invention, the above step 11, receiving real-time key information collected by the monitoring equipment in real time, including water quality indicators, equipment operating parameters, energy consumption data, and pre-processing the key information to obtain a processed data set, may include:

[0073] Step 111, dividing the real-time key information into multiple components, each component containing a fixed number of samples;

[0074] Step 112, initializing the parameters of the preprocessing model, including the mean and standard deviation, missing value filling scheme, and identifying and processing outliers and missing values ​​in each component data;

[0075] Step 113, standardize the component data according to the mean and standard deviation of the current pretreatment model to make the data evenly distributed, and extract key features that affect the sewage treatment process according to water quality indicators and equipment operating parameters to obtain standardization and feature extraction results;

[0076] Step 114, based on the standardization and feature extraction results, update the parameters of the preprocessing model, and repeat the standardization, feature extraction and parameter updating process until a preset number of iterations is reached, and merge all the preprocessed component data to obtain a processed data set.

[0077] In the embodiment of the present invention, in step 111, the real-time key information is divided into multiple components. , each component contains a fixed number of samples.

[0078] Step 111, initialize the parameters of the preprocessing model, including mean, standard deviation, missing value filling scheme (such as mean filling), and replace the missing values ​​in the data set with the mean of the component or the entire data set. According to the characteristics of the data and business needs, set one or more thresholds to identify outliers. For example, for a certain water quality indicator, an upper limit and a lower limit can be set, and values ​​outside this range are considered outliers. Calculate the component The mean and standard deviation , and then identify those values ​​that differ from the mean by more than 3 standard deviations as outliers, that is:

[0079] Outliers ;in, Indicates the amount A sample value in . Remove outliers, for components For example, using the component The mean to fill missing values.

[0080] Step 113, standardize the component data according to the mean and standard deviation of the current preprocessing model. Next, select water quality indicators and equipment operating parameters from the standardized data set as features to construct a feature matrix. Each row of the matrix represents a sample and each column represents a feature. Calculate the covariance matrix of the feature matrix.

[0081] Specifically, assuming that the feature matrix is a The matrix of is the sample size, is the number of features. The matrix It is expressed as: ; Among them, the first line [ , , …, ] represents the feature vector of the first sample, where is the value of the first sample on the first feature, is the value of the first sample on the second feature, and so on, until The first sample is The second line [ , , …, ] represents the feature vector of the second sample, where the meaning of each element is similar to that of the first row. And so on until the OK[ , , …, ] represents the The feature vector of the samples.

[0082] Covariance matrix ;in, Represents the feature matrix The mean row vector of is expanded into a matrix, that is, each row is the mean of the eigenvector; Represents the transpose of the matrix. The covariance matrix reflects the correlation between features. Perform eigenvalue decomposition to obtain eigenvalues ​​and corresponding eigenvectors. The eigenvalue represents the amount of information (i.e. variance) carried by each principal component, and the eigenvector defines the direction of the principal component.

[0083] According to the size of the eigenvalue, select the eigenvector corresponding to the eigenvalue as the principal component. Use the selected principal component eigenvector to construct the principal component matrix. Projection to principal component matrix On the above, we get the data matrix after dimensionality reduction . Analyze the principal component matrix The weight of each feature in the main component is used to understand the combination of water quality indicators and equipment operating parameters represented by each principal component. Based on the interpretability of the principal component, the key features that affect the sewage treatment process are identified. Based on the results of the principal component analysis, the features that have a large weight in the principal component and have an impact on the sewage treatment process are selected as key features.

[0084] Step 114, update the parameters of the preprocessing model according to the results of standardization and feature extraction. Repeat the standardization, feature extraction and parameter update process until the preset number of iterations is reached. Finally, merge all the preprocessed component data to obtain a processed data set.

[0085] Dividing real-time critical information into multiple components, each containing a fixed number of samples, helps to better manage and process data. This division makes data processing more flexible, and each component can be processed independently to improve processing efficiency. The quality of the data can be significantly improved by initializing the parameters of the preprocessing model and identifying and processing outliers and missing values ​​in each component data. The presence of outliers will affect the accuracy and reliability of the data, while missing values ​​may cause incomplete data. By effectively handling these issues, the accuracy and completeness of the data can be ensured. According to the mean and standard deviation of the current preprocessing model, the component data is standardized to make the data evenly distributed.

[0086] Extract key features that affect the sewage treatment process based on water quality indicators, equipment operating parameters, etc. This step helps to simplify the data and reduce the dimensionality of the data while retaining the most valuable information for modeling and analysis. Update the parameters of the preprocessing model based on the standardization and feature extraction results. This step ensures that the model is always trained based on the latest data, improving the accuracy and reliability of the model. The data processing results can be further optimized by repeating the standardization, feature extraction, and parameter update processes until the preset number of iterations is reached. This iterative processing method helps to gradually approach the true distribution of the data and improve the predictive ability of the model. Merge all preprocessed component data to obtain the processed data set. This step provides a unified, high-quality data foundation for subsequent modeling and analysis, which helps to improve the overall analysis effect.

[0087] In a preferred embodiment of the present invention, the above step 12, based on the processed data set, constructs a cloud database, and uses the cloud database to perform data analysis, including trend prediction and anomaly detection, to obtain analysis results, which may include:

[0088] Step 121, inputting the processed data set into a cloud database and determining the goals of data analysis, including trend prediction and anomaly detection;

[0089] Step 122, according to the goal of data analysis, determine the initial parameters of the time series model, including weights and biases, and initialize the parameters of the optimizer, including learning rate and momentum term coefficient;

[0090] Step 123, using historical data in the cloud database to train the time series model, during the training process, dynamically adjusting the learning rate of each parameter according to the first-order moment estimation and the second-order moment estimation of the gradient;

[0091] Step 124, evaluating the prediction performance of the time series model by symmetric percentage error to obtain an evaluation result, and adjusting and optimizing the time series model parameters according to the evaluation result to obtain a trained time series model;

[0092] Step 125, using the trained time series model to predict future water quality indicators and key information of equipment operating parameters, and generate trend prediction results;

[0093] Step 126, according to the trend prediction result, an abnormal threshold is set to distinguish normal data from abnormal data, and anomaly detection is performed on the real-time data in the cloud database to identify abnormal situations in the data and generate anomaly detection results, including abnormal data points, abnormal types, abnormal time and other information;

[0094] Step 127, analyzing the sewage treatment process based on the trend prediction results and the anomaly detection results, identifying potential problems, including equipment failure and water quality deterioration, to obtain analysis results.

[0095] In the embodiment of the present invention, a time series model (such as LSTM) is selected according to data characteristics and prediction targets, and initial parameters are set, including weights (randomly initialized) and biases (initialized to zero).

[0096] Choose an optimizer (such as Adam) and set its parameters, such as the learning rate (which controls the step size of parameter updates), the momentum term coefficient (which helps speed up convergence and reduce oscillations), etc.

[0097] Step 123, extract historical data from the cloud database, which contains time series information, such as water quality indicators, equipment operating parameters, etc. The extracted data is divided into a training set and a validation set for time series model training and performance evaluation. Use the training set data to train the selected model (such as the LSTM time series model). During the training process, the loss between the predicted value and the true value is calculated by forward propagation, and then the gradient is calculated by the back propagation algorithm, and the time series model parameters are updated to minimize the loss. In order to improve training efficiency and model performance, the learning rate of each parameter is dynamically adjusted according to the first-order moment estimate (i.e., gradient mean) and second-order moment estimate (i.e., gradient variance) of the gradient. The update rule using the Adam optimizer can be expressed as:

[0098] ;

[0099] ;

[0100] ;

[0101] ;

[0102] .

[0103] in, and They are the first and second moment estimates of the gradient, respectively; , is the decay rate, which is used to control the exponentially weighted average of the first-order moment and the second-order moment; and are bias-corrected first- and second-order moment estimates; is the learning rate; is a positive number to prevent the denominator from being zero; is a time series model parameter. By dynamically adjusting the learning rate, the Adam optimizer can converge to the optimal solution faster and can adaptively adjust the learning rate for different parameters, thereby improving training efficiency and model performance.

[0104] Step 124, evaluate the model using the validation set data, calculate the symmetric percentage error to quantify the prediction performance of the model. According to the evaluation results, adjust and optimize the parameters of the model, such as adjusting the weight, bias, learning rate, etc.

[0105] Step 125, prepare the input data for the future time points to be predicted (such as the last few time points of historical data). Input the input data into the trained time series model to obtain the predicted values ​​of future water quality indicators and equipment operating parameters. Save the prediction results to the cloud database or export them as files.

[0106] Step 126, according to the trend prediction results and actual needs, set an abnormal threshold to distinguish normal data from abnormal data. Obtain real-time data from the cloud database. Compare the real-time data with the threshold to identify abnormal data points, abnormal types, abnormal time and other information. Save the abnormal detection results to the cloud database.

[0107] Step 127, the trend prediction results and the anomaly detection results are integrated to form a complete analysis report. Through the analysis report, potential problems that may exist in the sewage treatment process, such as equipment failure, water quality deterioration, etc., are identified. According to the identified problems, corresponding suggestions or solutions are proposed, such as repairing equipment, adjusting process parameters, etc. The analysis report is saved to the cloud database or exported as a file.

[0108] Suppose there is a dataset of a sewage treatment plant, which contains time series data of water quality indicators (such as pH value, dissolved oxygen) and equipment operating parameters (such as pump speed, motor temperature) for the past year. Use this data to predict the water quality and equipment status for the next week and detect any possible anomalies. Import the processed dataset into a MySQL database, predict the pH value and pump speed for the next week, and detect any possible anomalies. Select the LSTM model as the time series prediction model. Initialize the weights and bias of the LSTM model using the random initialization method. Select the Adam optimizer and set the learning rate to 0.001 and the momentum term coefficients to 0.9 and 0.999. Extract the data for the past year from the MySQL database as the training set. Use the training set data to train the LSTM model and update the model parameters through the back propagation algorithm. During the training process, use the Adam optimizer to dynamically adjust the learning rate of each parameter. Use the validation set data to evaluate the LSTM model and calculate the symmetric percentage error to be 5%. According to the evaluation results, adjust the parameters of the LSTM model, such as increasing the number of hidden layer units and adjusting the learning rate. After multiple iterations of training and optimization, the trained LSTM model is obtained.

[0109] Prepare input data for the next week (such as pH value and pump speed data for the past week). Input the input data into the trained LSTM model to obtain the predicted values ​​of pH value and pump speed for the next week. Save the predicted results in the MySQL database. According to the predicted results and actual needs, set the abnormal threshold of pH value to [6.5, 8.5] and the abnormal threshold of pump speed to [50,100]. Get real-time data from the MySQL database and compare it with the abnormal threshold. Identify abnormal data points, such as the pH value of 9.0 on a certain day, which exceeds the abnormal threshold range and is marked as an abnormal data point. Save the abnormal detection results in the MySQL database. Integrate the prediction results and abnormal detection results to form an analysis report. Through the analysis report, identify potential problems that may exist in the sewage treatment process, such as a pump may fail, causing the pump speed to increase abnormally. Propose corresponding suggestions or solutions, such as inspecting and repairing the pump.

[0110] Entering the processed data set into the cloud database can achieve efficient management and fast access to the data. Cloud databases are usually highly available and scalable, and can cope with the storage and query needs of large amounts of data. By determining the initial parameters of the time series model according to the data analysis goals and training with historical data in the cloud database, the model can learn the time series characteristics in the data. Combined with the strategy of dynamically adjusting the learning rate, the model can converge to the optimal solution more quickly and improve the accuracy of the prediction. Using the trained trend prediction model, key information such as future water quality indicators and equipment operating parameters can be predicted to provide support for decision-making in the sewage treatment process.

[0111] Setting anomaly thresholds based on trend prediction results can more accurately distinguish normal data from abnormal data and reduce false positives and false negatives. Anomaly detection of real-time data in the cloud database can timely detect anomalies in the data, including abnormal data points, abnormal types, and abnormal time, providing strong support for timely response and handling of potential problems. Based on trend prediction results and anomaly detection results, the sewage treatment process can be deeply analyzed to identify potential equipment failures, water quality deterioration and other problems. This helps to take timely measures to repair or adjust, avoid further deterioration of the problem, and improve the stability and efficiency of the sewage treatment process. The results obtained through data analysis can provide a basis for operational decisions of sewage treatment plants. For example, adjusting treatment process parameters according to predicted water quality indicators, or arranging maintenance plans according to equipment operating status, thereby optimizing operating costs and resource utilization. The entire data analysis process is based on the processed data set and cloud database, realizing a data-driven decision-making method. This helps to promote the intelligent and automated development of sewage treatment processes and improve overall operational levels and efficiency.

[0112] In a preferred embodiment of the present invention, the calculation formula of the symmetric percentage error is:

[0113] ;

[0114] in,

[0115] ;

[0116] Indicates symmetrical percentage error; represents the overall scaling factor; represents the total number of samples; Represents an index variable; represents the time decay rate parameter; Indicates the maximum timestamp among all samples; Indicates The timestamp of each sample; represents the adjustment factor; Indicates The prediction error of samples; , represents a predefined error threshold; Indicates the offset; Indicates The predicted value of samples; Indicates The actual value of samples; represents a positive number; Indicates a condition.

[0117] In an embodiment of the present invention, the overall scaling factor ( ), time decay rate parameter ( )、Adjustment coefficient( ), a predefined error threshold ( and ), offset ( ) and positive numbers ( ). Get the total number of samples ( ), and the timestamp of each sample ( ), actual value ( ) and predicted values ​​( ).

[0118] Find the maximum timestamp among all samples ( ). For each sample ( From 1 to ), calculate the prediction error For each sample , if | | ≤ ,but = 1. If , then calculate the temporary value of the symmetric percentage error ,set up = .if > ,but .

[0119] For each sample , calculate the time decay factor . Initialize the value of S to 0.

[0120] For each sample Calculate the sample pair Contribution . Add the contribution value to After the calculation is completed, This is the final value of the symmetrical percentage error. The values ​​are output as an evaluation of the symmetrical percentage error.

[0121] pass The symmetric percentage error divides the prediction error into three intervals and gives different weights to each interval. This detailed error classification method can more accurately reflect the performance of the prediction model in different error ranges. , so that samples farther away from the current time contribute less to the evaluation. This is in line with the actual situation, because newer data can better reflect the current performance of the model.

[0122] Overall scaling factor , adjustment coefficient , Error threshold and , offset Parameters such as ε and positive number can be adjusted according to actual needs. This allows the symmetric percentage error to adapt to different evaluation scenarios and data characteristics. As a comprehensive performance indicator, the symmetric percentage error can fully reflect the accuracy and stability of the prediction model. In the fields of sewage treatment, financial forecasting, weather forecasting, etc., the symmetric percentage error can help decision makers better understand the performance of the prediction model and make reasonable decisions.

[0123] In a preferred embodiment of the present invention, the above step 13, based on the analysis results, constructs a sewage treatment data optimization model, and inputs water quality standards, equipment operating efficiency, and energy consumption cost parameters into the sewage treatment data optimization model to generate an optimization plan, which may include:

[0124] Step 131, according to the analysis results, including water quality data, equipment operation data, and energy consumption data, determine the objective function of the sewage treatment data optimization model and the decision variables of the sewage treatment data optimization model, so as to construct the constraint conditions of the sewage treatment data optimization model and obtain the sewage treatment data optimization model framework;

[0125] Step 132, inputting water quality standards, equipment operating efficiency and energy consumption cost parameters into the sewage treatment data optimization model framework, and setting initial values ​​of decision variables of the sewage treatment data optimization model to obtain a final sewage treatment data optimization model;

[0126] Step 133, according to the final sewage treatment data optimization model, set the relevant parameters of ant colony optimization, including population size, number of iterations, and mutation factor, and regard each ant as a solution;

[0127] Step 134, updating the position of the ants according to the concentration of pheromones in the current environment and the heuristic information, and calculating the objective function value corresponding to each ant in real time;

[0128] Step 135, continuously updating the position of each ant according to the concentration of pheromones in the current environment and heuristic information, and calculating the objective function value of each ant at the new position in real time. When the predetermined number of iterations is reached, the final solution is determined according to the objective function value of each ant;

[0129] Step 136, generating a wastewater treatment optimization plan based on the final solution, including the setting of water quality treatment parameters, expected water quality indicators, equipment operation efficiency and energy consumption cost.

[0130] In an embodiment of the present invention, step 131 determines the main objectives of the sewage treatment data optimization model based on water quality data, equipment operation data and energy consumption data, such as minimizing energy consumption costs, maximizing equipment operation efficiency or achieving specific water quality standards. For example, the objective function can be the weighted minimization of energy consumption costs and water quality treatment effects. Select variables that have a direct impact on the sewage treatment process and can be adjusted to optimize the target, such as dosage, aeration time, sludge return ratio, etc. Based on actual requirements such as water quality standards, equipment operation restrictions and energy consumption costs, construct constraints for the sewage treatment data optimization model. For example, water quality indicators (such as COD, BOD, NH3-N, etc.) must meet the prescribed emission standards; equipment operation efficiency cannot be lower than a certain threshold; energy consumption costs cannot exceed the budget, etc.

[0131] Step 132, input water quality standards (such as the limits of various water quality indicators), equipment operating efficiency (such as pump efficiency, aerator efficiency, etc.) and energy consumption cost (such as electricity costs, drug costs, etc.) parameters into the wastewater treatment data optimization model framework. Set reasonable initial values ​​for decision variables, which can be based on historical data and experience values.

[0132] Step 133, determine the size of the ant colony, that is, the number of ants searching at the same time. Set the maximum number of iterations of the ant colony algorithm to ensure that the algorithm has enough time to find the optimal solution. The mutation mechanism is used to increase the diversity of the search and prevent the algorithm from falling into the local optimal solution. Each ant is regarded as a potential solution. The ant continuously updates its position during the search process, that is, updates the value of the decision variable.

[0133] Step 134, update the position of the ant according to the concentration of pheromones in the current environment (reflecting the quality of the path previously searched by the ant) ​​and the heuristic information (reflecting the estimated distance from the current position to the target position). For each ant's decision variable value at the new position, calculate the corresponding objective function value to evaluate the quality of the solution.

[0134] Step 135, continuously repeat step 134 until a predetermined number of iterations is reached. The final solution is determined based on the objective function value of each ant.

[0135] Step 136, according to the decision variable values ​​in the final solution, determine the settings of water quality treatment parameters, such as dosage, aeration time, etc. Based on the water quality treatment effect predicted by the final solution, the water quality model is used to predict the water quality treatment effect to obtain the expected water quality index value. The water quality model is established based on the actual sewage treatment process and can reflect the relationship between parameters such as dosage, aeration time and water quality index. The expected water quality index value ;in, Indicates dosage coefficient; represents the aeration time coefficient; Indicates aeration time; Indicates the dosage; Represents the intercept term of the water quality model.

[0136] Suppose there is a wastewater treatment plant and its operation needs to be optimized to minimize energy costs while ensuring that water quality meets discharge standards. Specifically:

[0137] The objective function is set to minimize the energy cost (the sum of electricity and medicine costs) while ensuring that the water quality index (such as COD) is below a certain threshold. The decision variables are selected as the dosage and aeration time. Constraints include that the water quality index must meet the standard and the equipment operation time cannot exceed the specified range. Input water quality standards (such as COD≤50mg / L), equipment operation efficiency (such as aerator efficiency ≥80%) and energy cost parameters (such as electricity unit price, medicine unit price). Set initial values ​​for the dosage and aeration time, such as the initial dosage is 100mg / L and the initial aeration time is 2 hours. Set the ant colony size to 50 ants, the number of iterations to 100 times, and introduce a variation factor of 0.1 to increase the search diversity. The ants start searching based on the pheromone concentration and heuristic information, update the values ​​of the dosage and aeration time in each iteration, and calculate the corresponding objective function value (energy cost). After 100 iterations, the final solution is determined based on the objective function value of each ant, that is, the combination of dosage and aeration time with the lowest energy cost and water quality that meets the standard. An optimization plan is generated based on the final solution, such as adjusting the dosage to 80 mg / L and the aeration time to 1.5 hours. It is expected that under this optimization plan, the energy cost will be reduced by 10%, and the water quality indicator COD will be stabilized below 40 mg / L, meeting the emission standards.

[0138] By precisely controlling water treatment parameters, pollutants in sewage can be removed more effectively, effluent quality can be improved, and stricter water quality standards can be met. Optimizing equipment operating parameters can improve equipment operating efficiency, reduce equipment idling and ineffective operation, and thus improve the overall sewage treatment system's processing capacity. By reasonably controlling the dosage and aeration time, etc., the use of chemicals and energy consumption can be reduced, and the operating cost of the sewage treatment process can be reduced. The optimization plan helps reduce pollutant emissions.

[0139] In a preferred embodiment of the present invention, the calculation formula of the objective function value is:

[0140] ;

[0141] in, represents the objective function value; , , represents the weight coefficient; Indicates the total number of water quality indicators; , Represents an index variable; Indicates The weight coefficient of each water quality index; Indicates The actual value of each water quality indicator; Indicates Baseline values ​​of water quality indicators; Indicates Deviation penalty coefficient of each water quality index; Indicates the actual operating efficiency of the equipment; Indicates the maximum possible efficiency of the equipment; Indicates the total number of energy consumption types; Indicates The weight coefficient of energy consumption; Indicates The cost of energy consumption; Indicates The cost increasing coefficient of energy consumption; Indicates The benchmark cost of energy consumption; An exponent representing the efficiency penalty.

[0142] In the embodiment of the present invention, the total number of water quality indicators is determined and the total number of energy consumption types For each water quality indicator Determine the weight coefficient , actual value , Baseline value and the deviation penalty coefficient . Determine the actual operating efficiency of the equipment and maximum possible efficiency , and the efficiency penalty index For each energy consumption Determine the weight coefficient ,cost , cost increasing coefficient and baseline cost . Set the weight coefficient , , , which represent the relative importance of water quality index deviation, equipment operation efficiency and energy consumption cost in the objective function. , calculate the deviation ratio ( ). Square the deviation ratio and multiply it by the corresponding weight coefficient and the deviation penalty coefficient . Sum all water quality indicators to get the weighted sum of water quality indicator deviations . Calculate the ratio of the equipment's operating efficiency to its maximum possible efficiency . Set the new value of the ratio Raise the result to the power and subtract it from 1 to get the penalty term for equipment operating efficiency. For each energy consumption , calculate the cost With baseline cost and multiply it by the cost increment coefficient , then add 1. Multiply the above result by the cost of energy consumption and the corresponding weight coefficient . Sum all energy consumption types to get the weighted sum of energy consumption costs Multiply the weighted sum of water quality index deviation, equipment operation efficiency penalty term and energy consumption cost by their corresponding weight coefficients , , Add these three parts together to get the final objective function value .

[0143] By weight coefficient , , The setting can balance the performance of three key aspects: water quality index deviation, equipment operation efficiency and energy consumption cost. This helps to find a solution that achieves a balance between multiple objectives. By minimizing the value of the objective function, the system can perform well in water quality treatment, equipment operation and energy consumption management, thereby improving the overall treatment efficiency and economic benefits. The water quality index deviation part (composed of , , and composition) can reflect the degree of deviation between water quality indicators and benchmark values, and through the deviation penalty coefficient Increase the penalties for larger deviations. This helps motivate the system to more strictly control water quality and ensure that water quality meets standards. , and The energy cost part (composed of) can reflect the gap between the actual operating efficiency of the equipment and the maximum possible efficiency. By optimizing the equipment operating parameters, this gap can be narrowed, the equipment efficiency can be improved, and the energy consumption can be reduced. Optimizing the equipment operating efficiency can also help reduce equipment wear and failure rate, thereby extending the equipment life and reducing maintenance costs. , , and Energy consumption (composition) can reflect the cost of various energy consumption and its incremental situation in detail. This helps to manage energy consumption more finely and identify and optimize energy consumption items with high costs or high incremental rates. By optimizing energy consumption costs, energy consumption and emissions can be reduced.

[0144] In a preferred embodiment of the present invention, the above step 14, integrating the optimization scheme into the sewage treatment process and collecting water quality and energy consumption effect data after implementing the optimization scheme in real time, may include:

[0145] In an embodiment of the present invention, the sewage treatment process is adjusted according to the optimization scheme, including changing the operating conditions of the treatment unit, adjusting the dosage, optimizing the aeration volume, etc. The equipment that needs to be upgraded is replaced or modified, and the integration and coordinated operation of the new equipment with the entire treatment process are ensured. In order to collect data in real time, an automation system can be built or upgraded, including installing sensors, instruments, data acquisition systems, etc., for monitoring water quality indicators, equipment operating parameters and energy consumption. Monitoring points are set at key nodes of the sewage treatment process, such as water inlets, entrances and exits of each treatment unit, and water outlets, etc., to ensure that changes in water quality and energy consumption can be fully reflected. Corresponding sensors and instruments are configured at the monitoring points for real-time measurement of water quality indicators (such as COD, BOD, NH3-N, etc.), equipment operating parameters (such as flow, pressure, temperature, etc.) and energy consumption data (such as power consumption, drug consumption, etc.). A data acquisition system is constructed to collect, transmit and store the data collected by sensors and instruments in real time. Wired or wireless transmission methods can be used to ensure the timeliness and accuracy of data. A real-time monitoring platform is established to display and analyze the collected data in real time. Set up an alarm mechanism to issue an alarm message in time when water quality indicators or energy consumption data exceed the set threshold so that timely measures can be taken to make adjustments. Conduct in-depth analysis of the collected water quality and energy consumption data to identify the key factors affecting treatment effects and energy consumption, and evaluate the implementation effect of the optimization plan.

[0146] According to the data analysis results, the optimization plan is further adjusted and optimized. The process parameters, equipment operation mode, dosage, etc. can be adjusted to further improve the treatment effect and reduce energy consumption. The optimized and adjusted plan is integrated into the sewage treatment process again, and real-time data collection continues for analysis and monitoring. A continuous improvement cycle is formed to continuously improve the performance and efficiency of the sewage treatment system. The effect of the optimization plan is regularly evaluated, including the water quality compliance rate, energy consumption reduction, equipment operation stability, etc.

[0147] In a preferred embodiment of the present invention, the above step 15, based on the collected effect data, evaluates the effect of the sewage treatment data optimization model to obtain an evaluation result, and adjusts and optimizes the sewage treatment data optimization model based on the evaluation result, which may include:

[0148] Step 151, using the effect data and the evaluation index, quantitatively evaluate the effect of the sewage treatment data optimization model, calculate the actual value of each evaluation index, and compare the actual value of each evaluation index with the preset benchmark value to obtain an evaluation result, including the actual value of each evaluation index and the comparison result with the preset benchmark value;

[0149] Step 152, based on the evaluation results, formulate an adjustment and optimization plan for the sewage treatment data optimization model, and adjust and optimize the sewage treatment data optimization model according to the adjustment and optimization plan.

[0150] In an embodiment of the present invention, evaluation indicators that can fully reflect the effect of sewage treatment are selected, such as water quality compliance rate, energy consumption reduction percentage, equipment operation stability index, etc. A clear calculation method and benchmark value are set for each evaluation indicator for quantitative evaluation. Water quality data, energy consumption data and equipment operation data in the sewage treatment process are obtained from the real-time data monitoring system. Ensure the integrity and accuracy of the data, and eliminate or correct abnormal data. According to the selected evaluation indicators and calculation methods, the actual values ​​of various evaluation indicators are calculated using the collected effect data. For example, the water quality compliance rate can be obtained by calculating the ratio of the number of qualified water quality samples to the total number of samples; the energy consumption reduction percentage can be obtained by comparing the energy consumption data before and after optimization. The calculated actual value of the evaluation indicator is compared with the preset benchmark value to analyze the difference. If the actual value is better than the benchmark value, it means that the optimization model is better; if the actual value does not reach the benchmark value, it is necessary to further analyze the cause and consider adjusting the optimization model. The actual values ​​of various evaluation indicators, the comparison results with the benchmark values, and the analysis conclusions are summarized to form an evaluation report.

[0151] Step 152, carefully read the evaluation report, and deeply understand the performance of each evaluation indicator and its difference from the benchmark value. Identify the problems and deficiencies in the optimization model and clarify the direction of improvement. Develop specific adjustment and optimization plans for the problems found in the evaluation. The plan includes adjusting the parameters of the optimization model and improving the data processing method. Ensure that the plan is feasible and targeted and can effectively improve the effect of the optimization model. According to the formulated adjustment and optimization plan, modify and improve the sewage treatment data optimization model. Test the modified model to ensure that it functions normally and has improved performance. After implementing the adjustment and optimization, continue to monitor the sewage treatment process in real time and collect data. Regularly re-evaluate the effect of the optimization model to ensure that the model can continue to adapt to changes in the sewage treatment process and maintain good performance. Evaluation indicators: water quality compliance rate (benchmark value: 95%), energy consumption reduction percentage (benchmark value: 10%), equipment operation stability index (benchmark value: 80 points). The water quality data obtained through the real-time data monitoring system show that in the past month, the number of water quality samples that met the standard was 900 and the total number of samples was 1000. Therefore, the water quality compliance rate is 90%. The average monthly energy consumption before optimization is 1000 kWh, and the average monthly energy consumption after optimization is 920 kWh. Therefore, the energy consumption reduction percentage is =8%. According to the equipment operation data, the equipment operation stability index this month is 75 points. The water quality compliance rate did not reach the benchmark value (95%), and further analysis of the problems in the water quality treatment process and adjustment of the optimization model are required. The energy consumption reduction percentage did not reach the benchmark value (10%), indicating that the optimization model still has room for improvement in energy consumption control.

[0152] The equipment operation stability index is lower than the benchmark value (80 points), indicating that the equipment operation stability needs to be improved. Adjust the water quality treatment process parameters, increase the residence time of key treatment units, and improve the treatment efficiency. Optimize equipment operation parameters, such as adjusting aeration volume, pumping frequency, etc., to reduce energy consumption. After implementation, the sewage treatment process was monitored in real time and data was collected again. After one month of operation and evaluation, the water quality compliance rate increased to 96%, the energy consumption reduction percentage reached 12%, and the equipment operation stability index increased to 85 points. The effect of the optimization model has been improved.

[0153] By quantifying the evaluation indicators, the effect of the sewage treatment data optimization model can be objectively and accurately reflected, providing data support for subsequent optimization. By comparing with the benchmark values, problems and deficiencies in the model can be discovered in a timely manner, indicating the direction for optimization. Through targeted optimization measures, the effect of sewage treatment can be improved, such as improving effluent quality, reducing energy consumption, and reducing equipment operation failures. The optimized model can usually make more efficient use of resources, reduce treatment costs, and improve economic benefits. By improving equipment configuration and strengthening data monitoring, the stability of the sewage treatment system can be enhanced and treatment interruptions caused by equipment failures or data anomalies can be reduced. The optimized sewage treatment system is more environmentally friendly and efficient, which helps promote the sustainable development of the sewage treatment industry and contribute to environmental protection and resource conservation.

[0154] like Figure 2 As shown, an embodiment of the present invention further provides a sewage treatment data optimization system 20 based on cloud computing, including:

[0155] The acquisition module 21 is used to receive real-time key information collected by the monitoring equipment in real time, including water quality indicators, equipment operating parameters, and energy consumption data, and pre-process the key information to obtain a processed data set;

[0156] The analysis module 22 is used to construct a cloud database based on the processed data set, and use the cloud database to perform data analysis, including trend prediction and anomaly detection, to obtain analysis results;

[0157] A construction module 23 is used to construct a sewage treatment data optimization model according to the analysis results, and input water quality standards, equipment operation efficiency, and energy consumption cost parameters into the sewage treatment data optimization model to generate an optimization plan;

[0158] The optimization module 24 is used to integrate the optimization scheme into the sewage treatment process and collect water quality and energy consumption effect data after the optimization scheme is implemented in real time;

[0159] The processing module 25 is used to evaluate the effect of the sewage treatment data optimization model according to the collected effect data, obtain the evaluation result, and adjust and optimize the sewage treatment data optimization model according to the evaluation result.

[0160] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0161] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0162] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0163] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A sewage treatment data optimization method based on cloud computing, characterized in that: The method comprises: Receive real-time key information collected by monitoring equipment, including water quality indicators, equipment operating parameters, and energy consumption data, and pre-process the key information to obtain a processed data set; Based on the processed data set, a cloud database was built and used to perform data analysis, including trend prediction and anomaly detection, to obtain analysis results; According to the analysis results, a sewage treatment data optimization model is constructed, and water quality standards, equipment operation efficiency, and energy consumption cost parameters are input into the sewage treatment data optimization model to generate an optimization plan; Integrate the optimization plan into the sewage treatment process and collect real-time data on water quality and energy consumption after the optimization plan is implemented; Based on the collected effect data, the sewage treatment data optimization model is evaluated to obtain the evaluation results, and the sewage treatment data optimization model is adjusted and optimized according to the evaluation results; Among them, a cloud database was built based on the processed data set, and data analysis was performed using the cloud database, including trend prediction and anomaly detection, to obtain analysis results, including: Input the processed data set into the cloud database and determine the goals of data analysis, including trend prediction and anomaly detection; According to the goal of data analysis, determine the initial parameters of the time series model, including weights and biases, and initialize the parameters of the optimizer, including learning rate and momentum term coefficient; The time series model is trained using historical data in the cloud database. During the training process, the learning rate of each parameter is dynamically adjusted based on the first-order moment estimate and the second-order moment estimate of the gradient. The prediction performance of the time series model is evaluated by symmetric percentage error to obtain the evaluation results, and the parameters of the time series model are adjusted and optimized according to the evaluation results to obtain the trained time series model; Use the trained time series model to predict future water quality indicators and key information of equipment operating parameters and generate trend prediction results; According to the trend prediction results, set the abnormal threshold to distinguish normal data from abnormal data, and perform anomaly detection on the real-time data in the cloud database to identify anomalies in the data and generate anomaly detection results, including abnormal data points, anomaly types, abnormal time and other information; Based on the trend prediction results and anomaly detection results, the sewage treatment process is analyzed to identify potential problems, including equipment failure and water quality deterioration, to obtain analysis results.

2. The sewage treatment data optimization method based on cloud computing according to claim 1 is characterized in that: Receive real-time key information collected by monitoring equipment, including water quality indicators, equipment operating parameters, and energy consumption data, and pre-process the key information to obtain a processed data set, including: Divide the real-time critical information into multiple components, each containing a fixed number of samples; Initialize the parameters of the preprocessing model, including the mean and standard deviation, missing value filling scheme, and identify and handle outliers and missing values ​​in each component data; According to the mean and standard deviation of the current pretreatment model, the component data are standardized to make the data evenly distributed, and the key features affecting the sewage treatment process are extracted according to the water quality indicators and equipment operating parameters to obtain the standardization and feature extraction results; According to the results of standardization and feature extraction, the parameters of the preprocessing model are updated, and the standardization, feature extraction and parameter updating process are repeated until the preset number of iterations is reached, and all the preprocessed component data are merged to obtain the processed data set.

3. The cloud computing-based sewage treatment data optimization method and system according to claim 2, characterized in that: The calculation formula for the symmetric percentage error is: ; in, ; S represents the symmetric percentage error; γ represents the overall scaling factor; n represents the total number of samples; i represents the index variable; λ represents the time decay rate parameter; Indicates the maximum timestamp among all samples; represents the timestamp of the i-th sample; α represents the adjustment coefficient; represents the prediction error of the i-th sample; , represents the predefined error threshold; b represents the offset; Represents the predicted value of the i-th sample; represents the actual value of the i-th sample; represents a positive number; Indicates a condition.

4. The cloud computing-based sewage treatment data optimization method according to claim 3 is characterized in that: According to the analysis results, a sewage treatment data optimization model is constructed, and water quality standards, equipment operation efficiency, and energy consumption cost parameters are input into the sewage treatment data optimization model to generate an optimization plan, including: According to the analysis results, including water quality data, equipment operation data, and energy consumption data, the objective function of the sewage treatment data optimization model and the decision variables of the sewage treatment data optimization model are determined to construct the constraint conditions of the sewage treatment data optimization model and obtain the sewage treatment data optimization model framework; Input water quality standards, equipment operating efficiency and energy consumption cost parameters into the sewage treatment data optimization model framework, and set the initial values ​​of the decision variables of the sewage treatment data optimization model to obtain the final sewage treatment data optimization model; According to the final sewage treatment data optimization model, set the relevant parameters of ant colony optimization, including population size, number of iterations, and mutation factor, and regard each ant as a solution; According to the concentration of pheromones and heuristic information in the current environment, the position of the ants is updated, and the objective function value corresponding to each ant is calculated in real time; The position of each ant is continuously updated according to the concentration of pheromones and heuristic information in the current environment, and the objective function value of each ant at the new position is calculated in real time. When the predetermined number of iterations is reached, the final solution is determined according to the objective function value of each ant. Based on the final solution, a wastewater treatment optimization plan is generated, including the setting of water quality treatment parameters, expected water quality indicators, equipment operation efficiency and energy consumption cost.

5. The sewage treatment data optimization method based on cloud computing according to claim 4 is characterized in that: The calculation formula of the objective function value is: ； Where F represents the objective function value; , , represents the weight coefficient; N represents the total number of water quality indicators; i and j represent index variables; represents the weight coefficient of the i-th water quality index; represents the actual value of the i-th water quality index; represents the baseline value of the i-th water quality index; represents the deviation penalty coefficient of the i-th water quality index; Indicates the actual operating efficiency of the equipment; represents the maximum possible efficiency of the equipment; m represents the total number of energy consumption types; represents the weight coefficient of the j-th energy consumption; represents the cost of the j-th energy consumption; represents the cost increasing coefficient of the j-th energy consumption; represents the benchmark cost of the j-th energy consumption; An exponent representing the efficiency penalty.

6. The sewage treatment data optimization method based on cloud computing according to claim 5 is characterized in that: According to the collected effect data, the sewage treatment data optimization model is evaluated to obtain the evaluation results, and the sewage treatment data optimization model is adjusted and optimized according to the evaluation results, including: Using effect data and evaluation indicators, the effect of the sewage treatment data optimization model is quantitatively evaluated. Calculate the actual value of each evaluation indicator, and compare the actual value of each evaluation indicator with the preset benchmark value to obtain the evaluation result, including the actual value of each evaluation indicator and the comparison result with the preset benchmark value; Based on the evaluation results, formulate an adjustment and optimization plan for the sewage treatment data optimization model, and adjust and optimize the sewage treatment data optimization model according to the adjustment and optimization plan.

7. A sewage treatment data optimization system based on cloud computing, the system implementing the method according to any one of claims 1 to 6, characterized in that: include: The acquisition module is used to receive real-time key information collected by the monitoring equipment, including water quality indicators, equipment operating parameters, and energy consumption data, and pre-process the key information to obtain a processed data set; The analysis module is used to build a cloud database based on the processed data set and use the cloud database to perform data analysis, including trend prediction and anomaly detection, to obtain analysis results; A construction module is used to construct a sewage treatment data optimization model based on the analysis results, and input water quality standards, equipment operation efficiency, and energy consumption cost parameters into the sewage treatment data optimization model to generate an optimization plan; The optimization module is used to integrate the optimization scheme into the sewage treatment process and collect the water quality and energy consumption effect data after the optimization scheme is implemented in real time; The processing module is used to evaluate the effect of the sewage treatment data optimization model based on the collected effect data, obtain the evaluation results, and adjust and optimize the sewage treatment data optimization model based on the evaluation results.

8. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images