Demand-oriented intelligent sewage biological treatment process evaluation method and system

Through the intelligent sewage biological treatment process evaluation method that is oriented towards demand, machine learning is used to build an ecological risk prediction model, and the one-sided problem of evaluation results in the existing technology is solved, and multi-objective assessment of sewage biological treatment process and ecological risk reduction are achieved. The comprehensive evaluation of efficiency has selected the most effective process, which has enhanced the sewage treatment effect and ecological risk control.

CN117333030BActive Publication Date: 2024-11-01NANJING UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311169952.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-12
Publication Date
2024-11-01
Estimated Expiration
2043-09-12

AI Technical Summary

Technical Problem

The prediction and evaluation of existing sewage biological treatment processes is mainly oriented towards a single goal, and ignores the nonlinear coupling characteristics of multiple parameters in sewage biological treatment process, resulting in one-sided evaluation results, and it is impossible to comprehensively evaluate the ecological risks under different sewage biological treatment processes.

Method used

Construct a demand-oriented intelligent sewage biological treatment process evaluation method, collect and pretreat the basic water quality and ecological risk data of sewage treatment plants, build the optimal ecological risk prediction model for different sewage biological treatment processes based on machine learning, and evaluate different processes eco-risk reduction efficiency and select the most effective process.

Benefits of technology

A comprehensive assessment of multi-target attributes in the sewage treatment process has been achieved, more accurate and comprehensive ecological risk assessment results have been provided, helping to select the most effective biological treatment process, and enhancing the sewage treatment effect and ecological risk control capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure FDA0005040051650000011
    Figure FDA0005040051650000011
  • Figure FDA0005040051650000021
    Figure FDA0005040051650000021
  • Figure GDA0005040051660000021
    Figure GDA0005040051660000021
Patent Text Reader

Abstract

The present invention discloses a demand-oriented intelligent sewage biological treatment process evaluation method and system, the method comprising: constructing a basic water quality, ecological risk, and process information data set of the biological treatment process of a sewage treatment plant and merging them for pre-processing; determining an input water quality feature set and an ecological risk target; constructing and screening the optimal ecological risk prediction model for different sewage biological treatment processes based on machine learning; predicting the ecological risks of different biological treatment processes according to different sewage treatment scenarios, evaluating the ecological risk reduction efficiency of different sewage biological treatment processes, and selecting the most effective biological treatment process. The present invention takes into account the multi-objective attributes of the sewage treatment process, constructs an ecological risk prediction model for different sewage biological treatment processes, can comprehensively evaluate the ecological risks under different sewage biological treatment processes, and can select the most effective biological treatment process based on the evaluation results to enhance the sewage treatment effect.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to sewage treatment technology, and in particular to a demand-oriented intelligent sewage biological treatment process evaluation method and system. Background Art

[0002] As the national "dual carbon" and beautiful China strategies continue to advance, the field of water pollution control faces new and huge challenges. The selection of new sewage biological treatment processes faces multiple orientations such as pollution control, energy self-sufficiency and carbon neutrality, and effluent risk control. Among them, how to adapt to local conditions according to different water quality characteristics and evaluate and select sewage treatment processes for multiple purification targets is of great significance.

[0003] A Chinese patent application numbered 201910861998.7 discloses a prediction model for microbial dissolved organic nitrogen in sewage and its application. The prediction model is an ASM-mDON prediction model established using the operating parameters of the activated sludge treatment section of a sewage plant, the component concentrations in the influent, kinetic parameters, and stoichiometric coefficients. The application is specifically as follows: (1) establishing an ASM-mDON prediction model; (2) determining sewage components and ASM-mDON prediction model parameters; (3) predicting the concentration of microbial dissolved organic nitrogen in sewage. The ASM-mDON prediction model proposed in the present invention is suitable for the simulation of a steady-state sewage biological treatment system, and can solve the problem that microbial dissolved organic nitrogen is difficult to quantify in the actual sewage treatment process. The prediction method has the advantages of rapid prediction, accurate quantification, and simple operation, and is of practical significance for solving sewage quality problems.

[0004] The Chinese patent application number 202111627228.X discloses a method for evaluating the bioavailability of organic nitrogen in sewage based on machine learning, including collecting information on organic nitrogen molecular components and bioavailability data in sewage samples; establishing a prediction model for the bioavailability of organic nitrogen in sewage based on machine learning; determining the organic nitrogen molecular components in sewage samples from target sewage treatment plants; and predicting the bioavailability of organic nitrogen in sewage from target sewage treatment plants. This method has fast detection speed, high accuracy, and simple operation, and can be widely used in the evaluation of the bioavailability of organic nitrogen in sewage.

[0005] However, the predictive evaluation of the above-mentioned existing biological treatment processes is mainly guided by a single goal, ignoring the nonlinear coupling characteristics of multiple parameters in the sewage biological treatment process, making the evaluation results one-sided. Summary of the invention

[0006] Purpose of the invention: The present invention aims to provide a demand-oriented intelligent sewage biological treatment process evaluation method and system which takes into account the multi-objective attributes of the sewage treatment process, can comprehensively evaluate the ecological risks under different sewage biological treatment processes, and can select the most effective biological treatment process based on the evaluation results.

[0007] Technical solution: The demand-oriented intelligent sewage biological treatment process evaluation method described in the present invention comprises the following steps:

[0008] (1) Construct a data set of basic water quality, ecological risk, and process information of the biological treatment process of a sewage treatment plant and merge them for data preprocessing;

[0009] (2) Determine the input water quality characteristic set and ecological risk target based on demand orientation;

[0010] (3) Construct and screen the optimal ecological risk prediction model for different sewage biological treatment processes based on machine learning;

[0011] (4) According to different sewage treatment scenarios, the ecological risks of different biological treatment processes are predicted based on the optimal ecological risk prediction model of different sewage biological treatment processes, the ecological risk reduction efficiency of different sewage biological treatment processes is evaluated, and the most effective biological treatment process is selected.

[0012] Preferably, in step (1), the basic water quality characteristics include biochemical oxygen demand (BOD), chemical oxygen demand (COD), total phosphorus, total nitrogen, ammonia nitrogen, dissolved oxygen (DO), temperature, pH value, and fecal coliform group;

[0013] The ecological risk data characteristics include biological toxicity and micropollutant content;

[0014] The sewage biological treatment process includes an anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process, anoxic-aerobic or anaerobic-aerobic AO process, an oxidation ditch OD process, a membrane bioreactor MBR process, and a circulating activated sludge CAST process.

[0015] Preferably, in step (1), the data preprocessing includes abnormal data processing, default data processing, outlier data processing, repeatability and consistency detection, and data normalization and standardization of the basic water quality and ecological risk data of the inlet and outlet water of the sewage treatment plant.

[0016] Preferably, the construction of the optimal ecological risk prediction model for different sewage biological treatment processes in step (3) includes: selecting different machine learning algorithms, establishing ecological risk prediction models for different sewage biological treatment processes based on the input water quality feature set, and evaluating the performance of the ecological risk prediction model through cross-validation experiments, and selecting the optimal machine learning algorithm and the corresponding ecological risk prediction model.

[0017] Preferably, the model screening in step (3) is performed by comparing whether the prediction results of the optimal ecological risk prediction model meet the ecological risk target.

[0018] Preferably, the following four indicators are used as the basis for comparing and evaluating whether the prediction results of the optimal ecological risk prediction model meet the ecological risk target. The calculation formula of each indicator is:

[0019]

[0020] Among them, TP represents the amount of true positive data, TN represents the amount of true negative data, FP represents the amount of false positive data, and FN represents the amount of false negative data; SE represents sensitivity, SP represents specificity, and SE and SP respectively evaluate the prediction accuracy of the model for positive and negative examples; ACC represents the correct rate, and MCC represents the Matthews correlation coefficient, which are used to comprehensively evaluate the overall prediction performance of the model.

[0021] Preferably, step (4) includes: inputting water quality characteristics according to different sewage treatment scenarios, predicting ecological risks respectively based on the optimal ecological risk prediction models of different sewage biological treatment processes, and comparing them with the ecological risk targets, evaluating the ecological risk reduction efficiency of different sewage biological treatment processes, and selecting the most effective biological treatment process therefrom.

[0022] The demand-oriented intelligent sewage biological treatment process evaluation system described in the present invention includes:

[0023] Data collection and preprocessing module, used to build basic water quality, ecological risk, and process information data sets of the biological treatment process of the sewage treatment plant and merge them for data preprocessing;

[0024] Prediction model building and training module, used to determine the input water quality feature set and ecological risk target according to demand orientation; to build and screen the optimal ecological risk prediction model of different sewage biological treatment processes based on machine learning;

[0025] The sewage biological treatment process assessment and selection module is used to predict the ecological risks of different biological treatment processes based on the optimal ecological risk prediction models of different sewage biological treatment processes according to different sewage treatment scenarios, evaluate the ecological risk reduction efficiency of different sewage biological treatment processes, and select the most effective biological treatment process.

[0026] An electronic device described in the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is loaded into the processor, the above-mentioned demand-oriented intelligent sewage biological treatment process evaluation method is implemented.

[0027] The present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the above-mentioned demand-oriented intelligent sewage biological treatment process evaluation method.

[0028] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: 1. It takes into account multiple objective attributes such as basic water quality, ecological risks, and biological treatment process information data in the sewage treatment process, constructs an ecological risk prediction model for different sewage biological treatment processes under different sewage treatment scenarios, and evaluates the ecological risks under different sewage biological treatment processes, with more comprehensive evaluation results; 2. Based on the evaluation results, it can provide sewage treatment plants with effective biological treatment process selection plans, enhance sewage treatment effects, and at the same time provide supervision and early warning for the sewage treatment ecological risk management of sewage treatment plants. DETAILED DESCRIPTION

[0029] The demand-oriented intelligent sewage biological treatment process evaluation method described in the present invention comprises the following steps:

[0030] (1) Construct a data set of basic water quality, ecological risk, and process information of the biological treatment process of a sewage treatment plant and merge them for data preprocessing.

[0031] The basic water quality and ecological risk data sets of the inlet and outlet water of sewage treatment plants with different biological treatment processes were collected respectively. The basic water quality characteristics of the inlet and outlet water included biochemical oxygen demand (BOD), chemical oxygen demand (COD), total phosphorus, total nitrogen, ammonia nitrogen, dissolved oxygen (DO), temperature, pH value, fecal coliform group, etc.

[0032] Ecological risk data characteristics include biological toxicity, micropollutant content, etc.;

[0033] Biological sewage treatment processes include anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process, anoxic-aerobic or anaerobic-aerobic AO process, oxidation ditch OD process, membrane bioreactor MBR process, circulating activated sludge CAST process, etc.

[0034] The basic water quality and ecological risk data of the inlet and outlet water of the sewage treatment plant are preprocessed; the data preprocessing methods include abnormal data processing, default data processing, outlier data processing, repeatability and consistency detection, and data normalization and standardization.

[0035] (2) Determine the input water quality characteristic set and ecological risk targets based on demand.

[0036] (3) Construct the optimal ecological risk prediction model for different sewage biological treatment processes based on machine learning.

[0037] Different machine learning algorithms are selected, and based on the input water quality feature set, ecological risk prediction models are established for different sewage biological treatment processes. Through cross-validation experiments, the performance of the ecological risk prediction model is evaluated, and the optimal machine learning algorithm and the corresponding ecological risk prediction model are selected.

[0038] The input of the constructed optimal ecological risk prediction model is the determined set of input water quality characteristics and sewage biological treatment process characteristics. The output of the model is the ecological risk prediction result, and it is judged whether the ecological risk prediction result reaches the ecological risk target in step (2).

[0039] The following four indicators are used as the basis for comparing and evaluating whether the prediction results of the optimal ecological risk prediction model meet the ecological risk target. The calculation formula of each indicator is:

[0040]

[0041] Among them, TP represents the amount of true positive data, TN represents the amount of true negative data, FP represents the amount of false positive data, and FN represents the amount of false negative data; SE represents sensitivity, SP represents specificity, and SE and SP respectively evaluate the prediction accuracy of the model for positive and negative examples; ACC represents the correct rate, and MCC represents the Matthews correlation coefficient, which are used to comprehensively evaluate the overall prediction performance of the model.

[0042] (4) According to the input water quality characteristics of different sewage treatment scenarios, the ecological risks are predicted separately based on the optimal ecological risk prediction models of different sewage biological treatment processes, and compared with the ecological risk targets, the ecological risk reduction efficiency of different sewage biological treatment processes is evaluated, and the most effective biological treatment process is selected.

[0043] In order to further illustrate the method of the present invention, the sewage biological treatment process data of sewage treatment plants with different sewage biological treatment processes across the country are collected, and the sewage biological treatment process is evaluated and selected. The steps are as follows:

[0044] (1) Collect information on the biological treatment process of sewage from 180 sewage treatment plants in Jiangsu, Guangdong, Shaanxi, Sichuan, Beijing, Shanghai and other provinces and cities, and construct a data set of basic water quality, ecological risk, and process information of the biological treatment process of sewage treatment plants and merge them for data preprocessing.

[0045] Among them, basic water quality information includes total phosphorus, total nitrogen, ammonia nitrogen, total organic carbon TOC, dissolved oxygen DO and temperature; sewage biological treatment processes include anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process, anoxic-aerobic or anaerobic-aerobic AO process, oxidation ditch OD process and membrane bioreactor MBR process; the ecological risk characteristic is biological toxicity.

[0046] The collected sewage biological treatment process data of the sewage treatment plant are preprocessed. The data preprocessing methods include abnormal data processing, default data processing, outlier data processing, repeatability and consistency detection, and data normalization and standardization.

[0047] (2) Determine the input water quality characteristic set including total phosphorus, total nitrogen, ammonia nitrogen, total organic carbon TOC, dissolved oxygen DO and temperature, and the ecological risk target is a 70% reduction in biological toxicity.

[0048] (3) Construct the optimal ecological risk prediction model for different sewage biological treatment processes based on machine learning.

[0049] Two machine learning algorithms, support vector machine and random forest, were selected. Based on the input water quality feature set, ecological risk prediction models were established for different sewage biological treatment processes. The performance of the ecological risk prediction models was evaluated through cross-validation experiments, and the optimal machine learning algorithm and the corresponding ecological risk prediction model were selected.

[0050] In this embodiment, the random forest RF algorithm is used to construct the optimal ecological risk prediction model of the anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process, the anoxic-aerobic or anaerobic-aerobic AO process, and the support vector machine SVM algorithm is used to construct the optimal ecological risk prediction model of the oxidation ditch OD process and the membrane bioreactor MBR process.

[0051] Using the four indicators of SE, SP, ACC and MCC as the basis for comparing and evaluating whether the prediction results of the optimal ecological risk prediction model have reached the ecological risk target, the prediction evaluation index results of the optimal ecological risk prediction model corresponding to different sewage biological treatment processes are shown in Table 1. As can be seen from the table, the prediction evaluation indexes of SE, SP and ACC of the optimal ecological risk prediction model corresponding to different sewage biological treatment processes are all higher than the ecological risk target (biological toxicity reduction of 70%), so they can all reach the ecological risk target.

[0052]

[0053] (4) According to the input water quality characteristics of different sewage treatment scenarios, the ecological risks are predicted separately based on the optimal ecological risk prediction models of different sewage biological treatment processes, and compared with the ecological risk targets, the ecological risk reduction efficiency of different sewage biological treatment processes is evaluated, and the most effective biological treatment process is selected.

[0054] (4.1) Based on the optimal ecological risk prediction model of different sewage biological treatment processes, predict and evaluate whether the ecological risk reduction efficiency of different sewage biological treatment processes meets the standards.

[0055] Specifically, sewage treatment scenario one: input water quality characteristics include total phosphorus of 0.5 mg / L, total nitrogen of 7 mg / L, ammonia nitrogen of 3 mg / L, total organic carbon of 20 mg / L, and dissolved oxygen of 1.5 mg / L; the ecological risk target is a 70% reduction in biological toxicity.

[0056] The ecological risk prediction results of the optimal ecological risk prediction model for different sewage biological treatment processes are as follows:

[0057] The predicted biological toxicity reduction effect of the anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process is 71.39%, the predicted biological toxicity reduction effect of the anoxic-aerobic or anaerobic-aerobic AO process is 70.86%, the predicted biological toxicity reduction effect of the oxidation ditch OD process is 70.61%, and the predicted biological toxicity reduction effect of the membrane bioreactor MBR process is 68.14%. Therefore, the sewage biological treatment processes that meet the standards are the anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process, the anoxic-aerobic or anaerobic-aerobic AO process and the oxidation ditch process OD.

[0058] (4.2) According to the water quality characteristics of different sewage treatment scenarios, the ecological risks under different sewage biological treatment processes are predicted respectively, and the more suitable sewage biological treatment process is selected by comparing the ecological risk results.

[0059] Sewage treatment scenario two: Input water quality characteristics include total phosphorus of 1 mg / L, total nitrogen of 10 mg / L, ammonia nitrogen of 0.5 mg / L, total organic carbon of 15 mg / L, and dissolved oxygen of 1.2 mg / L; the ecological risk target is a 70% reduction in biological toxicity.

[0060] The ecological risk prediction results of the optimal ecological risk prediction model for different sewage biological treatment processes are as follows:

[0061] The predicted biological toxicity reduction effect of the anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process is 72.85%, the predicted biological toxicity reduction effect of the anoxic-aerobic or anaerobic-aerobic AO process is 71.54%, the predicted biological toxicity reduction effect of the oxidation ditch OD process is 69.13%, and the predicted biological toxicity reduction effect of the membrane bioreactor MBR process is 67.97%. It can be seen that the sewage biological treatment processes that meet the standards are the anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process and the anoxic-aerobic or anaerobic-aerobic AO process; the most effective biological treatment process is the anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process.

[0062] The demand-oriented intelligent sewage biological treatment process evaluation system described in the present invention includes:

[0063] Data collection and preprocessing module, used to build basic water quality, ecological risk, and process information data sets of the biological treatment process of the sewage treatment plant and merge them for data preprocessing;

[0064] Prediction model building and training module, used to determine the input water quality feature set and ecological risk target according to demand orientation; to build and screen the optimal ecological risk prediction model of different sewage biological treatment processes based on machine learning;

[0065] The sewage biological treatment process assessment and selection module is used to predict the ecological risks of different biological treatment processes based on the optimal ecological risk prediction models of different sewage biological treatment processes according to different sewage treatment scenarios, evaluate the ecological risk reduction efficiency of different sewage biological treatment processes, and select the most effective biological treatment process.

[0066] An electronic device described in the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is loaded into the processor, the above-mentioned demand-oriented intelligent sewage biological treatment process evaluation method is implemented.

[0067] The present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the above-mentioned demand-oriented intelligent sewage biological treatment process evaluation method.

Claims

1. A demand-oriented intelligent sewage biological treatment process evaluation method, characterized in that: The following steps are involved: (1) Construct a data set of basic water quality, ecological risk, and process information of the biological treatment process of a sewage treatment plant and merge them for data preprocessing; (2) Determine the input water quality characteristic set and ecological risk target based on demand orientation; (3) Construct and screen the optimal ecological risk prediction model for different sewage biological treatment processes based on machine learning; The step (3) of constructing the optimal ecological risk prediction model for different sewage biological treatment processes includes: selecting different machine learning algorithms, establishing ecological risk prediction models for different sewage biological treatment processes based on the input water quality feature set, and evaluating the performance of the ecological risk prediction model through cross-validation experiments, and selecting the optimal machine learning algorithm and the corresponding ecological risk prediction model; The model screening in step (3) is to screen by comparing whether the prediction results of the optimal ecological risk prediction model meet the ecological risk target; The following four indicators are used as the basis for comparing and evaluating whether the prediction results of the optimal ecological risk prediction model meet the ecological risk target. The calculation formula of each indicator is: Among them, TP represents the amount of true positive data, TN represents the amount of true negative data, FP represents the amount of false positive data, and FN represents the amount of false negative data; SE represents sensitivity, SP represents specificity, and SE and SP respectively evaluate the prediction accuracy of the model for positive and negative examples; ACC represents the correct rate, and MCC represents the Matthews correlation coefficient, which is used to comprehensively evaluate the overall prediction performance of the model; (4) According to different sewage treatment scenarios, the ecological risks of different biological treatment processes are predicted based on the optimal ecological risk prediction model of different sewage biological treatment processes, the ecological risk reduction efficiency of different sewage biological treatment processes is evaluated, and the most effective biological treatment process is selected.

2. The demand-oriented intelligent sewage biological treatment process evaluation method according to claim 1 is characterized in that: In step (1), the basic water quality characteristics include biochemical oxygen demand (BOD), chemical oxygen demand (COD), total phosphorus, total nitrogen, ammonia nitrogen, dissolved oxygen (DO), temperature, pH value, and fecal coliform group; The ecological risk data characteristics include biological toxicity and micropollutant content; The sewage biological treatment process includes an anaerobic-anoxic-aerobic or anoxic-anaerobic-aerobic AAO process, anoxic-aerobic or anaerobic-aerobic AO process, an oxidation ditch OD process, a membrane bioreactor MBR process, and a circulating activated sludge CAST process.

3. The demand-oriented intelligent sewage biological treatment process evaluation method according to claim 1 is characterized in that: In step (1), the data preprocessing includes abnormal data processing, default data processing, outlier data processing, repeatability and consistency detection, and data normalization and standardization of the basic water quality and ecological risk data of the inlet and outlet water of the sewage treatment plant.

4. The demand-oriented intelligent sewage biological treatment process evaluation method according to claim 1 is characterized in that: Step (4) includes: inputting water quality characteristics according to different sewage treatment scenarios, predicting ecological risks respectively based on the optimal ecological risk prediction models of different sewage biological treatment processes, and comparing them with the ecological risk targets, evaluating the ecological risk reduction efficiency of different sewage biological treatment processes, and selecting the most effective biological treatment process.

5. A demand-oriented intelligent sewage biological treatment process evaluation system, characterized in that: include: Data collection and preprocessing module, used to build basic water quality, ecological risk, and process information data sets of the biological treatment process of the sewage treatment plant and merge them for data preprocessing; Prediction model building and training module, used to determine the ecological risk target of the input water quality characteristic set according to demand orientation; Based on machine learning, the optimal ecological risk prediction model of different sewage biological treatment processes is constructed and screened; The construction of the optimal ecological risk prediction model for different sewage biological treatment processes includes: selecting different machine learning algorithms, establishing ecological risk prediction models for different sewage biological treatment processes based on the input water quality feature set, and evaluating the performance of the ecological risk prediction model through cross-validation experiments, and selecting the optimal machine learning algorithm and the corresponding ecological risk prediction model; Model screening is to screen by comparing the prediction results of the optimal ecological risk prediction model to see whether they meet the ecological risk target; The following four indicators are used as the basis for comparing and evaluating whether the prediction results of the optimal ecological risk prediction model meet the ecological risk target. The calculation formula of each indicator is: Among them, TP represents the amount of true positive data, TN represents the amount of true negative data, FP represents the amount of false positive data, and FN represents the amount of false negative data; SE represents sensitivity, SP represents specificity, and SE and SP respectively evaluate the prediction accuracy of the model for positive and negative examples; ACC represents the correct rate, and MCC represents the Matthews correlation coefficient, which is used to comprehensively evaluate the overall prediction performance of the model; The sewage biological treatment process assessment and selection module is used to predict the ecological risks of different biological treatment processes based on the optimal ecological risk prediction models of different sewage biological treatment processes according to different sewage treatment scenarios, evaluate the ecological risk reduction efficiency of different sewage biological treatment processes, and select the most effective biological treatment process.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into the processor, it implements the demand-oriented intelligent sewage biological treatment process evaluation method according to any one of claims 1-4.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements the demand-oriented intelligent sewage biological treatment process evaluation method according to any one of claims 1-4.