Electrochemical reaction-based wastewater treatment optimization method

Abstract

The invention relates to a wastewater treatment optimization method based on electrochemical reaction. Comprising the following steps: monitoring the concentration of pollutants in wastewater in real time by configuring various sensors including a heavy metal sensor, a chemical oxygen demand sensor and an ammonia nitrogen sensor; sensor data is transmitted to the central control unit of the processing system through the wireless communication module; constructing a reaction kinetic model according to the type and concentration of pollutants in the wastewater and the properties of electrochemical reaction; an optimization target is set, the pollutant removal rate is calculated to be maximized, and energy consumption and consumables including abrasion of electrodes and use of chemical reagents are reduced; dynamically adjusting parameters of electrochemical reaction by adopting a self-adaptive optimization algorithm; the parameters comprise current density, voltage of an electrolytic cell, reaction time and temperature; in the wastewater treatment process, the sensor feeds back the pollutant concentration of the wastewater and the voltage and current density information in the electrolytic bath in real time; and the data is transmitted to a central unit of the control system through the data transmission module.

Description

Technical Field

[0001] The invention relates to a wastewater treatment optimization method, in particular to a wastewater treatment optimization method based on electrochemical reaction. Background Art

[0002] The current optimization methods for wastewater treatment, especially the traditional model-based optimization methods, have some significant deficiencies and drawbacks. Taking the Chinese invention patent publication number CN118747419A as an example, this method manages and controls the industrial waste gas and wastewater treatment system through model optimization. Although it achieves some optimization, its limitations and defects cannot be ignored.

[0003] First, the existing wastewater treatment system usually relies on a preset optimization model. This model-based optimization method relies on a large number of assumptions and known conditions, and ignores the dynamic changes of the actual environment during the wastewater treatment process. The accuracy of the model can usually only be guaranteed under ideal conditions. In actual applications, the composition of wastewater and the concentration of pollutants are affected by many factors, such as seasonal changes, fluctuations in pollution sources, and differences in influent water quality, which all lead to insufficient adaptability of the model. Therefore, the existing model optimization method is often difficult to achieve the expected treatment effect in dealing with complex wastewater treatment processes, especially when the concentration of pollutants changes greatly. Secondly, in the existing method, although the step of introducing the wastewater into the regulating tank after pretreatment can provide a certain temporary regulation function, this method still cannot effectively deal with the complex characteristics of diversified pollutants in wastewater. The removal effect of organic pollutants in the pretreatment process is limited by the singleness and limitations of the treatment method, especially when there are difficult-to-degrade organic matter or heavy metals in the wastewater, the existing microbial degradation method is difficult to completely eliminate these pollutants, resulting in low removal rate and difficulty in meeting higher standards of treatment requirements. Furthermore, although microbial degradation is one of the common wastewater treatment methods, it has certain limitations. Microbial degradation depends on specific microbial populations and conditions, such as temperature, pH value, etc. These factors may fluctuate greatly during the actual wastewater treatment process, resulting in reduced microbial activity or inability to fully function, thereby affecting the treatment effect.

[0004] In addition, the existing methods rely on simulated degradation data sets and measured degradation data sets to control and optimize the wastewater treatment process. However, in the process of generating and using these data, there may be inaccurate data collection or errors in data processing. For each link in wastewater treatment, the accuracy and real-time nature of the data are crucial. However, due to the interference of various external factors, the existing processing and analysis modules may not be able to fully obtain accurate degradation data, which will directly affect the effect of the optimization algorithm, resulting in deviation or instability in the treatment results. In the further deep treatment stage of sedimentation separation, although some suspended solids and heavy metals can be effectively removed, the method still faces some challenges. First, the efficiency of the sedimentation separation process is greatly affected by various factors such as water quality and temperature. Especially in the case of high concentrations of pollutants, the sedimentation efficiency may be greatly reduced; secondly, sedimentation separation usually requires a long processing time, which means that the overall wastewater treatment cycle is long, especially in large-scale industrial wastewater treatment situations. Slow processing speed may lead to excessive resources and time, thereby increasing costs. Another major problem is that although the existing evaluation and control module can generate instructions to control the wastewater treatment process based on the optimization algorithm model, its effect usually depends on the complexity and accuracy of the algorithm model, and the existing optimization algorithm often faces the problems of insufficient computing resources and poor real-time processing. In actual operation, the concentration and composition of pollutants in wastewater often change dynamically, and existing algorithms are unable to respond to these changes in real time. As a result, in actual operation, the wastewater treatment process is not flexible enough, and cannot be effectively adjusted in the shortest time, and cannot respond to sudden changes in the environment in a timely manner. Summary of the invention

[0005] The purpose of the present invention is to provide a wastewater treatment optimization method based on electrochemical reaction, so as to solve some of the drawbacks and shortcomings pointed out in the background technology.

[0006] The present invention solves the above-mentioned technical problems by adopting the following technical solution, which includes the following steps:

[0007] S1. Wastewater composition analysis and electrochemical reaction model construction:

[0008] S1.1. By configuring a variety of sensors including heavy metal sensors, chemical oxygen demand sensors, and ammonia nitrogen sensors, the concentration of pollutants in wastewater is monitored in real time; the sensor data is transmitted to the central control unit of the treatment system through a wireless communication module;

[0009] S1.2. Construct a reaction kinetic model based on the type and concentration of pollutants in the wastewater and the properties of the electrochemical reaction;

[0010] S2. Adaptive optimization algorithm design:

[0011] S2.1. Set optimization goals, calculate the maximum pollutant removal rate, and reduce energy consumption and consumables including electrode wear and chemical reagent use;

[0012] S2.2, using an adaptive optimization algorithm to dynamically adjust the parameters of the electrochemical reaction; the parameters include current density, voltage of the electrolytic cell, reaction time, and temperature;

[0013] S3. Real-time data feedback and system self-adjustment mechanism:

[0014] S3.1. During the wastewater treatment process, the sensor provides real-time feedback on the pollutant concentration of the wastewater, the voltage in the electrolytic cell, and the current density information; the data is transmitted to the central unit of the control system through the data transmission module;

[0015] S3.2, by analyzing the data, the current treatment effect including the removal rate of pollutants and the treatment speed is evaluated in real time, and the results are transmitted to the next treatment link;

[0016] S3.3. Based on the real-time monitoring data, the parameters of the electrochemical reaction including current and voltage will be dynamically adjusted according to the preset objective function and optimization algorithm;

[0017] S4. Intelligent prediction and advance adjustment strategy:

[0018] S4.1. Based on historical data and real-time monitoring data, a prediction model for wastewater pollutant concentration is established using machine learning methods; based on past wastewater composition and treatment effects, the changing trend of wastewater pollutant concentration in the future is predicted;

[0019] S4.2. Make adjustments based on the prediction results; if it is predicted that the pollutant concentration in the wastewater will increase, increase the current or extend the reaction time; if it is predicted that the pollutant concentration will decrease, reduce the reaction intensity.

[0020] Furthermore, the real-time data feedback and system self-adjustment mechanism includes:

[0021] Based on the real-time collected pollutant concentration data, the current treatment effect is evaluated by dynamically calculating the pollutant removal rate of the wastewater; the pollutant removal rate represents the reduction ratio of the pollutant concentration in the wastewater compared to before treatment; and the pollutant removal rate is calculated using the formula:

[0022]

[0023] Among them: F removal (t) is the pollutant removal rate at time t, which indicates the removal ratio of the pollutant concentration after wastewater treatment relative to the initial concentration; C t is the pollutant concentration at time t; C0 is the initial pollutant concentration at the beginning of wastewater treatment;

[0024] The real-time treatment speed assessment is the rate of change of pollutant concentration in wastewater over time. The progress of wastewater treatment is assessed by the relationship between pollutant concentration and time. The real-time treatment speed assessment formula is used:

[0025]

[0026] Where: P rate (t) is the wastewater treatment rate at time t, which represents the rate of change of pollutant concentration over time; It is the time derivative of the pollutant concentration in wastewater, reflecting the change in pollutant concentration per unit time.

[0027] Furthermore, the real-time data feedback and system self-adjustment mechanism includes:

[0028] A multi-objective optimization function is introduced to comprehensively calculate multiple factors such as pollutant removal rate, energy efficiency, and equipment wear; the optimization objective function dynamically adjusts the weight according to real-time data changes, and optimizes the treatment parameters in real time according to the concentration of pollutants in the wastewater and the treatment environment; the comprehensive optimization objective function is:

[0029]

[0030] in:

[0031] F opt (t) is the comprehensive optimization objective function at time t, which represents the comprehensive processing objective of each objective in the optimization process; w1, w2, w3 are the weight coefficients of each objective; It indicates the pollutant removal rate, reflecting the degree of reduction of pollutant concentration in wastewater; is the energy efficiency ratio, which indicates the ratio of the electric power to the current density of the processing speed; σ is the electrode wear coefficient, which reflects the degree of equipment wear; ρ(t) is the current density, which indicates the current intensity in the electrolytic reaction; t is the reaction time, which determines the duration of the electrochemical reaction; is the rate of change of pollutant concentration, indicating the processing speed; γ is the influence coefficient of processing speed on the optimization objective function, regulating the contribution of speed to the objective function.

[0032] Furthermore, the real-time data feedback and system self-adjustment mechanism includes:

[0033] Through real-time data feedback and the results of the optimization algorithm, the wastewater treatment system dynamically adjusts the current density, voltage, and reaction time parameters. The dynamic adjustment formula for current density, voltage, and reaction time is:

[0034]

[0035] in:

[0036] I(t) is the current density at time t, which determines the intensity of the electrolytic reaction; V(t) is the voltage at time t, which determines the voltage intensity of the reaction; T r is the reaction time, indicating the duration of the electrochemical reaction; I0, V0, T0 are the initial current density, voltage and reaction time, respectively, representing the initial conditions of wastewater treatment; β1, β2 are the adjustment coefficients of current density and voltage, reflecting the influence of pollutant concentration on current and voltage; γ is the adjustment coefficient of reaction time, characterizing the relationship between reaction time and pollutant concentration.

[0037] Furthermore, the intelligent prediction and advance adjustment strategy method includes:

[0038] The data is standardized so that the distribution is within the same scale range; the filtering algorithm is used to smooth the data, and then the rate of change of pollutant concentration over time is calculated to capture the dynamic trend of the data and provide time data input for the machine learning model; the data fusion and preprocessing calculation formula is:

[0039]

[0040] in:

[0041] X proc (t) is the pre-processed data at time t; X(t) is the original pollutant concentration, current density, voltage, and temperature parameters at time t; μ X and σ X is the mean and standard deviation of the parameter, used for standardization; α is the smoothing coefficient (0<α<1), which controls the strength of data denoising; X proc (t-1) is the smoothed data of the previous moment; ΔX(t)=X(t)X(tΔt) is the time series change, reflecting the time change trend of the pollutant concentration.

[0042] Furthermore, the intelligent prediction and advance adjustment strategy method includes:

[0043] Construct a machine learning model to learn the change pattern of pollutant concentration over time during wastewater treatment; use the sliding window method to obtain pollutant concentration, current density, and voltage information over a period of time to identify short-term and long-term trends; select a time series prediction model and train it in combination with feature data; define the loss function and optimize the model parameters; use the mean square error MSE as the objective function and add the L2 regularization term to prevent overfitting; and the model training and optimization formula is expressed as:

[0044]

[0045] in:

[0046] L modelis the loss function used for optimization, including error minimization and model complexity constraints; N is the number of training data samples; y i is the actual pollutant concentration of the i-th sample; is the pollutant concentration predicted by the model; λ is the regularization coefficient; w j is the weight parameter in the model; M is the total number of parameters in the model.

[0047] Furthermore, the intelligent prediction and advance adjustment strategy method includes:

[0048] Using the trained machine learning model, the pollutant concentration change trend in the future is predicted based on the current pollutant concentration and current density environmental factors. The incremental learning method is used to optimize the model parameters using the gradient update method each time new data is input. The real-time prediction and model update formula is:

[0049]

[0050] in:

[0051] is the predicted future pollutant concentration; C(t) is the current pollutant concentration; X env (t) is the environmental characteristics, including temperature, current density, and voltage; θ is the parameter set of the current model; fC(t), X env (t), θ is the prediction function of the machine learning model; η is the learning rate, which controls the step size of parameter update; is the gradient of the loss function with respect to the model parameters, which is used to update the model.

[0052] The wastewater treatment optimization method of the present invention has significant beneficial effects, which are mainly reflected in the following aspects:

[0053] By dynamically adjusting key parameters such as current density, voltage, and reaction time in real time, it is possible to optimize in time according to changes in the concentration of pollutants in the wastewater, maximize the efficiency of the electrochemical reaction, and achieve efficient pollutant removal. This refined control not only improves the treatment effect, but also reduces unnecessary energy consumption. Using real-time data feedback and machine learning models, the present invention can achieve intelligent prediction and adaptive adjustment in the wastewater treatment process. The system can automatically adjust operating parameters according to changes in environmental factors such as wastewater pollutant concentration, current density, and voltage to ensure the best removal effect in each treatment cycle and reduce manual intervention.

[0054] The wastewater treatment process can be precisely controlled by calculating the pollutant removal rate and real-time processing speed, combined with the dynamic adjustment of electrochemical reaction parameters. Even in the case of large fluctuations in pollutant concentration, the system can effectively respond and make real-time adjustments through optimization algorithms to avoid over- or under-processing during the treatment process. Over-reaction and energy waste are avoided by optimizing parameters such as current density, voltage, and reaction time. The system adjusts the treatment intensity in real time according to changes in pollutant concentration to ensure the highest removal effect at the lowest energy consumption, thereby effectively reducing the operating cost of wastewater treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of the wastewater treatment optimization method based on electrochemical reaction of the present invention.

[0056] Figure 2 It is a flow chart of the real-time data feedback and system self-adjustment mechanism method of the present invention.

[0057] Figure 3 This is a flow chart of the intelligent prediction and advance adjustment strategy method of the present invention. DETAILED DESCRIPTION

[0058] The specific implementation modes of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0059] The core of the wastewater treatment optimization method based on electrochemical reaction lies in optimizing the wastewater treatment process by accurately monitoring the concentration of pollutants in the wastewater and applying the kinetic principles of electrochemical reactions. First of all, the composition analysis of wastewater and the construction of electrochemical reaction models are an indispensable step in the entire optimization process. At this stage, it is necessary to configure a variety of sensors, including heavy metal sensors, chemical oxygen demand sensors, ammonia nitrogen sensors, etc. The function of these sensors is to monitor the concentration of various pollutants in wastewater in real time. Heavy metal sensors can accurately detect the concentration of heavy metal ions in water, chemical oxygen demand sensors are used to measure the oxidation demand of organic matter in water bodies, and ammonia nitrogen sensors quantitatively monitor nitrogen pollution in wastewater. Through the joint use of these sensors, pollutant data in wastewater can be fully and real-time mastered.

[0060] These sensor data are transmitted to the central control unit of the wastewater treatment system through the wireless communication module, which is responsible for receiving, analyzing and further decision-making of the data. The application of the wireless communication module makes data transmission more flexible and efficient, ensuring that all kinds of data are transmitted to the control system in a short time, reducing the need for manual intervention and improving the automation of the system. After receiving the real-time monitoring data, the central control unit uses data analysis methods to determine the current pollution status of the wastewater, including the concentration and type of pollutants. Based on these data, the control system can establish a corresponding reaction model based on the kinetic principle of electrochemical reactions, thereby predicting the performance of electrochemical reactions in wastewater treatment.

[0061] The construction of the electrochemical reaction model mainly considers the interaction between pollutants in the wastewater and the electrode surface, including redox reactions, deposition reactions, and adsorption reactions. The rates of these reactions usually depend on multiple factors, such as current density, voltage, temperature, etc. By modeling these factors, the behavior of electrochemical reactions under different conditions can be simulated, and the most suitable treatment scheme can be derived from them. This reaction kinetic model not only needs to consider the concentration and type of pollutants in the wastewater, but also requires in-depth analysis of the reaction rates of various pollutants in electrochemical reactions and their influencing factors.

[0062] By dynamically adjusting the key parameters of the electrochemical reaction, the pollutant removal rate in wastewater treatment can be maximized, while reducing energy consumption, equipment wear and the use of chemical reagents. First of all, setting the optimization goal is the first step of the algorithm. This goal not only includes maximizing the pollutant removal rate, but also considering other important factors in the wastewater treatment process, such as energy efficiency, equipment life, and the amount of chemical reagents used. Pollutant removal rate is an important indicator for evaluating the effect of wastewater treatment, but in actual operation, resource conservation and environmental impact must also be considered comprehensively. Therefore, the optimization goal requires a multi-dimensional trade-off. In order to ensure the diversity and flexibility of the objective function, a weighted summation method can be used to comprehensively optimize factors such as pollutant removal rate, energy efficiency, equipment wear and chemical reagent consumption to obtain a multi-objective optimization function. The establishment of this function provides a theoretical basis for subsequent algorithm design.

[0063] Next, an adaptive optimization algorithm is used to dynamically adjust the key parameters in the electrochemical reaction. The adaptive optimization algorithm has the ability to automatically adjust system parameters according to real-time changes in the wastewater treatment process, especially when the pollutant concentration changes and the reaction environment is unstable, it can intelligently find the most suitable treatment parameters. The key parameters that need to be adjusted include current density, electrolytic cell voltage, reaction time and temperature. These parameters directly affect the efficiency of the electrochemical reaction and the purification effect of the wastewater. Among them, the current density and electrolytic cell voltage determine the intensity of the reaction, and the reaction time and temperature determine the continuity and speed of the reaction. Through the adaptive optimization algorithm, the system can adjust these parameters according to real-time monitoring data to avoid excessive or low treatment intensity during the reaction process, so as to achieve the best balance between energy efficiency and removal rate. Specifically, when the pollutant concentration is high, the system can automatically increase the current density or voltage to accelerate the reaction and improve the removal efficiency; when energy saving is required, the system will reduce energy consumption by reducing the current density or reaction temperature, while avoiding excessive wear and overreaction of the electrode due to high current or high temperature, thereby reducing the maintenance cost of the equipment and extending its service life. Adjustment of the reaction time can avoid excessive processing or energy waste due to too long a reaction time, or the reaction time is insufficient to achieve the desired purification effect.

[0064] During the entire wastewater treatment process, the system can obtain real-time information such as the pollutant concentration in the wastewater, the voltage and current density in the electrolytic cell through various sensors installed in the system, such as pollutant concentration sensors, voltage sensors and current density sensors. This information is transmitted to the central unit of the control system in a timely manner through the data transmission module. The function of the data transmission module is to transmit the data collected on site to the central processing unit in a reliable manner for further processing and analysis. After receiving this data, the central control unit analyzes it in real time through the data analysis algorithm to evaluate key performance indicators such as pollutant removal rate and processing speed in the current wastewater treatment process. The removal rate reflects the degree of purification of pollutants in the wastewater, while the processing speed reflects the working efficiency of the wastewater treatment system.

[0065] By analyzing these data, the control system can instantly understand the wastewater treatment effect and pass the results to the next treatment link so that necessary adjustments can be made according to the current treatment situation. At this time, the system will use the preset objective function and optimization algorithm to dynamically adjust the key parameters in the electrochemical reaction process, including current density, voltage, reaction time, etc. The objective function is usually multi-objective optimization with the goal of maximizing pollutant removal rate, minimizing energy consumption, reducing equipment wear and chemical reagent use. Based on the feedback data of real-time monitoring, the optimization algorithm can automatically adjust the reaction conditions such as current and voltage according to different pollutant concentrations and system requirements. For example, when the pollutant concentration is high, the system may increase the current density or voltage to increase the removal rate; when energy saving is required, the current density and voltage will be automatically lowered to reduce energy consumption and avoid damage to the equipment due to excessive reaction. The self-adjustment mechanism of the system ensures that the wastewater treatment process always remains in the optimal operating state, and can dynamically adjust the reaction conditions according to the changes in real-time data, thereby effectively improving the treatment effect, optimizing energy efficiency and extending the life of the equipment.

[0066] By integrating historical data and real-time monitoring data, a prediction model for wastewater pollutant concentration is established. This process first collects wastewater composition data and treatment effect data over a period of time. These historical data can reflect the pollutant concentration, treatment parameters (such as current density, voltage, temperature, etc.), and corresponding reaction effects in the wastewater treatment process. At the same time, real-time monitoring data are constantly updated during the treatment process. These data include key parameters such as the current wastewater pollutant concentration, current density, and voltage. Through the joint analysis of historical data and real-time data, the machine learning model can reveal the laws and potential trends of changes in wastewater pollutant concentrations. Based on this information, the system can predict changes in pollutant concentrations in wastewater over a period of time in the future, and use this prediction result to guide the dynamic adjustment of wastewater treatment.

[0067] When the prediction results show that the concentration of pollutants in the wastewater will rise, the system will respond in advance and take measures such as increasing the current density or extending the reaction time to enhance the intensity of the electrochemical reaction, thereby more effectively removing pollutants; and when it is predicted that the pollutant concentration may decrease, the system will reduce the reaction intensity accordingly, such as reducing the current density or shortening the reaction time, to avoid energy waste and overreaction. Through this prediction-based advance adjustment strategy, the wastewater treatment system can achieve more efficient and accurate operation, avoiding the passive response to future changes in traditional methods, making the treatment process more flexible and efficient. The predictive ability of the machine learning model enables the entire system to make scientific and reasonable adjustments when facing fluctuations in pollutant concentrations, effectively improving the treatment effect, and also has significant advantages in energy saving and reducing equipment wear.

[0068] Embodiment 1:

[0069] This example treats a batch of industrial wastewater from a chemical plant. The main pollutants in the wastewater are heavy metal ions (such as lead and cadmium). The initial pollutant concentration is C0 = 1500 mg / L. The treatment system removes pollutants through electrochemical reactions. During the treatment process, the pollutant concentration in the wastewater is monitored in real time by sensors, and the above two formulas are used to evaluate the wastewater treatment effect and self-adjust the system.

[0070] After 10 minutes of treatment, the sensor detects that the pollutant concentration in the wastewater has dropped to C t =1000mg / L.

[0071] According to the pollutant removal rate formula:

[0072]

[0073] Substitute the specific values ​​into the formula:

[0074]

[0075] Therefore, after 10 minutes of treatment, the pollutant removal rate is 33.33%. This means that 33.33% of the pollutants in the wastewater have been removed, and the remaining pollutant concentration is 1000mg / L. The result will be fed back to the central control unit in real time through the data transmission module to help the system evaluate the current treatment effect.

[0076] To further evaluate the processing speed, the system calculates the rate of change of pollutant concentration over time through real-time data. The function of the change of pollutant concentration in wastewater over time is set as C(t). If the pollutant concentration drops from 1500mg / L to 1000mg / L during the 10-minute processing, the real-time processing speed can be calculated by the following formula.

[0077]

[0078] In this example, the concentration changes are:

[0079] ΔC=1500-1000=500mg / L

[0080] The time changes are:

[0081] Δt=10min

[0082] Therefore, the real-time processing speed is:

[0083]

[0084] This means that within one minute, the pollutant concentration decreases by 50 mg / L, reflecting the rate of wastewater treatment. The results will also be fed back to the control system through real-time data to optimize the parameters of the electrochemical reaction.

[0085] Based on the real-time feedback of pollutant removal rate and processing speed, the control system will dynamically adjust the electrochemical reaction parameters according to the preset objective function and optimization algorithm. For example, when the pollutant removal rate is low or the processing speed is slow, the system may automatically increase the current density or extend the reaction time to enhance the removal effect of the electrochemical reaction. If, in the above process, the pollutant removal rate is lower than the preset target (such as 80%), the system will adjust the reaction parameters according to an adaptive optimization algorithm (such as a particle swarm optimization algorithm or a genetic algorithm). For example, the system can increase the current density from 10A / m 2 Increase to 15A / m 2 , or extend the reaction time to further improve the treatment effect.

[0086] The range of coefficients involved in the calculation formula is:

[0087] Current density i: generally in the range of 10A / m 2 Up to 30A / m 2 The specific value depends on the wastewater pollutant concentration and treatment requirements. Temperature T: Usually controlled between 25°C and 45°C. Too high a temperature will cause equipment wear, while too low a temperature may affect the reaction rate. Reaction time t: Depending on the pollutant concentration and the target removal rate, the reaction time is generally adjusted between 30 minutes and 2 hours.

[0088] This example treats a batch of wastewater from a chemical plant. The wastewater contains heavy metal ions such as lead and cadmium, and the initial pollutant concentration is C0 = 2000 mg / L. During the wastewater treatment process, the current density of the electrolytic cell is initially set to ρ0 = 15 A / m 2 , voltage is 10V, and reaction time is 60 minutes. Through real-time data monitoring, the system will feedback information such as changes in pollutant concentrations in wastewater, current density, electrolytic cell voltage, equipment wear, etc., and make intelligent adjustments in combination with multi-objective optimization functions.

[0089] The comprehensive optimization objective function used is:

[0090]

[0091] in:

[0092] F opt (t) is the comprehensive optimization objective function at time t, which represents the comprehensive processing goal of each objective in the optimization process.

[0093] w1, w2, and w3 are objective weight coefficients, which are used to adjust the impact of pollutant removal rate, energy efficiency, equipment wear, and processing speed on the objective function. It indicates the pollutant removal rate, reflecting the degree of reduction in pollutant concentration in wastewater. is the energy efficiency ratio, which represents the ratio of the electric power to the current density of the processing speed. σ is the electrode wear coefficient, which reflects the degree of equipment wear. ρ(t) is the current density, which represents the current intensity in the electrolytic reaction. t is the reaction time, which determines the duration of the electrochemical reaction. is the rate of change of pollutant concentration, indicating the processing speed. γ is the influence coefficient of processing speed on the optimization objective function, regulating the contribution of speed to the objective function.

[0094] It is assumed that after 60 minutes of reaction, the pollutant concentration in the wastewater drops to C t =1200mg / L. According to the pollutant removal rate formula, the removal rate can be calculated:

[0095]

[0096] That is, after 60 minutes, 40% of the pollutants in the wastewater were removed.

[0097] During the wastewater treatment process, the power P(t) and current density I(t) of the electrolytic cell are monitored in real time. When set at 60 minutes, the power of the electrolytic cell is P(t) = 300W and the current density is ρ(t) = 20A / m 2 , then the energy efficiency ratio is:

[0098]

[0099] The wear of the electrode will increase with the increase of reaction time. Set it to 60 minutes later, the electrode wear coefficient is σ = 0.05g / min, the reaction time is t = 60min, and the current density is ρ(t) = 20A / m 2 , then the influence of electrode wear is:

[0100] σ·ρ(t)·t=0.05·20·60=60g

[0101] This indicates that the total wear of the electrode was 60 g during the 60 min treatment.

[0102] It is assumed that during the reaction, the pollutant concentration changes linearly within 60 minutes, from C0 = 2000 mg / L to C t =1200mg / L, the rate of change of pollutant concentration is:

[0103]

[0104] This represents a decrease in pollutant concentration of 13.33 mg / L per minute.

[0105] Now, the system will dynamically adjust the current density, voltage, reaction time and other parameters of the electrolyzer based on real-time data feedback and the weight coefficient of the optimization objective function. Set the weight coefficient to the following values:

[0106] w1=0.4,

[0107] w2=0.3,

[0108] w3=0.2,

[0109] γ=0.1.

[0110] Substitute these values ​​into the comprehensive optimization objective function:

[0111]

[0112] Perform the calculation:

[0113] F opt (60) = 0.4 0.4 + 0.3 15 + 0.2 60 + (-13.33) 0-1

[0114] F opt (60) = 0.16 + 4.5 + 121.333 = 15.327

[0115] Comprehensive optimization objective function F opt (60) is 15.327, which means that during the 60-minute treatment process, the effects of pollutant removal rate, energy efficiency, equipment wear and treatment speed are comprehensively considered. The system will dynamically adjust the treatment parameters according to this value to ensure the optimal effect. Through the comprehensive evaluation of the multi-objective optimization function, the system can adjust the current density, voltage, reaction time and other parameters of the electrolytic cell in real time during the wastewater treatment process, which not only improves the removal efficiency of pollutants, but also optimizes energy efficiency and equipment wear, ensuring the high efficiency and sustainability of wastewater treatment.

[0116] The weight coefficients w1, w2, w3 are usually in the range of 0.1 to 0.5, which are used to balance the influence of each optimization target. Current density ρ is usually 10A / m 2 Up to 30A / m 2 Adjust between 0.01 g / min and 0.1 g / min, depending on the electrode material and treatment conditions. Reaction time t: Usually controlled between 30 minutes and 2 hours, adjusted according to the pollutant concentration and target removal rate.

[0117] The wastewater in this example contains a large amount of heavy metal ions such as lead and cadmium, and the initial pollutant concentration is C0 = 2000 mg / L. During the electrolytic cell treatment process, the initial current density I0 = 20 A / m 2 , voltage V0 = 10V, and reaction time T0 = 60min. The system monitors the pollutant concentration C in the wastewater in real time through sensors. t , and dynamically optimizes current density, voltage and reaction time according to preset adjustment formulas to achieve higher processing efficiency and lower energy consumption.

[0118] In practical applications, the dynamic adjustment of current density, voltage and reaction time depends on the ratio of the pollutant concentration in the wastewater to the initial concentration. The specific formula is as follows:

[0119]

[0120] in:

[0121] I(t) is the current density at time t, which determines the intensity of the electrolysis reaction;

[0122] V(t) is the voltage at time t, which determines the voltage intensity of the reaction;

[0123] T r is the reaction time, which indicates the duration of the electrochemical reaction;

[0124] I0, V0, and T0 are the initial current density, voltage, and reaction time, respectively, representing the initial conditions for wastewater treatment;

[0125] β1 and β2 are the adjustment coefficients of current density and voltage, reflecting the influence of pollutant concentration on current and voltage;

[0126] γ is the adjustment coefficient of reaction time, which characterizes the relationship between reaction time and pollutant concentration.

[0127] According to actual wastewater treatment experiments and data analysis, the following are the value ranges of each coefficient:

[0128] β1 = 0.02 to 0.1, indicating the effect of pollutant concentration on current density, and the specific value depends on the nature of the wastewater;

[0129] β2 = 0.01 to 0.05, which indicates the effect of pollutant concentration on voltage. A smaller coefficient means a relatively small voltage change. γ = 0.05 to 0.2, which indicates the effect of pollutant concentration on reaction time. A larger reaction time adjustment coefficient can accelerate or delay the reaction.

[0130] Set at 30 minutes, the pollutant concentration in the wastewater C t =1500mg / L, adjust the parameters according to the above formula. The initial conditions are:

[0131] Current density I0 = 20A / m 2 ;

[0132] Voltage V0 = 10V;

[0133] Reaction time T0 = 60 min.

[0134] The selected adjustment factors are:

[0135] β1=0.05, current density adjustment coefficient;

[0136] β2 = 0.02, voltage adjustment coefficient;

[0137] γ=0.1, reaction time adjustment coefficient.

[0138] According to the formula, the dynamic adjustment value of each parameter is calculated.

[0139] Current density adjustment:

[0140]

[0141] I(30)=20·(1+0.05·0.75)

[0142] I(30)=20·(1+0.0375)=20·1.0375=20.75A / m 2

[0143] Therefore, at 30 minutes, the current density was adjusted to 20.75 A / m 2 .

[0144] Voltage Regulation:

[0145]

[0146] V(30)=10·(1+0.02·0.75)

[0147] V (30) = 10·(1+0.015) = 10·1.015 = 10.15 V

[0148] Therefore, at 30 minutes, the voltage was adjusted to 10.15V.

[0149] Reaction time adjustment:

[0150]

[0151] T r =60·(1+0.1·0.75)

[0152] Tr =60·(1+0.075)=60·1.075=64.5min

[0153] Therefore, the reaction time was adjusted to 64.5 min at 30 min.

[0154] By dynamically adjusting the current density, voltage and reaction time, the system can adapt to changes in pollutant concentration and optimize the wastewater treatment effect. At this time, the current density I(30) = 20.75A / m 2 , voltage V(30) = 10.15 V and reaction time T r =64.5min will accelerate the electrolysis reaction, improve the removal rate of pollutants in wastewater, and reduce energy consumption and equipment wear. If the pollutant concentration monitored by the system in real time is further reduced, the system will continue to optimize the parameters according to the set adjustment rules to ensure that the wastewater treatment efficiency is always in the best state.

[0155] Embodiment 2:

[0156] Take the wastewater treatment of a certain electroplating plant as an example. The main pollutants in the wastewater are nickel (Ni) and chromium (Cr). The initial pollutant concentration is C0 = 800 mg / L. The laboratory monitors the real-time concentration of pollutants in the wastewater during the treatment process. The initial current density, temperature and voltage are set to I0 = 25 A / m 2 , temperature T0 = 35 °C, voltage V0 = 8 V. It is hoped that through intelligent prediction and advance adjustment strategies, historical data can be used to predict the trend of future pollutant concentration changes, and control parameters such as current density, voltage and temperature can be adjusted in advance according to the prediction results.

[0157] First, the real-time collected data needs to be standardized to ensure that the data is within the same scale range, which is convenient for subsequent analysis and model training. Assume that at the beginning of the experiment, the original pollutant concentration data X(t) at a certain moment is 600 mg / L, and there is a data set X(t1), X(t2), ..., X(t n ), and calculate its mean μ X and standard deviation σ X :

[0158]

[0159]

[0160] Set the calculated mean value to μ X =650mg / L, standard deviation is σ X =50mg / L. Then, the standardized value of the original data at a certain time t is:

[0161]

[0162] For example, at t = 30 minutes, the real-time monitored pollutant concentration is X(30) = 600 mg / L, then the standardized result is:

[0163]

[0164] Next, use a smoothing algorithm (such as weighted moving average) to filter the data, remove noise, and capture the dynamic trend of pollutant concentration changes. Introduce a smoothing coefficient α (0<α<1), which controls the strength of data denoising. Set α = 0.2, then the smoothed data X proc (t) will be updated based on the smoothed data of the previous moment and the current data. The specific formula is:

[0165] X proc (t) = αX(t) + (1-α)X proc (t-1)

[0166] For example, when t=30min, the smoothed data X at the previous moment proc (29) = -1.2, then we can calculate:

[0167] X proc (30) = 0.2 × 600 + 0.8 × (-1.2) = 120 - 0.96 = 119.04

[0168] Here, the smoothed data is 119.04. In this way, the system can remove high-frequency noise in the data, making the trend of pollutant concentration changing over time more stable. In order to capture the dynamic trend of pollutant concentration changing over time, it is also necessary to calculate the time series change of pollutant concentration ΔX(t), which reflects the rate of change of pollutant concentration over time. This change is defined as:

[0169] ΔX(t)=X(t)-X(t-Δt)

[0170] Where Δt is the time step. For example, if at t = 30 min, the pollutant concentration is X(30) = 600 mg / L, and at t = 29 min, the concentration is X(29) = 620 mg / L, then the time series change is:

[0171] ΔX(30)=600-620=-20mg / L / min

[0172] This means that the pollutant concentration dropped by 20 mg / L in 1 minute.

[0173] Based on the above data preprocessing and dynamic change calculation, the system can input the processed data into the machine learning model to predict the pollutant concentration. Set a regression analysis model (such as support vector machine regression SVM-R) to predict the changing trend of pollutant concentration, and adjust parameters such as current density, voltage and reaction time in advance according to the prediction results to optimize the treatment effect. For example, through prediction, it is known that the pollutant concentration will continue to decrease in the next 30 minutes. The system can predict and adjust the current density and reaction time to maintain the best treatment efficiency when the pollutant concentration gradually decreases.

[0174] Through the intelligent prediction algorithm, the system predicts that the pollutant concentration will drop to 550 mg / L in the next 30 minutes, so the current density can be slightly reduced to save energy and extend the reaction time to ensure complete purification of the wastewater. Finally, the system adjusts the current density I(t) and reaction time T according to the prediction. r :

[0175] I(t)=I0·(1-β1·ΔX(t))

[0176] T r =T0·(1+γ·ΔX(t))

[0177] In this example, β1 = 0.05 and γ = 0.1 are set, and according to the calculation:

[0178] I(30)=25·(1-0.05×(-20))=25·1.1=27.5A / m 2

[0179] T r =60·(1+0.1×(-20))=60·0.8=48min

[0180] By standardizing, smoothing and calculating changes in data, combined with intelligent prediction algorithms, the wastewater treatment system can monitor pollutant concentrations in real time and adjust parameters such as current density, voltage and response time in advance.

[0181] Furthermore, the wastewater treated in this embodiment is wastewater from a certain electroplating plant, the main pollutants are nickel and chromium, and the initial pollutant concentration is C0 = 800 mg / L. To achieve optimization, a time series-based machine learning model will be constructed to predict the change of pollutant concentration over time in wastewater treatment, and adjust the treatment parameters based on these predictions. The sliding window method is used to extract features from past pollutant concentration, current density, and voltage information to identify the trend of pollutant concentration changes, and train the model to optimize various parameters in the treatment process.

[0182] The sliding window method is a commonly used technique for processing time series data. It captures short-term changes by fixing the window size and helps identify trends in pollutant concentrations. A window size of 10 minutes is selected, and the pollutant concentrations C(t), C(t-1), ..., C(t-9), current density I(t), I(t-1), ..., I(t-9) and voltage V(t), V(t-1), ..., V(t-9) in the past 10 minutes are extracted from each window as input feature data. In this way, the model can learn how these historical data affect future changes in pollutant concentrations.

[0183] After the feature data is prepared, the time series prediction model is used for training. The support vector regression (SVR) model is selected because it performs well in dealing with nonlinear relationships. The model predicts future pollutant concentrations by learning the changing patterns of pollutant concentrations, current density, and voltage.

[0184] The goal of model training is to minimize the error between the predicted pollutant concentration and the actual concentration, so the loss function L is defined model To optimize:

[0185]

[0186] Among them, L model is the loss function, including error minimization and model complexity constraints; N is the number of training data samples; y i is the actual pollutant concentration of the i-th sample; is the pollutant concentration predicted by the model; λ is the regularization coefficient; w j is the weight parameter in the model; M is the total number of parameters in the model. By optimizing this loss function, the optimal model parameters can be obtained.

[0187] The following dataset is used to train the model (all units are mg / L and A / m 2 ):

[0188] Time (minutes) Pollutant concentration CtCtCt Current density Voltage VtVtVt 0 800 25 8 10 780 26 8.1 20 760 27 8.2 30 740 28 8.3 40 720 29 8.4 50 700 30 8.5 60 680 31 8.6

[0189] Select a sliding window size of 10 minutes. For the first sliding window (from 0 to 10 minutes), there are pollutant concentration, current density and voltage data as feature inputs. After setting the model training, it is hoped that these input features can be used to predict the pollutant concentration at 20 minutes.

[0190] Select a sliding window size of 10 minutes. For the first sliding window (from 0 to 10 minutes), there are pollutant concentration, current density and voltage data as feature inputs. After setting the model training, it is hoped that these input features can be used to predict the pollutant concentration at 20 minutes.

[0191] By calculating the loss function, the model training process can be completed through optimization methods such as gradient descent. The regularization coefficient λ is set to 0.1. After each iteration, the model adjusts the weight parameter w according to the error between the actual pollutant concentration and the predicted concentration. j , thereby gradually reducing the error. By continuously optimizing the model, a model that can accurately predict the concentration of pollutants will eventually be obtained. The pollutant concentration predicted by the trained model at the 20th minute is set to Compared with the actual measured value y 20 =740mg / L is exactly the same. This shows that the model can accurately predict the pollutant concentration and provide an effective reference for subsequent wastewater treatment.

[0192] Once the model is trained and verified for accuracy, it can be used to make real-time predictions and adjust the parameters of the wastewater treatment system in advance based on the prediction results. The model is set to predict that the pollutant concentration will continue to decrease in the next 30 minutes, and it is expected that control parameters such as current density, temperature and voltage can be slightly adjusted to save energy.

[0193] For example, the system determines to reduce the current density and reaction time in order to optimize energy consumption and ensure the wastewater treatment effect. By dynamically adjusting these parameters, the system can reduce energy consumption while ensuring the removal rate and improve overall efficiency.

[0194] This example is treating wastewater from a chemical plant. The pollutant concentrations are mainly benzene and chloride. The current pollutant concentration is C(t) = 500 mg / L. The current density in wastewater treatment is I(t) = 35 A / m 2 , voltage V(t) = 9V, temperature T(t) = 25°C, these environmental factors will affect the removal efficiency of pollutants.

[0195] A machine learning model based on historical data has been trained, and the model can predict the changing trend of pollutant concentrations in the future based on these input parameters. However, in actual operation, it is necessary to continuously optimize the model based on new real-time data to ensure the accuracy of the prediction results. To this end, an incremental learning method is introduced. Whenever new data is input, the model will use the gradient update method to update its parameters, so that the model can adapt to new environments and data changes in real time.

[0196] In order to achieve real-time prediction and model update, an incremental learning method is adopted. The real-time prediction formula of the model is set as:

[0197]

[0198] in:

[0199] is the predicted pollutant concentration at the future time t+Δt;

[0200] C(t) is the pollutant concentration at the current time t;

[0201] X env (t) is the current environmental characteristics (including temperature, current density, voltage, etc.);

[0202] θ is the parameter set of the current model;

[0203] f(C(t),X env (t), θ) is the prediction function based on the current pollutant concentration and environmental characteristics;

[0204] η is the learning rate, which controls the step size of each parameter update;

[0205] is the gradient of the loss function with respect to the model parameters, which is used to update the model.

[0206] After training with historical data, a preliminary model is obtained, whose parameters θ include multiple weights and bias terms. In practical applications, these parameters will be dynamically updated according to real-time input data. Set the current pollutant concentration to C(t) = 500 mg / L and the current density to I(t) = 35 A / m 2 , voltage V(t) = 9V, temperature T(t) = 25°C, and the initial prediction result of the model is the pollutant concentration in the next 10 minutes

[0207] On this basis, new real-time data is assumed to be obtained and input into the model:

[0208] Current pollutant concentration C(t) = 500 mg / L

[0209] Current density I(t) = 35A / m 2

[0210] Current voltage V(t) = 9V

[0211] Current temperature T(t) = 25°C

[0212] Learning rate η = 0.01

[0213] Set the loss function L model Gradient

[0214] By inputting these real-time data into the model, the model will calculate updated parameters and optimize its prediction results using incremental learning methods. The prediction error of the model at the 10th minute is set to Where ΔC = 500480 = 20 mg / L. Based on this error, the model will perform gradient updates on its parameters θ to reduce future prediction errors.

[0215] According to the incremental learning formula, the model will update the model parameters based on real-time data and the current prediction error:

[0216]

[0217] Set the initial model parameters to θ old =[0.5, -0.2, 0.1, 0.2], that is, the weight corresponding to each feature (pollutant concentration, current density, voltage, temperature). Through the gradient update method, the updated parameters are:

[0218] θ new =[0.5, -0.2, 0.1, 0.2] 0.01 [0.2, -0.1, 0.05, 0.1] = [0.498, -0.199, 0.095, 0.199]

[0219] Through such updates, the model is able to adjust its parameters in real time, making future predictions more accurate.

[0220] After the model is updated, the new forecast result is set as the pollutant concentration in the next 10 minutes. That is, the pollutant concentration is further reduced compared to the previous prediction value. This prediction result suggests that the wastewater treatment system can respond in advance according to the new prediction results, such as appropriately adjusting the current density and voltage to improve the treatment efficiency.

[0221] Through the incremental learning method, the model can be dynamically updated and future prediction results can be optimized based on the real-time pollutant concentration and environmental characteristics. This method can effectively predict the changing trend of future pollutant concentrations and adjust the operating parameters of the wastewater treatment system in advance based on the prediction results, thereby improving the efficiency and accuracy of wastewater treatment. Through this real-time prediction and model update mechanism, the system can achieve intelligent control and adaptive adjustment in the wastewater treatment process, greatly improving the effect and energy efficiency of wastewater treatment.

Claims

1. Wastewater treatment optimization method based on electrochemical reaction, characterized in that The following steps are involved: S1. Wastewater composition analysis and electrochemical reaction model construction: S1.

1. By configuring a variety of sensors including heavy metal sensors, chemical oxygen demand sensors, and ammonia nitrogen sensors, the concentration of pollutants in wastewater is monitored in real time; the sensor data is transmitted to the central control unit of the treatment system through a wireless communication module; S1.

2. Construct a reaction kinetic model based on the type and concentration of pollutants in the wastewater and the properties of the electrochemical reaction; S2. Adaptive optimization algorithm design: S2.

1. Set optimization goals, calculate the maximum pollutant removal rate, and reduce energy consumption and consumables including electrode wear and chemical reagent use; S2.2, using an adaptive optimization algorithm to dynamically adjust the parameters of the electrochemical reaction; the parameters include current density, voltage of the electrolytic cell, reaction time, and temperature; S3. Real-time data feedback and system self-adjustment mechanism: S3.

1. During the wastewater treatment process, the sensor provides real-time feedback on the pollutant concentration of the wastewater, the voltage in the electrolytic cell, and the current density information; the data is transmitted to the central unit of the control system through the data transmission module; S3.2, by analyzing the data, the current treatment effect including the removal rate of pollutants and the treatment speed is evaluated in real time, and the results are transmitted to the next treatment link; S3.

3. Based on the real-time monitoring data, the parameters of the electrochemical reaction including current and voltage will be dynamically adjusted according to the preset objective function and optimization algorithm; S4. Intelligent prediction and advance adjustment strategy: S4.

1. Based on historical data and real-time monitoring data, a prediction model for wastewater pollutant concentration is established using machine learning methods; based on past wastewater composition and treatment effects, the changing trend of wastewater pollutant concentration in the future is predicted; S4.

2. Make adjustments based on the prediction results; if it is predicted that the pollutant concentration in the wastewater will increase, increase the current or extend the reaction time; if it is predicted that the pollutant concentration will decrease, reduce the reaction intensity.

2. The wastewater treatment optimization method based on electrochemical reaction according to claim 1 is characterized in that The real-time data feedback and system self-adjustment mechanism includes: Based on the real-time collected pollutant concentration data, the current treatment effect is evaluated by dynamically calculating the pollutant removal rate of the wastewater; the pollutant removal rate represents the reduction ratio of the pollutant concentration in the wastewater compared to before treatment; and the pollutant removal rate is calculated using the formula: Among them: F removal (t) is the pollutant removal rate at time t, which indicates the removal ratio of the pollutant concentration after wastewater treatment relative to the initial concentration; C t is the pollutant concentration at time t; C0 is the initial pollutant concentration at the beginning of wastewater treatment; The real-time treatment speed assessment is the rate of change of pollutant concentration in wastewater over time. The progress of wastewater treatment is assessed by the relationship between pollutant concentration and time. The real-time treatment speed assessment formula is used: Where: P rate (t) is the wastewater treatment rate at time t, which represents the rate of change of pollutant concentration over time; It is the time derivative of the pollutant concentration in wastewater, reflecting the change in pollutant concentration per unit time.

3. The wastewater treatment optimization method based on electrochemical reaction according to claim 2 is characterized in that The real-time data feedback and system self-adjustment mechanism includes: introducing a multi-objective optimization function to comprehensively calculate multiple factors such as pollutant removal rate, energy efficiency, and equipment wear; the optimization objective function dynamically adjusts the weight according to real-time data changes, and optimizes the treatment parameters in real time according to the concentration of pollutants in the wastewater and the treatment environment.

4. The wastewater treatment optimization method based on electrochemical reaction according to claim 3 is characterized in that The real-time data feedback and system self-adjustment mechanism includes: Through real-time data feedback and the results of the optimization algorithm, the wastewater treatment system dynamically adjusts the current density, voltage, and reaction time parameters. The dynamic adjustment formula for current density, voltage, and reaction time is: in: I(t) is the current density at time t, which determines the intensity of the electrolytic reaction; V(t) is the voltage at time t, which determines the voltage intensity of the reaction; T r is the reaction time, indicating the duration of the electrochemical reaction; I0, V0, T0 are the initial current density, voltage and reaction time, respectively, representing the initial conditions of wastewater treatment; β1, β2 are the adjustment coefficients of current density and voltage, reflecting the influence of pollutant concentration on current and voltage; γ is the adjustment coefficient of reaction time, characterizing the relationship between reaction time and pollutant concentration.

5. The wastewater treatment optimization method based on electrochemical reaction according to claim 1 is characterized in that The intelligent prediction and advance adjustment strategy method includes: The data is standardized so that the distribution is within the same scale range; the filtering algorithm is used to smooth the data, and then the rate of change of pollutant concentration over time is calculated to capture the dynamic trend of the data and provide time data input for the machine learning model; the data fusion and preprocessing calculation formula is: in: X proc (t) is the pre-processed data at time t; X(t) is the original pollutant concentration, current density, voltage, and temperature parameters at time t; μ X and σ X is the mean and standard deviation of the parameter, which is used for standardization; α is the smoothing coefficient (0<α<1), which controls the strength of data denoising; X proc (t-1) is the smoothed data of the previous moment; ΔX(t)=X(t)X(tΔt) is the time series change, reflecting the time change trend of the pollutant concentration.

6. The wastewater treatment optimization method based on electrochemical reaction according to claim 5 is characterized in that The intelligent prediction and advance adjustment strategy method includes: Construct a machine learning model to learn how pollutant concentrations change over time during wastewater treatment. Use the sliding window method to obtain information on pollutant concentrations, current density, and voltage over a period of time to identify short-term and long-term trends. Select a time series prediction model and train it in combination with feature data. Define the loss function and optimize the model parameters, using the mean square error (MSE) as the objective function and adding an L2 regularization term to prevent overfitting.

7. The wastewater treatment optimization method based on electrochemical reaction according to claim 6 is characterized in that The intelligent prediction and advance adjustment strategy method includes: Using the trained machine learning model, the pollutant concentration change trend in the future is predicted based on the current pollutant concentration and current density environmental factors. The incremental learning method is used to optimize the model parameters using the gradient update method each time new data is input. The real-time prediction and model update formula is: in: is the predicted future pollutant concentration; C(t) is the current pollutant concentration; X env (t) is the environmental characteristics, including temperature, current density, and voltage; θ is the parameter set of the current model; fC(t),X env(t),θ is the prediction function of the machine learning model; η is the learning rate, which controls the step size of parameter update; is the gradient of the loss function with respect to the model parameters, which is used to update the model.

Images