Microbial fuel cell coupling system for industrial wastewater treatment and energy regeneration

Abstract

The invention discloses a microbial fuel cell coupling system for industrial wastewater treatment and energy regeneration, and the system comprises a wastewater treatment and energy conversion module which is used for introducing industrial wastewater into a microbial fuel cell, realizing efficient decomposition of organic matters under the metabolism effect of microbiota in an anode chamber, generating purified water reaching the standard, and recovering clean energy; the intelligent parameter control module is used for dynamically adjusting operation parameters of each unit through an MLP model; and the dynamic monitoring feedback module carries a high-precision sensor array and an internet-of-things transmission device, acquires parameters in the fuel cell in real time, provides a predictive regulation and control strategy for the system by applying an MLP model, and forms closed-loop optimization through an energy-material-information three-flow coupling mechanism. Through cooperative work of all the modules, efficient and stable operation of the system is achieved, the prospect in the field of industrial wastewater treatment is wide, the resource utilization efficiency can be remarkably improved, and the cost is reduced.

Description

Technical Field

[0001] The present invention relates to a microbial fuel cell coupling system, in particular to a microbial fuel cell coupling system for treating industrial wastewater and realizing energy regeneration. Background Art

[0002] In recent years, with the rapid development of industry, the discharge of industrial wastewater has continued to increase, causing serious environmental pollution. At the same time, the growing global demand for energy has also placed tremendous pressure on energy supply. Against this backdrop, the development of technologies that can simultaneously address the problems of industrial wastewater treatment and energy recovery is of paramount importance. Microbial fuel cells (MFCs), as an emerging bio-electrochemical system, can utilize the metabolic activity of microorganisms to convert organic matter in wastewater into electricity, effectively integrating wastewater treatment with energy recovery. MFC technology offers the following advantages: First, it can effectively remove organic pollutants from wastewater, reducing chemical oxygen demand (COD) and biochemical oxygen demand (BOD); second, it can generate electricity, achieving energy self-sufficiency and even export; and third, it offers low operating costs, simple operation, and environmental friendliness. Despite its numerous theoretical advantages, MFC technology still faces several challenges in practical application. For example, the power generation efficiency and wastewater treatment capacity of MFCs are influenced by multiple factors, such as microbial community structure, electrode materials, reactor configuration, and operating parameters, requiring further optimization and regulation. In addition, how to realize intelligent control of MFC system and improve its adaptability and stability is also a hot topic in current research. Summary of the Invention

[0003] Purpose of the invention: The purpose of the present invention is to provide a microbial fuel cell coupling system for industrial wastewater treatment and energy regeneration. By integrating advanced sensor technology, intelligent control algorithms and MFC technology, efficient treatment of industrial wastewater and energy recovery and utilization can be achieved, thereby improving the overall performance of the system and resource utilization efficiency.

[0004] Technical solution: The present invention provides a microbial fuel cell coupled system for industrial wastewater treatment and energy regeneration, comprising:

[0005] Wastewater treatment and energy conversion module: This module directs industrial wastewater into the microbial fuel cell, where it is metabolized by the microbial community in the anode chamber to efficiently decompose organic matter, simultaneously releasing electrons and generating bioelectricity. The cathode chamber then undergoes redox reactions to generate a stable current output, producing qualified purified water and recovering clean energy.

[0006] Intelligent parameter control module: including microbial activity control unit, electrolyte concentration control unit, oxygen concentration control unit, water flow rate control unit, temperature control unit, pH adjustment unit and substrate concentration control unit, dynamically adjusting the operating parameters of each unit through the MLP model;

[0007] Dynamic monitoring feedback module: equipped with a high-precision sensor array and IoT transmission device, it collects real-time data in the fuel cell, including but not limited to biofilm activity, ion migration rate, dissolved oxygen gradient, hydraulic retention time, reaction temperature field, pH fluctuation curve and organic load distribution parameters. It uses the MLP model to provide a predictive control strategy for the system, and forms a closed-loop optimization through the energy-matter-information three-flow coupling mechanism.

[0008] Preferably, the wastewater treatment and energy conversion module includes: passing the wastewater through a screen to remove larger solid impurities, flowing it into a grit chamber to allow the sand to settle, and then balancing the water quality, water volume and pH value of the wastewater in a regulating tank to ensure the stability of various indicators of the wastewater; the pretreated wastewater enters the microbial fuel cell module, and under the action of microbial metabolism, the organic matter in the wastewater is converted into electrical energy, and purified water is produced and discharged, thereby realizing wastewater treatment and energy recovery.

[0009] Preferably, in the wastewater treatment and energy conversion module, the microbial fuel cell uses microorganisms as anode catalysts to convert chemical energy in the wastewater into electrical energy; in the anode chamber, the microorganisms oxidize and decompose organic matter in the wastewater into small molecules and release electrons; the electrons are transferred to the anode through the cell membrane and transferred to the cathode along the external circuit to form an electric current; the electrons transferred to the cathode react with electron acceptors to generate non-polluting products; anions in the solution are transferred to the anode, and cations are transferred to the cathode; the treated purified water is output, and the generated electrical energy is stored or used for energy supply.

[0010] Preferably, the intelligent parameter control module and the microbial activity control unit regulate the living environment of the microorganisms, including providing nutrients, controlling the concentration of harmful substances, and enhancing the ability to decompose organic matter in wastewater; the temperature control unit regulates the temperature in the system according to the suitable living temperature range of the microorganisms.

[0011] Preferably, the dynamic monitoring feedback module collects key data, including the chemical oxygen demand, electricity generation, and microbial activity data of the wastewater, and transmits them to the intelligent parameter control module in real time. The intelligent parameter control module adjusts each control unit based on these data to achieve dynamic balance and collaborative work between modules.

[0012] Preferably, the intelligent parameter control module uses an MLP model to regulate system operating parameters, and the implementation process includes:

[0013] (1) Obtain the electricity data generated by the treated wastewater, the wastewater quality data, and the operating status data of each module of the system from the wastewater treatment and energy conversion module, and preset the target amount of industrial wastewater treatment efficiency and electricity output; based on this data, calculate the adjustment direction and amplitude of each control unit in the parameter control module, and then perform preliminary regulation on the system operating parameters;

[0014] (2) The MLP model is optimized using the Adam optimization algorithm. During the model training process, the training data is continuously monitored to analyze the impact of the current parameter adjustment on the model prediction accuracy and convergence speed. A gradient calculation model is established to evaluate the effectiveness of each parameter update. If the model performance is significantly improved after the update, the update strategy is strengthened. If the update effect is not good, the update strategy is recalculated based on the gradient information to optimize the weight parameters of each layer.

[0015] Preferably, step (1) includes collecting key data of the wastewater treatment and energy conversion modules, including data on electricity generated by treating wastewater, data on chemical oxygen demand of wastewater, data on microbial activity, and operating status data of each module of the system; measuring the current wastewater treatment efficiency according to the change in chemical oxygen demand of wastewater, and calculating the current electricity output using the collected electricity data; and calculating the error between the current performance index and the target amount:

[0016] Wastewater treatment efficiency error:

[0017] e COD =COD target -COD current

[0018] Power output error:

[0019] e P =P target -P current

[0020] Among them, COD target is the target wastewater treatment efficiency, COD current is the current wastewater treatment efficiency, P target is the target power output, P current is the current electrical energy output;

[0021] Combining the wastewater treatment efficiency error and the power output error, the objective function is constructed:

[0022]

[0023] Among them, e COD is the wastewater treatment efficiency error, e P is the power output error, α and β weight coefficients, satisfying α + β = 1;

[0024] Calculate the gradient of the objective function J with respect to the parameters of each control unit; the control unit parameter is θ, and the gradient calculation formula is:

[0025]

[0026] According to the direction and magnitude of the gradient, the adjustment direction and amplitude of the parameters of each control unit are determined, and the adjustment amplitude is determined based on the magnitude of the gradient and the learning rate.

[0027] Preferably, the adjustment direction is: if the gradient is positive, then the parameter value is reduced; conversely, if the gradient is negative, then the parameter value is increased;

[0028] The adjustment range calculation formula is:

[0029]

[0030] Among them, Δθ is the parameter adjustment amplitude, η is the learning rate;

[0031] According to the calculated adjustment direction and amplitude, update the parameters of each control unit. The update formula is:

[0032] θ new =θ old +Δθ

[0033] Among them, θ new is the updated parameter value, θ old is the parameter value before updating.

[0034] Preferably, the step (2) includes: calculating the gradient g under the current parameters during the optimization of the MLP model t , the Adam optimization algorithm estimates m by calculating the first-order moment of the gradient t and the second moment estimate v t To dynamically adjust the learning rate, the formula is as follows:

[0035] m t =β1m t-1 +(1-β1)g t

[0036]

[0037] Among them, β1 and β2 are the exponential decay rates of the first-order moment and the second-order moment respectively, and then m t and v t To perform bias correction:

[0038]

[0039] Finally update the model parameters:

[0040]

[0041] Here, η is the learning rate and ε is a small constant to prevent division by zero.

[0042] Preferably, the step (2) includes: using mean square error to perform performance evaluation, the formula is as follows:

[0043]

[0044] Among them, y i and are the true value and the predicted value respectively, and n is the number of samples.

[0045] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0046] (1) The present invention adopts the MLP model to balance wastewater treatment efficiency and power output by dynamically adjusting parameters during system operation, effectively avoiding resource waste, achieving a balance between system exploration and exploitation, and ensuring that the system continuously and stably operates towards the goal of maximizing resource utilization in complex and changeable industrial wastewater treatment scenarios.

[0047] (2) In the wastewater treatment and energy conversion module, the present invention utilizes microorganisms as anode catalysts to decompose organic matter in the wastewater and release electrons to form electric current, thereby solving the problem that traditional wastewater treatment technology cannot simultaneously recover energy, and achieving the dual benefits of wastewater purification and electricity production.

[0048] (3) The present invention utilizes an intelligent parameter control module to dynamically optimize the MLP model through the Adam optimization algorithm, and adjusts the parameters of each control unit in combination with gradient calculation and deviation correction mechanism, thereby solving the problems of parameter control lag and low convergence efficiency in traditional MFC systems and improving the model prediction accuracy and system response speed.

[0049] (4) The present invention solves the problems of delayed acquisition of fuel cell operating parameters and single data dimension in traditional systems by equipping a dynamic monitoring feedback module with a high-precision sensor array and an Internet of Things transmission device, and achieves the effect of real-time acquisition of multi-dimensional parameters such as biofilm activity and ion migration rate, providing accurate data support for system regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0051] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0052] The present invention provides a microbial fuel cell coupling system for industrial wastewater treatment and energy regeneration, such as Figure 1Shown, including:

[0053] Wastewater treatment and energy conversion module: This module directs industrial wastewater into the microbial fuel cell, where it is metabolized by the microbial community in the anode chamber to efficiently decompose organic matter, simultaneously releasing electrons and generating bioelectricity. The cathode chamber then undergoes redox reactions to generate a stable current output, producing qualified purified water and recovering clean energy.

[0054] Intelligent parameter control module: including microbial activity control unit, electrolyte concentration control unit, oxygen concentration control unit, water flow rate control unit, temperature control unit, pH adjustment unit and substrate concentration control unit, dynamically adjusting the operating parameters of each unit through the MLP model;

[0055] Dynamic monitoring feedback module: equipped with a high-precision sensor array and IoT transmission device, it collects real-time data in the fuel cell, including but not limited to biofilm activity, ion migration rate, dissolved oxygen gradient, hydraulic retention time, reaction temperature field, pH fluctuation curve and organic load distribution parameters. It uses the MLP model to provide a predictive control strategy for the system, and forms a closed-loop optimization through the energy-matter-information three-flow coupling mechanism.

[0056] The wastewater treatment and energy conversion module uses a collection system consisting of collection pipes and pumping stations throughout the factory to efficiently collect industrial wastewater. Subsequently, the wastewater enters the pretreatment stage. First, it passes through a screen to remove larger solid impurities, then flows into a grit chamber to precipitate the sand particles, and then the water quality, water volume and pH value of the wastewater are balanced and adjusted in the regulating tank to ensure that various indicators of the wastewater are stable. The pretreated wastewater enters the microbial fuel cell module. Under the action of microbial metabolism, the organic matter in the wastewater is converted into electrical energy, and at the same time, purified water is produced and discharged, realizing wastewater treatment and energy recovery, and improving resource utilization efficiency and environmental benefits.

[0057] In the wastewater treatment and energy conversion module, the microbial fuel cell uses microorganisms as anode catalysts to convert the chemical energy in the wastewater into electrical energy. In the anode chamber, microorganisms oxidize and decompose the organic matter in the wastewater into small molecules and release electrons; the electrons are transferred to the anode through the cell membrane and transferred to the cathode along the external circuit to form an electric current; the electrons transferred to the cathode react with the electron acceptor to generate non-polluting products; the anions in the solution transfer to the anode and the cations transfer to the cathode. The treated purified water is output from the module, and the generated electricity is stored or used for power supply. This process not only effectively reduces the organic pollutants in the wastewater and reduces environmental pollution, but also achieves efficient resource conversion through energy recycling and utilization, with significant environmental and economic benefits.

[0058] The microbial activity control unit in the intelligent parameter control module regulates the living environment of the microorganisms, such as providing suitable nutrients and controlling the concentration of harmful substances, so as to maintain the optimal activity state of the microorganisms and enhance their ability to decompose organic matter in wastewater. The temperature control unit adjusts the temperature within the system according to the suitable living temperature range of the microorganisms to ensure that the metabolic process of the microorganisms can proceed stably, thereby improving system performance and power output.

[0059] The monitoring and feedback module collects key data, such as the wastewater's chemical oxygen demand (COD), electricity generation, and microbial activity data, and transmits it to the parameter control module in real time. Based on this data, the parameter control module adjusts each control unit accordingly. For example, if a decrease in COD removal rate is detected, the substrate concentration control unit is adjusted to increase the substrate concentration to enhance microbial metabolic activity, thereby improving wastewater treatment efficiency and achieving dynamic balance and collaborative operation between modules.

[0060] The wastewater treatment and energy conversion module leverages the metabolic activity of microorganisms to efficiently convert chemical energy in wastewater into electrical energy. A portion of this generated energy is used to maintain system operation, optimize system performance through the intelligent parameter control module's adjustment devices, and support data collection and transmission in the monitoring and feedback module. The remaining energy is stored in an energy storage device or connected to the grid for external output, achieving energy recovery, improving resource utilization efficiency, and reducing reliance on external energy sources.

[0061] The intelligent parameter control module uses the MLP model to regulate system operating parameters, optimizing industrial wastewater treatment efficiency and power output, thereby maximizing resource utilization. The implementation process is as follows:

[0062] (1) Data on the electricity generated by treated wastewater, wastewater quality, and the operating status of each system module are obtained from the wastewater treatment and energy conversion modules. Targets for industrial wastewater treatment efficiency and electricity output are also preset. Based on this data, the adjustment direction and amplitude of each control unit in the parameter control module are calculated, and preliminary adjustments to the system operating parameters are then made.

[0063] (2) The MLP model is optimized using the Adam optimization algorithm. During model training, the training data is continuously monitored to analyze the impact of current parameter adjustments on the model's prediction accuracy and convergence speed. By establishing a gradient calculation model, the effectiveness of each parameter update is evaluated. If the model performance improves significantly after the update, the update strategy is strengthened; if the update effect is not good, the update strategy is recalculated based on the gradient information, and the weight parameters of each layer are optimized to avoid the waste of computing resources caused by invalid updates and ensure that the model always converges to the optimal state.

[0064] The step (1) specifically includes:

[0065] Data collection and performance evaluation: Collect key data from the wastewater treatment and energy conversion modules, including data on electricity generated by wastewater treatment, chemical oxygen demand (COD) data, microbial activity data, and operating status data for each system module (such as temperature, pH value, water flow rate, etc.). The current wastewater treatment efficiency is measured based on changes in the chemical oxygen demand (COD) of the wastewater. The collected electricity data is used to calculate the current electricity output. Calculate the error between the current performance indicator and the target amount:

[0066] Wastewater treatment efficiency error:

[0067] e COD =COD target -COD current

[0068] Power output error:

[0069] e P =P target -P current

[0070] Among them, COD target is the target wastewater treatment efficiency, COD current is the current wastewater treatment efficiency, P target is the target power output, P current is the current electrical energy output.

[0071] Gradient calculation and adjustment determination: Combine the wastewater treatment efficiency error and the power output error to construct the objective function:

[0072]

[0073] Among them, e COD is the wastewater treatment efficiency error, e P is the power output error, and α and β are weight coefficients, satisfying α+β=1.

[0074] Calculate the gradient of the objective function J with respect to the parameters of each control unit. Assuming that the control unit parameter is θ, the gradient calculation formula is:

[0075]

[0076] According to the direction and magnitude of the gradient, the adjustment direction and magnitude of the control unit parameters are determined. The adjustment magnitude can be determined based on the magnitude of the gradient and the learning rate:

[0077] Adjustment direction: The sign of the gradient determines the direction of parameter adjustment. If the gradient is positive, it means that increasing the parameter value will increase the objective function, and the parameter value should be decreased; conversely, if the gradient is negative, the parameter value should be increased.

[0078] Adjustment range: The adjustment range is calculated as follows:

[0079]

[0080] Among them, Δθ is the parameter adjustment amplitude and η is the learning rate.

[0081] Parameter update and verification feedback: Update the parameters of each control unit according to the calculated adjustment direction and amplitude. The update formula is:

[0082] θ new =θ old +Δθ

[0083] Among them, θ new is the updated parameter value, θ old is the parameter value before updating

[0084] After updating the parameters, collect system operation data again, calculate the new objective function value, and verify the effect of the parameter adjustment. If the new objective function value indicates that the system performance has improved, continue to use the current adjustment strategy; otherwise, recalculate the gradient and adjustment range and proceed to the next round of parameter optimization.

[0085] The step (2) specifically includes:

[0086] Gradient calculation and parameter update: In each iteration, the gradient g under the current parameters is calculated t The Adam optimization algorithm estimates m by calculating the first moment of the gradient t and the second moment estimate v t To dynamically adjust the learning rate, the formula is as follows:

[0087] m t =β1m t-1 +(1-β1)g t

[0088]

[0089] Among them, β1 and β2 are the exponential decay rates of the first-order moment and the second-order moment, respectively, which are usually 0.9 and 0.999. Then, for m t and v t To perform bias correction:

[0090]

[0091] Finally update the model parameters:

[0092]

[0093] Where η is the learning rate and ε is a small constant to prevent division by zero, usually 10 -8In this way, the Adam optimization algorithm can adaptively adjust the learning rate of each parameter to improve convergence speed and stability.

[0094] After each parameter update, the effectiveness of the update is evaluated using performance metrics on the validation or training set. If the model performance improves significantly after the update, the current update strategy is continued. If the update is ineffective, the update strategy is recalculated based on the gradient information and the weight parameters of each layer are adjusted to avoid wasting computing resources due to ineffective updates and ensure that the model always converges to the optimal state.

[0095] Performance evaluation and feedback adjustment: Performance evaluation uses multiple indicators to comprehensively measure model performance. The mean square error (MSE) is used:

[0096]

[0097] Among them, y i and where η and β are the true and predicted values, respectively, and n is the number of samples. Performance evaluation results are fed back to the parameter optimization module to guide its optimization strategy. If the model's MSE decreases slowly or shows an upward trend over multiple iterations, this indicates that the current learning rate may be too large or too small. The parameter optimization module can then appropriately adjust the learning rate η or the momentum parameter β1 to accelerate convergence and avoid local optimality. This interaction of data flow, control flow, and information flow forms a closed-loop optimization loop, achieving efficient utilization of computing resources and continuous improvement of model performance.

Claims

1. A microbial fuel cell coupling system for industrial wastewater treatment and energy regeneration, characterized in that: The system comprises: Wastewater treatment and energy conversion module: This module directs industrial wastewater into the microbial fuel cell, where it is metabolized by the microbial community in the anode chamber to efficiently decompose organic matter, simultaneously releasing electrons and generating bioelectricity. The cathode chamber then undergoes redox reactions to generate a stable current output, producing qualified purified water and recovering clean energy. Intelligent parameter control module: including microbial activity control unit, electrolyte concentration control unit, oxygen concentration control unit, water flow rate control unit, temperature control unit, pH adjustment unit and substrate concentration control unit, dynamically adjusting the operating parameters of each unit through the MLP model; Dynamic monitoring feedback module: equipped with a high-precision sensor array and IoT transmission device, it collects real-time data in the fuel cell, including but not limited to biofilm activity, ion migration rate, dissolved oxygen gradient, hydraulic retention time, reaction temperature field, pH fluctuation curve and organic load distribution parameters. It uses the MLP model to provide a predictive control strategy for the system, and forms a closed-loop optimization through the energy-matter-information three-flow coupling mechanism.

2. The microbial fuel cell coupling system for industrial wastewater treatment and energy regeneration according to claim 1, characterized in that: The wastewater treatment and energy conversion module includes: passing the wastewater through a screen to remove larger solid impurities, flowing it into a grit chamber to allow the sand to settle, and then balancing the water quality, water volume and pH value of the wastewater in a regulating tank to ensure the stability of various indicators of the wastewater; the pretreated wastewater enters the microbial fuel cell module, where, under the action of microbial metabolism, organic matter in the wastewater is converted into electrical energy, and purified water is simultaneously produced and discharged, realizing wastewater treatment and energy recovery.

3. The microbial fuel cell coupled system for industrial wastewater treatment and energy regeneration according to claim 1, characterized in that: In the wastewater treatment and energy conversion module, the microbial fuel cell uses microorganisms as anode catalysts to convert chemical energy in the wastewater into electrical energy. In the anode chamber, the microorganisms oxidize and decompose organic matter in the wastewater into small molecules and release electrons. Electrons are transferred to the anode through the cell membrane and then to the cathode along the external circuit, forming an electric current. The electrons transferred to the cathode react with electron acceptors to generate non-polluting products. Anions in the solution transfer to the anode, and cations transfer to the cathode; the treated purified water is output, and the generated electrical energy is stored or supplied.

4. The microbial fuel cell coupling system for industrial wastewater treatment and energy regeneration according to claim 1, characterized in that: The intelligent parameter control module and the microbial activity control unit regulate the living environment of microorganisms, including providing nutrients, controlling the concentration of harmful substances, and enhancing the ability to decompose organic matter in wastewater; the temperature control unit regulates the temperature in the system according to the suitable living temperature range of microorganisms.

5. The microbial fuel cell coupled system for industrial wastewater treatment and energy regeneration according to claim 1, characterized in that: The dynamic monitoring feedback module collects key data, including the chemical oxygen demand of the wastewater, electricity generation, and microbial activity data, and transmits it to the intelligent parameter control module in real time. The intelligent parameter control module adjusts each control unit based on this data to achieve dynamic balance and collaborative work between modules.

6. The microbial fuel cell coupled system for industrial wastewater treatment and energy regeneration according to claim 1, characterized in that: The intelligent parameter control module uses the MLP model to control system operating parameters. The implementation process includes: (1) Obtain the electricity data generated by the treated wastewater, the wastewater quality data, and the operating status data of each module of the system from the wastewater treatment and energy conversion module, and preset the target amount of industrial wastewater treatment efficiency and electricity output; based on this data, calculate the adjustment direction and amplitude of each control unit in the parameter control module, and then perform preliminary regulation on the system operating parameters; (2) The MLP model is optimized using the Adam optimization algorithm. During the model training process, the training data is continuously monitored to analyze the impact of the current parameter adjustment on the model prediction accuracy and convergence speed. A gradient calculation model is established to evaluate the effectiveness of each parameter update. If the model performance is significantly improved after the update, the update strategy is strengthened. If the update effect is not good, the update strategy is recalculated based on the gradient information to optimize the weight parameters of each layer.

7. The microbial fuel cell coupled system for industrial wastewater treatment and energy regeneration according to claim 6, characterized in that: The step (1) includes collecting key data of the wastewater treatment and energy conversion modules, including data on electricity generated by treating wastewater, data on chemical oxygen demand of the wastewater, data on microbial activity, and data on the operating status of each module of the system; Measure the current wastewater treatment efficiency based on the change in chemical oxygen demand of the wastewater, and use the collected electricity data to calculate the current electricity output; Calculate the error between the current performance indicator and the target quantity: Wastewater treatment efficiency error: it is COD =COD target -CODE current Power output error: e P =P target -P current Among them, COD target is the target wastewater treatment efficiency, COD current is the current wastewater treatment efficiency, P target is the target power output, P current is the current electrical energy output; Combining the wastewater treatment efficiency error and the power output error, the objective function is constructed: Among them, e COD is the wastewater treatment efficiency error, e P is the power output error, α and β weight coefficients, satisfying α + β = 1; Calculate the gradient of the objective function J with respect to the parameters of each control unit; the control unit parameter is θ, and the gradient calculation formula is: According to the direction and magnitude of the gradient, the adjustment direction and amplitude of the parameters of each control unit are determined, and the adjustment amplitude is determined based on the magnitude of the gradient and the learning rate.

8. The microbial fuel cell coupled system for industrial wastewater treatment and energy regeneration according to claim 6, characterized in that: The adjustment direction is: if the gradient is positive, then the parameter value is reduced; conversely, if the gradient is negative, then the parameter value is increased; The adjustment range calculation formula is: Among them, Δθ is the parameter adjustment amplitude, η is the learning rate; According to the calculated adjustment direction and amplitude, update the parameters of each control unit. The update formula is: i new =θ old +Δθ Among them, θ new is the updated parameter value, θ old is the parameter value before updating.

9. The microbial fuel cell coupled system for industrial wastewater treatment and energy regeneration according to claim 6, characterized in that: The step (2) includes: calculating the gradient g under the current parameters during the optimization of the MLP model t , the Adam optimization algorithm estimates m by calculating the first-order moment of the gradient t and the second-order moment estimate v t To dynamically adjust the learning rate, the formula is as follows: m t =β1m t-1 +(1-β1)g t Among them, β1 and β2 are the exponential decay rates of the first-order moment and the second-order moment respectively, and then m t and v t To perform bias correction: Finally update the model parameters: Here, η is the learning rate and ε is a small constant to prevent division by zero.

10. The microbial fuel cell coupled system for industrial wastewater treatment and energy regeneration according to claim 6, characterized in that: The step (2) includes: using mean square error to perform performance evaluation, the formula is as follows: Among them, y i and are the true value and the predicted value respectively, and n is the number of samples.

Images