Activated sludge compound control method capable of directly sensing biomechanical signals

Abstract

The invention discloses an activated sludge compound control method capable of directly sensing biomechanical signals, and belongs to the technical field of sewage treatment. According to the method, a hierarchical porous three-dimensional graphene skeleton is used as a microbial carrier, a biocompatible coating is modified on the surface to directionally fix a composite microbial community, and polyvinylidene fluoride (PVDF) piezoelectric material layers cover two sides of the skeleton to construct a piezoelectric sensing unit. A carrier deformation signal caused by microbial metabolic activity is monitored, a biological characteristic component is extracted, and the sludge reflux amount is dynamically adjusted in combination with a feedforward-feedback composite control strategy. The limitation of traditional chemical parameter detection is broken through, the physiological state of microorganisms is directly sensed, the stability, the pollutant removal rate and the energy efficiency ratio of the sewage treatment process are remarkably improved, and the problem that an existing activated sludge method cannot be adjusted in time under the unsteady-state working condition with large water inflow change is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of sewage treatment, and in particular relates to an activated sludge composite control method based on direct perception of biomechanical signals. Background Art

[0002] The activated sludge process is a biological wastewater treatment technology, primarily based on activated sludge. This technology mixes wastewater with activated sludge (microorganisms), agitates them, and aerates them to decompose organic pollutants in the wastewater. Biosolids are then separated from the treated wastewater, with some being returned to the aeration tank as needed. The activated sludge process involves continuously introducing air into the wastewater. Over time, aerobic microorganisms multiply, forming sludge-like flocs. These flocs are inhabited by microorganisms, primarily flocs, which have a strong ability to adsorb and oxidize organic matter.

[0003] The activated sludge process has the advantages of high treatment capacity and good effluent quality, and is the most widely used biological wastewater treatment method in the world today. However, existing sensors are mostly based on chemical parameter detection, which cannot directly reflect the physiological state of microorganisms. It is difficult to establish an accurate activity correlation model and cannot be used to carry out automatic regulation of activated sludge. In addition, the current method mainly relies on regular measurement of sludge concentration, and the sludge return flow rate of the secondary sedimentation tank is adjusted based on the concentration measurement results. In non-steady-state conditions such as large changes in influent volume, timely processing cannot be carried out, resulting in low regulation accuracy, inaccurate dosage, over- or under-adjustment of adjustment results, and ultimately causing unqualified chemical oxygen demand (CODcr). Summary of the Invention

[0004] The technical problem to be solved by the present invention is: to provide a composite control method for activated sludge with direct perception of biomechanical signals, so as to solve the problem that the activated sludge method in the existing technology cannot directly reflect the physiological state of microorganisms based on chemical parameter detection, it is difficult to establish an accurate activity correlation model, and cannot be used to carry out automatic adjustment of activated sludge; it mainly relies on regular measurement of sludge concentration, and adjusts the sludge return flow rate of the secondary sedimentation tank according to the measurement results of the concentration. It cannot be processed in time under non-steady-state conditions such as large changes in water inlet volume, which will result in low adjustment accuracy, inaccurate input amount, over-adjustment or under-adjustment of adjustment results, and ultimately cause the chemical oxygen demand CODcr to be unqualified.

[0005] Technical solution of the present invention: Beneficial effects of the present invention: The present invention directly senses the metabolic activity of microorganisms through biomechanical signals, breaking through the indirect limitations of traditional chemical parameter detection and realizing the essential state monitoring of the treatment process; the integrated design of the piezoelectric sensing unit and the carrier structure effectively captures the dynamic changes of the biofilm and significantly improves the system's response speed to fluctuations in the influent load; the analytical algorithm converts the mechanical signal into a biological activity indicator, establishes a closed-loop control system, and greatly improves the stability and energy efficiency of the sewage treatment process.

[0006] The invention solves the problem that the activated sludge method in the existing technology cannot directly reflect the physiological state of microorganisms based on chemical parameter detection, it is difficult to establish an accurate activity correlation model, and cannot be used to carry out automatic adjustment of activated sludge; it mainly relies on regular measurement of sludge concentration, and adjusts the sludge return flow rate of the secondary sedimentation tank according to the concentration measurement results. It cannot be processed in time under non-steady-state conditions such as large changes in water inflow, which will result in low adjustment accuracy, inaccurate feed amount, over-adjustment or under-adjustment of adjustment results, and ultimately cause unqualified chemical oxygen demand CODcr. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is a flow chart of the present invention.

[0008] Figure 2 This is a diagram of the device of the present invention, in which 1 is a three-dimensional graphene skeleton + polyimide coating, 2 is a piezoelectric material layer + interdigital electrode array, and 3 is a signal lead + silicone rubber. DETAILED DESCRIPTION

[0009] An activated sludge composite control method based on direct biomechanical signal sensing, comprising: A hierarchical porous three-dimensional graphene skeleton is used as a microbial carrier, and the internal surface of the skeleton is modified with a biocompatible coating to directionally fix a composite microbial community of nitrifying bacteria, denitrifying bacteria and polyphosphate bacteria.

[0010] A piezoelectric sensing unit is constructed by covering the skeleton surfaces on both sides of the three-dimensional graphene with a polyvinylidene fluoride (PVDF) piezoelectric material layer.

[0011] The carrier deformation signal caused by microbial metabolic activity is monitored by a piezoelectric sensing unit, and the signal component related to biofilm metabolic activity is extracted; Based on the rate of change of water inlet flow, a feedforward reference regulation amount is generated through fuzzy reasoning. At the same time, a feedback compensation amount is generated through the fuzzy PID algorithm according to the deviation of the biological characteristic component. After superposition, the servo motor is driven to adjust the speed of the reflux pump to achieve the total regulation amount.

[0012] Example 1 A three-dimensional graphene skeleton with hierarchical pores is used, with a porosity of 90%-98%. The porosity calculation formula is: Porosity = V 孔隙 / V 总体积 ×100% in, V 孔隙 is the pore volume, V 总体积 is the total volume of the graphene skeleton.

[0013] The internal surface of the skeleton is modified with a biocompatible coating, which fixes the composite microbial community including nitrifying bacteria, denitrifying bacteria and polyphosphate bacteria through covalent bonds to ensure the high efficiency and stability of microbial metabolic activities.

[0014] The piezoelectric sensing unit includes a piezoelectric material layer conformally bonded to the carrier surface, which is PVDF (polyvinylidene fluoride) or its derivatives, has high sensitivity and biocompatibility, and can efficiently convert the carrier deformation into an electrical signal.

[0015] The piezoelectric material layer is evenly covered on both sides of the three-dimensional graphene skeleton through spin coating or spray coating process, with a thickness of 10-50 microns, ensuring close bonding with the skeleton.

[0016] The piezoelectric sensing unit also includes an interdigital electrode array with a preset spacing, and a signal lead structure encapsulated in a biocompatible material. The encapsulation material is polyimide or silicone rubber to ensure the stability and reliability of signal transmission.

[0017] The original signal is processed using a wavelet packet decomposition algorithm. The decomposition depth of the wavelet packet decomposition algorithm in the retained signal is 4 layers, which is used to filter out effective information of high-frequency mechanical noise with a frequency higher than 500 Hz.

[0018] The biological characteristic components in the multi-channel signal are extracted by blind source separation technology, wherein the blind source separation technology adopts FastICA algorithm to screen out signal components that are highly correlated with the metabolic activity of the biofilm.

[0019] A feedforward-feedback composite control strategy is adopted, with the control target being the sewage return flow, combining the predictability of feedforward control with the correctiveness of feedback control.

[0020] The feedforward module predicts the load fluctuation trend by inferring the change of water inlet flow rate, and based on the water inlet flow rate change rate ( ΔQ 进水 ), predict the baseline adjustment amount of sludge return flow by inference method ( ΔQ 基准 ), the reasoning method includes fuzzy reasoning or neural network reasoning.

[0021] It is characterized by: ΔQ 基准 =F 推理 (ΔQ 进水 ), where: ΔQ 基准Reference water intake, F 推理 is the fuzzy inference or neural network inference function, ΔQ 进水 is the rate of change of water inlet flow.

[0022] According to the deviation of the biometric component from the target value ( e ), using fuzzy PID algorithm to dynamically adjust the sludge return flow compensation ( ΔQ 补偿 ).

[0023] The parameter formula of the fuzzy PID algorithm is: Among them: Kp0, Ki0, Kd0 are initial gains, μ p (e), μ i (e), μ d (e) is the fuzzy adjustment coefficient, and e is the deviation between the biometric component and the target value.

[0024] After the outputs of the feedforward-feedback composite control module are superimposed, a high-precision servo motor is driven to adjust the speed of the reflux pump to ensure dynamic optimization and stable control of the reflux volume during sewage treatment. The superposition formula is: ΔQ 总 =ΔQ 基准 +ΔQ 补偿 Where: ΔQ 总 is the total regulated flow, ΔQ 基准 is the reference water inflow, ΔQ 补偿 To compensate for the water inflow, it is used to drive the servo motor to adjust the speed of the reflux pump.

[0025] Example 2 In order to verify the beneficial effects of the present invention, scientific demonstration was carried out through experiments. 1 Experimental equipment and materials - 3D graphene skeleton (porosity: 95%, pore volume: 0.8 cm³ / cm³) - Biocompatible coating (polyimide) - Piezoelectric material layer (PVDF, thickness: 30 microns) - Interdigitated electrode array (pitch: 0.5 mm) - Signal lead structure (packaging material: silicone rubber) - Servo motor and reflux pump - Data acquisition and control system - Sewage treatment experimental device (simulating changes in influent flow) Microbial community: - Composite microbial communities of nitrifying bacteria, denitrifying bacteria and phosphate accumulation bacteria - Directed fixation on the surface of three-dimensional graphene skeleton through covalent bonds Experimental parameter settings: - Water flow rate range: 10-50 L / h - Wastewater pollutant concentration: CODcr 200-500 mg / L - Experimental period: 72 hours - Data collection frequency: record once every 5 minutes 2. Experimental Procedure Step 1: Preparation of 3D graphene framework and microbial immobilization - The three-dimensional graphene skeleton is prepared using chemical vapor deposition, ensuring a porosity of 95%.

[0026] - The inner surface of the skeleton is modified with a polyimide coating, which fixes the composite microbial community of nitrifying bacteria, denitrifying bacteria and phosphate-accumulating bacteria through covalent bonds.

[0027] Step 2: Piezoelectric Sensing Unit Construction - Spin-coat a PVDF piezoelectric material layer on both sides of the three-dimensional graphene skeleton with a thickness of 30 microns and an area of ​​80% to 90% of the area on both sides of the three-dimensional graphene skeleton.

[0028] - An array of interdigitated electrodes is mounted on the surface of the piezoelectric material layer with a spacing of 0.5 mm.

[0029] - Encapsulate the signal leads in silicone rubber to ensure the stability and reliability of signal transmission.

[0030] The interdigitated electrode array consists of alternating positive and negative electrode fingers. The electrode fingers are formed into periodic gold / platinum metal stripes on the surface of the PVDF film through photolithography. Adjacent electrode fingers are connected to the positive and negative buses respectively. The leads are pre-embedded with micro-channels in the silicone encapsulation layer through laser micromachining, extending the electrode bus to the corrosion-resistant gold-plated copper pins outside the encapsulation layer. The ends of the pins are connected to the data acquisition system using a standard BNC interface. Step 3: Signal Processing and Control System Setup - Perform wavelet noise reduction on the original signal (decomposition depth is 4 layers, filtering out high-frequency mechanical noise above 500 Hz).

[0031] - FastICA algorithm is used to extract signal components that are highly correlated with biofilm metabolic activity.

[0032] - Set up feedforward-feedback composite control strategy: - Feedforward module: Based on the water flow rate change rate ( ΔQ 进水 ) Predicted sludge return flow benchmark adjustment amount ( ΔQ 基准).

[0033] - Feedback module: Based on the deviation of the biometric component from the target value ( e ) Dynamically adjust the sludge return flow compensation amount ( ΔQ 补偿 ).

[0034] - Total adjustment formula: ΔQ 总 = ΔQ 基准 + ΔQ 补偿 .

[0035] Step 4: Experimental Run - Simulate water inlet flow rate changes (10-50 L / h) and record the signal output of the piezoelectric sensor unit.

[0036] - Real-time monitoring of biometric signals and dynamic adjustment of reflux pump speed through the control system.

[0037] - Record return flow rate adjustments, pollutant removal rates, and system energy consumption.

[0038] 3. Experimental Results and Analysis Experimental data record: Table 1: Changes in inlet flow and return flow adjustment Table 2: Pollutant removal rate and system energy consumption Result analysis: Through the feedforward-feedback composite control strategy, the system can quickly respond to changes in the inlet flow rate, and the return flow adjustment amount (ΔQtotal) is highly consistent with the inlet flow rate change trend, ensuring the stability of the sewage treatment process.

[0039] The CODcr removal rate remained between 75% and 90% during the experiment, with an average removal rate of 84%, which was significantly higher than the traditional activated sludge method (about 70%).

[0040] The average system energy consumption is 0.41 kWh / m³, which is about 15% lower than the traditional method, demonstrating the high efficiency and energy saving of this method.

[0041] This experiment demonstrated that a hybrid activated sludge control method based on direct sensing of biomechanical signals can effectively address influent flow fluctuations, dynamically optimize return flow, improve pollutant removal rates, and reduce system energy consumption. This method transcends the limitations of traditional chemical parameter detection and enables essential state monitoring of the wastewater treatment process, demonstrating its high practicality and potential for widespread adoption.

Claims

1. A composite activated sludge control method based on direct biomechanical signal sensing, characterized in that: The following steps are involved: A hierarchical porous three-dimensional graphene skeleton is used as a microbial carrier, and the internal surface of the skeleton is modified with a biocompatible coating to directionally fix a composite microbial community of nitrifying bacteria, denitrifying bacteria, and phosphate-accumulating bacteria. A piezoelectric sensing unit is constructed by covering the surfaces of both sides of the three-dimensional graphene skeleton with a polyvinylidene fluoride (PVDF) piezoelectric material layer; The carrier deformation signal caused by microbial metabolic activity is monitored by a piezoelectric sensing unit, and the biological characteristic components related to biofilm metabolic activity are extracted; Based on the water inlet flow rate change rate, a feedforward reference adjustment amount is generated through fuzzy reasoning. At the same time, a feedback compensation amount is generated through the fuzzy PID algorithm according to the biological characteristic component deviation. After superposition, the servo motor is driven to adjust the speed of the reflux pump to achieve the total adjustment amount.

2. The method according to claim 1, characterized in that The porosity of the three-dimensional graphene skeleton is 90%-98%, and the proportion of pore volume to total volume is calculated by the formula: Porosity Total pore volume Porosity = Vpore / Vtotal volume × 100% Where Vpore is the pore volume and Vtotal volume is the total volume of the graphene skeleton.

3. The method according to claim 1, characterized in that The piezoelectric material layer has a thickness of 10-50 microns and is covered on the surface of the three-dimensional graphene skeleton by spin coating or spray coating, and is combined with an interdigitated electrode array and a biocompatible packaging material to form a piezoelectric sensing unit.

4. The method according to claim 1, wherein The extraction of the biometric component includes: The original signal is decomposed by wavelet packet with a decomposition depth of 4 layers to filter out high-frequency mechanical noise with a frequency higher than 500 Hz; FastICA algorithm was used for blind source separation to screen signal components that were highly correlated with biofilm metabolic activities.

5. The method according to claim 1, wherein A feedforward-feedback composite control strategy is adopted, and the control target is the sewage return flow.

6. The method according to claim 5, characterized in that The generation formula of the feedforward reference adjustment amount is: ΔQ 基准 =F 推理 (ΔQ 进水 ) Where: ΔQ 基准 Reference water intake, F 推理 is the fuzzy inference or neural network inference function, ΔQ 进水 is the rate of change of water inlet flow.

7. The method according to claim 5, characterized in that The feedback compensation is dynamically adjusted by the fuzzy PID algorithm, and its parameter formula is: Among them: Kp0, Ki0, Kd0 are initial gains, μ p (e), μ i (e), μ d (e) is the fuzzy adjustment coefficient, and e is the deviation between the biometric component and the target value.

8. The method according to claim 5, characterized in that The calculation formula of the total adjustment amount is: ΔQ 总 =ΔQ 基准 +ΔQ 补偿 Where: ΔQ 总 is the total regulated flow, ΔQ 基准 is the reference water inflow, ΔQ 补偿 To compensate for the water inflow, it is used to drive the servo motor to adjust the speed of the reflux pump.

9. The method according to claim 5, characterized in that The method is suitable for sewage treatment under non-steady-state conditions, and can dynamically optimize the return flow, improve the CODcr removal rate and reduce system energy consumption.

Images