Settling basin flocculation settling control method and device, electronic equipment and storage medium

Abstract

The application provides a sedimentation tank flocculation sedimentation control method and device, electronic equipment and storage medium, including: obtaining raw water quality target parameters; based on the raw water quality target parameters, feature extraction and noise reduction processing are performed through an adaptive mode decomposition algorithm to obtain a feature vector; the feature vector is processed by using an extreme learning machine model optimized by an improved genetic algorithm to obtain a coagulant dosage prediction value; based on the coagulant dosage prediction value, coagulant is added to the raw water, and the operating parameters of the sedimentation tank are monitored; based on the operating parameters, the dry sludge amount in the sedimentation tank is determined through a dry sludge amount model; if it is determined that the sludge discharge condition is met based on the dry sludge amount and a pre-set sludge level threshold value, sludge is discharged; the effluent turbidity of the sedimentation tank is monitored in real time during the sludge discharge process, and the coagulant dosage is dynamically adjusted based on the effluent turbidity through the adaptive mode decomposition algorithm. In this way, the coagulant dosage accuracy and process synergy are improved, and the sedimentation tank is discharged on demand.

Description

Technical Field

[0001] This invention relates to the field of flocculation and sedimentation technology, and in particular to a method, apparatus, electronic device, and storage medium for controlling flocculation and sedimentation in sedimentation tanks. Background Technology

[0002] Coagulation and sedimentation are the core series of processes in water treatment plants for removing suspended solids, colloidal particles and algae from raw water: first, coagulants are added to make impurities in the water form flocs, and then the flocs are separated by sedimentation in a sedimentation tank.

[0003] In traditional water treatment plants, coagulant dosing and sedimentation tank sludge removal are often controlled as independent processes without a synergistic optimization mechanism. Adjustments to coagulant dosage fail to consider the sludge load and sedimentation capacity of the subsequent sedimentation tank, leading to increased sludge accumulation when the dosage is too high. Furthermore, feedback on sedimentation tank sludge removal performance does not provide guidance for optimizing coagulant dosage. When incomplete sludge removal results in excessive effluent turbidity, it is impossible to adjust coagulant dosage parameters in a timely manner to improve floc formation. This disconnect in processes leads to low operating efficiency of the entire coagulation-sedimentation system, making it difficult to achieve the dual goals of meeting water quality standards and controlling costs. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a method, device, electronic equipment and storage medium for controlling flocculation and sedimentation in sedimentation tanks, which improves the accuracy of coagulant dosing and process synergy, realizes on-demand sludge discharge from sedimentation tanks, thereby improving the treatment efficiency and water quality stability of water supply plants, and reducing energy consumption and operation and maintenance costs.

[0005] In a first aspect, embodiments of the present invention provide a method for controlling flocculation and sedimentation in a sedimentation tank. The method includes: obtaining target parameters of raw water quality; extracting and denoising features based on the target parameters of raw water quality using an adaptive mode decomposition algorithm to obtain feature vectors; processing the feature vectors using an optimized extreme learning machine model based on an improved genetic algorithm to obtain a predicted value for coagulant dosage; adding coagulant to the raw water based on the predicted value for coagulant dosage and monitoring the operating parameters of the sedimentation tank; determining the amount of dry sludge in the sedimentation tank based on the operating parameters using a dry sludge quantity model; if the sludge discharge conditions are met based on the amount of dry sludge and a pre-set sludge level threshold, then discharging sludge; and monitoring the turbidity of the effluent from the sedimentation tank in real time during the sludge discharge process, and dynamically adjusting the coagulant dosage based on the effluent turbidity using an adaptive mode decomposition algorithm.

[0006] In a preferred embodiment of the present invention, the above-mentioned acquisition of raw water quality target parameters includes: collecting raw water quality parameters; normalizing the raw water quality parameters to obtain raw water quality parameter mapping values; and using Pearson correlation analysis based on the raw water quality parameter mapping values ​​to screen out parameters that are significantly correlated with the coagulant dosage as raw water quality target parameters.

[0007] In a preferred embodiment of the present invention, the above-mentioned feature vector is obtained by feature extraction and noise reduction based on the target parameters of raw water quality through an adaptive mode decomposition algorithm, including: setting the search range of the number of modes and performing multimodal decomposition on the time series data of water quality parameters; iteratively updating the modal components, center frequencies and Lagrange multipliers until the convergence condition is met; constructing a comprehensive evaluation index based on the average sample entropy and average kurtosis, and selecting the optimal number of modes; retaining the key modal components and splicing them in order to form a feature vector, and normalizing the feature vector.

[0008] In a preferred embodiment of the present invention, the above-mentioned use of an optimized Extreme Learning Machine (ELM) model based on an improved genetic algorithm to process feature vectors and obtain predicted values ​​for coagulant dosage includes: defining the architecture of the ELM model; setting the parameters of the improved genetic algorithm; globally optimizing the weights and bias parameters of the ELM model using the improved genetic algorithm; training the ELM model based on the optimized parameters; and inputting the feature vectors into the trained ELM model to obtain predicted values ​​for coagulant dosage.

[0009] In a preferred embodiment of the present invention, the above-mentioned determination of meeting the sludge discharge conditions based on the amount of dry sludge and a preset sludge level threshold includes: determining the total amount of dry sludge in the sedimentation tank through a dry sludge amount model; and determining that the sludge discharge conditions are met when the sludge level of the total amount of dry sludge meets the sludge level threshold, or when the amount of dry sludge accumulates to a preset proportion of the effective volume of the sedimentation tank.

[0010] In a preferred embodiment of the present invention, the above-mentioned sludge discharge includes: opening the sludge discharge valves sequentially according to the first-in-first-out strategy; monitoring the sludge discharge concentration in real time and dynamically adjusting the sludge discharge time based on the sludge discharge concentration; and closing the sludge discharge valves when the sludge discharge concentration drops to a preset concentration threshold or the sludge level drops to a zero threshold.

[0011] In a preferred embodiment of the present invention, the coagulant dosage is dynamically adjusted based on the effluent turbidity using an adaptive mode decomposition algorithm, including: when the effluent turbidity exceeds a preset turbidity threshold, increasing the weight of the turbidity parameter in the adaptive mode decomposition algorithm; and re-determining the coagulant dosage using the adaptive mode decomposition algorithm.

[0012] Secondly, embodiments of the present invention also provide a sedimentation tank flocculation and sedimentation control device, comprising: a raw water quality target parameter acquisition module for acquiring raw water quality target parameters; a feature vector acquisition module for obtaining feature vectors based on the raw water quality target parameters through an adaptive mode decomposition algorithm for feature extraction and noise reduction; a feature vector processing module for processing the feature vectors using an improved genetic algorithm-optimized extreme learning machine model to obtain a predicted value for coagulant dosage; an operating parameter monitoring module for adding coagulant to the raw water based on the predicted value for coagulant dosage and monitoring the operating parameters of the sedimentation tank; a dry sludge quantity determination module for determining the amount of dry sludge in the sedimentation tank based on the operating parameters and a dry sludge quantity model; a sludge discharge module for discharging sludge if the sludge discharge conditions are met based on the dry sludge quantity and a pre-set sludge level threshold; and a coagulant dosage adjustment module for monitoring the effluent turbidity of the sedimentation tank in real time during the sludge discharge process and dynamically adjusting the coagulant dosage based on the effluent turbidity through an adaptive mode decomposition algorithm.

[0013] Thirdly, embodiments of the present invention also provide an electronic device, including a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the sedimentation tank flocculation and sedimentation control method of the first aspect described above.

[0014] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the sedimentation tank flocculation and sedimentation control method of the first aspect described above.

[0015] The embodiments of the present invention bring the following beneficial effects:

[0016] This invention provides a method, apparatus, electronic device, and storage medium for controlling flocculation and sedimentation in a sedimentation tank. The method involves acquiring target parameters of the raw water quality, extracting and denoising features using an adaptive mode decomposition algorithm to obtain feature vectors, and then processing these feature vectors using an optimized extreme learning machine model based on an improved genetic algorithm to obtain a predicted coagulant dosage. Based on this predicted dosage, coagulant is added to the raw water, and the operating parameters of the sedimentation tank are monitored. Based on these parameters, the amount of dry sludge in the sedimentation tank is determined using a dry sludge quantity model. If the dry sludge quantity and a pre-set sludge level threshold indicate that the conditions for sludge discharge are met, sludge is discharged. During the sludge discharge process, the turbidity of the effluent from the sedimentation tank is monitored in real time, and the coagulant dosage is dynamically adjusted based on the effluent turbidity using the adaptive mode decomposition algorithm. This method improves the accuracy of coagulant dosing and process synergy, enabling on-demand sludge discharge from the sedimentation tank.

[0017] Other features and advantages of this disclosure will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above.

[0018] To make the above-mentioned objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 A flowchart of a sedimentation tank flocculation and sedimentation control method provided in an embodiment of the present invention;

[0021] Figure 2 A flowchart of another sedimentation tank flocculation and sedimentation control method provided in an embodiment of the present invention;

[0022] Figure 3 A flowchart of another sedimentation tank flocculation sedimentation control method provided in an embodiment of the present invention;

[0023] Figure 4 This is a schematic diagram of a sedimentation tank flocculation and sedimentation control device provided in an embodiment of the present invention;

[0024] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Coagulation and sedimentation are the core series of processes in water treatment plants for removing suspended solids, colloidal particles and algae from raw water: first, coagulants are added to make impurities in the water form flocs, and then the flocs are separated by sedimentation in a sedimentation tank.

[0027] In traditional water treatment plants, coagulant dosing and sedimentation tank sludge removal are often controlled as independent processes without a synergistic optimization mechanism. Adjustments to coagulant dosage fail to consider the sludge load and sedimentation capacity of the subsequent sedimentation tank, leading to increased sludge accumulation when the dosage is too high. Furthermore, feedback on sedimentation tank sludge removal performance does not provide guidance for optimizing coagulant dosage. When incomplete sludge removal results in excessive effluent turbidity, it is impossible to adjust coagulant dosage parameters in a timely manner to improve floc formation. This disconnect in processes leads to low operating efficiency of the entire coagulation-sedimentation system, making it difficult to achieve the dual goals of meeting water quality standards and controlling costs.

[0028] Based on this, the present invention provides a sedimentation tank flocculation and sedimentation control method, device, electronic equipment, and storage medium. This method involves acquiring target parameters of the raw water quality, extracting and denoising features using an adaptive mode decomposition algorithm to obtain feature vectors based on these parameters, processing the feature vectors using an improved genetic algorithm-optimized extreme learning machine model to obtain a predicted coagulant dosage, adding coagulant to the raw water based on this predicted dosage, and monitoring the sedimentation tank's operating parameters. Based on these operating parameters, the dry sludge volume in the sedimentation tank is determined using a dry sludge volume model. If the dry sludge volume and a pre-set sludge level threshold indicate that the sludge discharge conditions are met, sludge discharge is performed. During the sludge discharge process, the effluent turbidity of the sedimentation tank is monitored in real time, and the coagulant dosage is dynamically adjusted based on the effluent turbidity using an adaptive mode decomposition algorithm. This method improves the accuracy of coagulant dosing and process synergy, enabling on-demand sludge discharge from the sedimentation tank.

[0029] To facilitate understanding of this embodiment, a detailed description of a sedimentation tank flocculation and sedimentation control method disclosed in this embodiment of the invention will be provided first.

[0030] Example 1

[0031] This invention provides a method for controlling flocculation and sedimentation in a sedimentation tank. Figure 1 This is a flowchart illustrating a method for controlling flocculation and sedimentation in a sedimentation tank, as provided in an embodiment of the present invention. Figure 1 As shown, the flocculation and sedimentation control method in this sedimentation tank may include the following steps:

[0032] Step S101: Obtain the target parameters for raw water quality.

[0033] Specifically, obtaining the target parameters for raw water quality may include: collecting raw water quality parameters; normalizing the raw water quality parameters to obtain raw water quality parameter mapping values; and using Pearson correlation analysis based on the raw water quality parameter mapping values ​​to screen out parameters that are significantly correlated with the amount of coagulant added as the target parameters for raw water quality.

[0034] Among them, online sensors can be used to collect parameters of the raw water inlet. Online sensors may include: flow sensors, turbidity sensors, COD sensors, pH sensors, ammonia nitrogen sensors, and water temperature sensors.

[0035] Among them, the flow sensor has a range of 0-5000m³ / h (accuracy ±1%), the turbidity sensor has a range of 0-100NTU (accuracy ±2%), the COD sensor has a range of 0-20mg / L (accuracy ±5%), the pH sensor has a range of 0-14 (accuracy ±0.02), and the ammonia nitrogen and water temperature sensors have a range of 0-50℃ (accuracy ±0.1℃).

[0036] Step S102: Based on the target parameters of raw water quality, feature vectors are obtained by feature extraction and noise reduction using an adaptive mode decomposition algorithm.

[0037] The dimensional differences can be eliminated by max-min normalization, as shown in the following formula:

[0038] Where: x is the original value of the water quality parameter; x min The minimum value of the original water quality parameters; x max is the maximum value of the original water quality parameter; x' is the value that maps the original water quality parameter value to the interval [0, 1].

[0039] Then, through Pearson correlation analysis, seven water quality parameters that are significantly correlated with the coagulant dosage, such as flow rate, turbidity, COD, ammonia nitrogen, pH, and water temperature with a two-sided confidence level ≤0.01, were selected. These six water quality parameters were used as the input to the next model, and the coagulant dosage was used as the output.

[0040] The selected parameters can first be decomposed using a three-layer adaptive modality number optimization algorithm. Each layer decomposes the signal into low-frequency components (reflecting long-term trends) and high-frequency components (reflecting short-term fluctuations). The first layer decomposition retains the high-frequency components, the second layer decomposition retains the high-frequency components, and the third layer decomposition retains the low-frequency components, ultimately yielding four feature components, effectively removing data noise. Then, the improved genetic algorithm is used to optimize the parameters to achieve the training of the ELM model and the prediction of coagulant addition.

[0041] Step S103: The feature vector is processed using an improved genetic algorithm-optimized extreme learning machine model to obtain the predicted value of coagulant dosage.

[0042] Step S104: Add coagulant to the raw water based on the predicted coagulant dosage and monitor the operating parameters of the sedimentation tank.

[0043] One set of raw water data is collected every hour, normalized, and then input into the adaptive modal number optimization algorithm-PSO-BP prediction model module. The model first decomposes and extracts data features through a three-layer adaptive modal number optimization algorithm, and then calculates the predicted value of coagulant dosage using a BP neural network optimized by PSO. The variable frequency metering pump adds coagulant according to the predicted value, and the suspended solids and colloidal particles in the raw water form dense flocs, which enter the sedimentation tank to wait for settling.

[0044] The resulting flocs gradually settle in the sedimentation tank, forming a bottom sludge layer. A sludge level sensor installed at the bottom of the sedimentation tank monitors the sludge accumulation height in real time, an influent turbidity sensor monitors the turbidity (NTU) of the floc-laden water entering the sedimentation tank, an influent flow sensor monitors the influent flow rate (Q), and a sludge concentration detection device is on standby. All data is transmitted to the data analysis layer via the communication layer.

[0045] Step S105: Based on the operating parameters, determine the amount of dry sludge in the sedimentation tank using a dry sludge quantity model.

[0046] Specifically, the data analysis layer substitutes the monitoring data into the dry sludge volume calculation formula:

[0047] TDS total=Q(T×E1+A×E2×K pH +(COD×E3+NH4 + ×E4)C×K org ×10 -6 .

[0048] Where TDS is the total dry sludge volume, t / h; Q is the influent volume, m³ / h. 3 / h; T is the raw water turbidity, NTU; E1 is the conversion factor between turbidity and SS, 0.7~2.2; A is the aluminum salt coagulant injection rate, mg / L; E2 is the conversion factor between Al2O3 and Al(OH)3, taken as 1.53, K pH E1 represents the correction factor, categorized according to the raw water pH. When pH = 6.5~7.5 (optimal reaction range), the value is 1.0; when pH = 5.5~6.5 or 7.5~8.5, the value is 0.8~0.9; when pH < 5.5 or > 8.5, the value is 0.5~0.7. E3 represents the conversion factor for the dry sludge amount of COD organic matter complexes, experimentally calibrated to 0.08~0.12 (for every 1 mg / L increase in COD, the dry sludge amount increases by 0.08~0.12 mg / L); E4 represents the conversion factor for the dry sludge amount of ammonia nitrogen reaction products, experimentally calibrated to 0.05~0.07 (for every 1 mg / L increase in ammonia nitrogen, the dry sludge amount increases by 0.05~0.07 mg / L); K orgThis represents the correction factor for the impact of organic matter. It is divided according to the COD value: when COD < 5 mg / L, the value is 0.9~1.0; when 5 mg / L ≤ COD < 10 mg / L, the value is 0.8~0.9; when COD ≥ 10 mg / L, the value is 0.6~0.8.

[0049] To address the issue of "calculation deviations caused by sensor errors and fluctuations in reagent dosage," the basic dry sludge quantity was adjusted based on actual operational feedback parameters from the sedimentation tank (effluent turbidity and sludge moisture content). Dynamic adjustments are made to form a two-stage model of "theoretical calculation + feedback adjustment", as shown in the following formula:

[0050] ;

[0051] ;

[0052] .

[0053] in, K turb This represents the turbidity correction factor for the effluent. T out Tout,set represents the measured value of turbidity in the sedimentation tank effluent; Tout,set represents the set value of turbidity in the effluent. K turb The value ranges from 0 to 0.1; when T out ≤ T out When setting, K turb =0.

[0054] K moist This represents the sludge moisture content correction factor, with a value range of 0~0.08 (when P...). 1,act ≤P 1,des At that time, K moist =0); P 1,act This indicates the actual moisture content of the sludge (%, collected by an online moisture content sensor); P 1,des This indicates the design moisture content of the sludge (generally between 99.0% and 99.5%).

[0055] The amount of dry sludge in the flocculation sedimentation tank was estimated. TDS final Assuming the sludge moisture content in the flocculation sedimentation tank is P1, and the sludge density is ρ (g / m³),... 3 If the sludge production at the bottom of the sedimentation tank is Q1, the calculation formula is as follows: .

[0056] When the sludge level detection sensor shows that the sludge accumulation height H reaches the upper limit threshold Hset (the threshold depends on the actual situation), or when the sludge production accumulates to 80% of the effective volume of the sedimentation tank, the data analysis layer triggers a sludge discharge command, setting the initial sludge discharge duration to 120s.

[0057] The formula for calculating the sludge accumulation height H is as follows: .

[0058] H: Sludge level height in sedimentation tank (m); t: Accumulation time of dry sludge (h); ρ: Sludge density (g / m³); S: Cross-sectional area of ​​sedimentation tank (m²).

[0059] Sludge discharge trigger condition: when H ≥ Hset (sludge level threshold) or (V) eff When the effective volume of the sedimentation tank is reached, sludge discharge is triggered.

[0060] Step S106: If the conditions for sludge discharge are met based on the dry sludge volume and the preset sludge level threshold, then sludge discharge is performed.

[0061] Specifically, determining whether the sludge discharge conditions are met based on the amount of dry sludge and a pre-set sludge level threshold can include: determining the total amount of dry sludge in the sedimentation tank using a dry sludge volume model; and determining whether the sludge discharge conditions are met when the total amount of dry sludge meets the sludge level threshold or when the amount of dry sludge accumulates to a preset proportion of the effective volume of the sedimentation tank.

[0062] Specifically, sludge removal can include: opening the sludge removal valves sequentially according to a first-in-first-out strategy; monitoring the sludge concentration in real time and dynamically adjusting the sludge removal time based on the sludge concentration; and closing the sludge removal valves when the sludge concentration drops to a preset concentration threshold or the sludge level drops to the zero threshold.

[0063] The control system, upon receiving the sludge discharge command, opens the sludge discharge valve according to the "first-in, first-out" strategy (opening sequentially at 30-second intervals). The sludge concentration detection device monitors the sludge concentration in real time. The initial sludge concentration is pd1, and after 50 seconds, the concentration drops to pd2. The PLC controller shortens the sludge discharge duration to 90 seconds. When the sludge concentration drops to pd3 or the sludge level drops to the zero threshold, the sludge discharge valve is closed to avoid excessive drainage.

[0064] Step S107: During the sludge discharge process, the turbidity of the effluent from the sedimentation tank is monitored in real time, and the amount of coagulant added is dynamically adjusted based on the effluent turbidity using an adaptive mode decomposition algorithm.

[0065] Specifically, dynamically adjusting the coagulant dosage based on effluent turbidity using an adaptive mode decomposition algorithm may include: increasing the weight of the turbidity parameter in the adaptive mode decomposition algorithm when the effluent turbidity exceeds a preset turbidity threshold; and redetermining the coagulant dosage using the adaptive mode decomposition algorithm.

[0066] If the turbidity monitoring value of the sedimentation tank effluent is greater than the preset value (assuming the preset value is 1.0 NTU), the data analysis layer will feed this data back to the coagulant intelligent dosing unit. The adaptive modal number optimization algorithm - PSO-BP prediction model module will automatically increase the weight of the turbidity parameter. The calculation formula for the dynamic adjustment of the weight is as follows: assuming the original turbidity setting... When the effluent turbidity exceeds the preset value of 1.0 NTU, the new weighting is as follows: .

[0067] The coagulant dosage is recalculated, and the coagulant dosing execution module is adjusted according to the new dosage. The turbidity of the effluent from the subsequent sedimentation tank is reduced to below the preset value, thus achieving process synergistic optimization.

[0068] The beneficial effects of this application's embodiments are centered on "solving the disconnect between coagulation and sedimentation processes," constructing a complete closed loop of "data acquisition → feature extraction → dosing prediction → sludge discharge control → feedback optimization." Taking a practical engineering problem as an example, the main advantages of this application's embodiments can be described from the following aspects: First, by using adaptive mode decomposition and improved GA-ELM, the problems of "low accuracy and slow response" in traditional experience-based dosing are solved; second, sludge discharge is triggered by dual conditions of dry sludge volume model and sludge level threshold, combined with dynamic adjustment of sludge discharge concentration duration, avoiding the blindness of "timed sludge discharge"; third, during the sludge discharge process, effluent turbidity is used as a feedback signal to dynamically adjust the parameter weights of adaptive mode decomposition, achieving mutual feedback between "dosing" and "sedimentation," breaking through the limitations of independent control in traditional processes.

[0069] The sedimentation tank flocculation and sedimentation control method provided in this invention can obtain the target parameters of raw water quality, extract features and reduce noise based on these parameters using an adaptive mode decomposition algorithm to obtain a feature vector, process the feature vector using an optimized extreme learning machine model with an improved genetic algorithm to obtain a predicted value for coagulant dosage, add coagulant to the raw water based on the predicted dosage, and monitor the operating parameters of the sedimentation tank. Based on the operating parameters, the dry sludge volume in the sedimentation tank is determined using a dry sludge volume model. If the dry sludge volume and a pre-set sludge level threshold determine that the sludge discharge conditions are met, sludge discharge is performed. During the sludge discharge process, the turbidity of the effluent from the sedimentation tank is monitored in real time, and the coagulant dosage is dynamically adjusted based on the effluent turbidity using an adaptive mode decomposition algorithm. This method improves the accuracy of coagulant dosage and process synergy, and achieves on-demand sludge discharge from the sedimentation tank.

[0070] Example 2

[0071] This invention also provides another method for controlling flocculation and sedimentation in sedimentation tanks; this method is implemented based on the method in the above embodiments; the method focuses on describing the specific implementation of obtaining feature vectors by feature extraction and noise reduction processing based on the target parameters of raw water quality through an adaptive mode decomposition algorithm.

[0072] Figure 2 A flowchart of another sedimentation tank flocculation and sedimentation control method provided in an embodiment of the present invention is shown below. Figure 2 As shown, the feature vector obtained by feature extraction and noise reduction based on the raw water quality target parameters through an adaptive mode decomposition algorithm may include the following steps:

[0073] Step S201: Set the search range for the number of modes and perform multimodal decomposition on the time series data of water quality parameters.

[0074] First, the search range of the modality number K is set: K∈[k1,k2]. The search range is preset with a reasonable interval based on the characteristics of water quality parameters, and the value range of each water quality parameter is determined according to the actual situation.

[0075] The basic parameters of the adaptive modality number optimization algorithm are set as follows: penalty parameter α=180, balancing reconstruction error and bandwidth constraint; convergence criterion ε=10. -7 Control the iteration termination condition; the maximum number of iterations T=1000; input water quality parameter time series data x(t), flow rate, turbidity, COD, ammonia nitrogen, pH, and water temperature.

[0076] Step S202 involves iteratively updating the modal components, center frequency, and Lagrange multiplier until the convergence condition is met.

[0077] For each candidate K value (K=2, 3, ..., 10), the following decomposition is performed:

[0078] Step A1, set K modal components u k and its center frequency ω k Initialize the Lagrange multiplier λ and the iteration counter n=0.

[0079] Step A2: Perform iterative updates.

[0080] Step A21: Update the modal components by solving a variational problem in the frequency domain. The formula is: ;in, Fourier transform of the k-th modal component in the (n+1)-th iteration; : Represents the original Fourier transform of the raw water quality time series; The sum of the Fourier transforms of all modal components except the k-th component; : Fourier transform of Lagrange multipliers (used for constraint optimization); ω: angular frequency (frequency domain variable); ωk: center frequency of the k-th modal component.

[0081] Step A22: Update the center frequency using the following formula: ;in, The center frequency of the k-th modal component in the (n+1)-th iteration; : Modulus square of the Fourier transform of the k-th modal component (energy spectral density); Numerator: Frequency-weighted energy integral (reflecting the frequency position of energy concentration); Denominator: Total energy integral (normalized).

[0082] Step A23: Update the Lagrange multipliers using the following formula: Where τ is the update step size (taken as 0.001); : The Lagrange multiplier Fourier transform of the (n+1)th iteration; Reconstruction error: The difference between the original signal and the sum of all modal components.

[0083] Step A24, perform convergence judgment: if the following conditions are met... Then stop the iteration and output K modal components {u1, u2, ..., u}. K}; otherwise, n = n + 1, return to the previous step and continue iterating and updating. Wherein, : The squared L2 norm (difference) of the two iteration results of the kth modal component; : The squared L2 norm of the current iteration result of the k-th modal component (energy baseline).

[0084] Step S203: Construct a comprehensive evaluation index based on the average sample entropy and average kurtosis, and select the optimal number of modes.

[0085] Specifically, regarding the calculation of average sample entropy:

[0086] Construct an m-dimensional embedding vector (m=2, adapted to the short-term correlation features of water quality time series data);

[0087] Xi=[uk(i),uk(i+1),...,uk(i+m 1) (i=1,2,...,N) m+1), where N: the number of samples of modal component uk, corresponding to the frequency of raw water data collection (e.g., 1 set per hour).

[0088] Calculate the distance between vectors: d[Xi, Xj] = maxl = 1, 2, ..., m |Xi(l) Xj(l)∣, where Xi(l): the l-th element of the i-th embedding vector, reflecting the value of the vector at the l-th time node.

[0089] Calculate the similarity vector ratio: (r=0.2×std( ), Θ is the step function.

[0090] Calculate the sample entropy of a single modal component: .

[0091] Calculate the average sample entropy: .

[0092] Where, (Xi): the i-th m-dimensional embedding vector (composed of m consecutive data points of the modal components); m: embedding dimension (taken as 2 in this algorithm to reflect the local correlation of the signal); r: similarity threshold (taken as 0.2×std( (std is the standard deviation). Step function (1 if the value in parentheses is ≥0, otherwise 0); : The proportion of vectors similar to Xi (reflecting the regularity of the signal); SEk: The sample entropy of the k-th modal component (the smaller the value, the more regular the signal); SEavg: The average sample entropy of all modal components (an indicator of overall regularity). : The maximum distance between vectors Xi and Xj (measures similarity).

[0093] Specifically, regarding the calculation of average kurtosis:

[0094] Kurtosis of the k-th modal component: (μ is the mean, E is the expected value).

[0095] Average kurtosis:

[0096] Where, Kurtk: the kurtosis of the k-th modal component (measures the degree to which the signal deviates from the normal distribution; a larger value indicates a more significant peak); : The fourth power of the difference between the kth modal component and the mean (fourth central moment). : Variance of the k-th modal component (second-order central moment); u: Mean of the k-th modal component; Kurtavg: Average kurtosis of all modal components (overall peak characteristic index).

[0097] Specifically, regarding the construction of comprehensive evaluation indicators:

[0098] F(K) = 0.3 × SE avg 0.7×Kurt avg .

[0099] In this context, a smaller SE indicates a more stable feature, while a larger Kurt indicates a greater ability to capture mutations; hence, the signs are opposite.

[0100] Specifically, regarding the selection of the optimal number of modes:

[0101] Optimal number of modes Kopt: Iterate through K=2~10 and select the K value that minimizes F(K) to ensure that the decomposed modal components have both low noise and strong mutation capture capability.

[0102] Step S204: Retain the key modal components and concatenate them in order to form a feature vector, and then normalize the feature vector.

[0103] Specifically, regarding eigenvector recombination:

[0104] The first four key components out of the Kopt modal components are retained. Contribution analysis shows that the first four components contain more than 95% of the energy and can represent the core features of the original data. They are concatenated in the order of "low-frequency trend component → high-frequency fluctuation component" to form a 28-dimensional feature vector. Each component has 7 time-series nodes, for a total of 28 dimensions.

[0105] The steps for calculating the energy percentage are as follows:

[0106] For each modal component u K Calculate its energy Total energy Through calculation .

[0107] The 28 eigenvectors are normalized using the following formula: Where x: the original value of a certain element in the 28-dimensional feature vector; x min x max : The minimum and maximum values ​​of this element in the historical dataset; : The feature values ​​are normalized and mapped to the interval [0, 1] to avoid the impact of dimensional differences on subsequent model training.

[0108] Example 3

[0109] This invention also provides another method for controlling flocculation and sedimentation in sedimentation tanks; this method is implemented based on the method in the above embodiments; the method focuses on describing the specific implementation of using an improved genetic algorithm-optimized extreme learning machine model to process feature vectors and obtain the predicted value of coagulant dosage.

[0110] Figure 3 A flowchart of another sedimentation tank flocculation and sedimentation control method provided in an embodiment of the present invention is shown below. Figure 3 As shown, the process of using an optimized extreme learning machine model based on an improved genetic algorithm to process feature vectors and obtain predicted values ​​for coagulant dosage can include the following steps:

[0111] Step S301: Define the architecture of the Extreme Learning Machine model.

[0112] Specifically, the Extreme Learning Machine model includes an input layer, a hidden layer, and an output layer.

[0113] Input layer: 28 nodes, corresponding to 28-dimensional normalized feature vectors, which are matched with the output of the adaptive modality number optimization algorithm.

[0114] Hidden layer: L nodes, parameters to be optimized, value range [10, 50], activation function is Sigmoid function. It adapts to the nonlinear mapping requirements of water quality data.

[0115] Output layer: 1 node, coagulant dosage A: mg / L, which is the core prediction target of the model;

[0116] Parameters to be optimized: There are a total of 28L+L+1=29L+1 parameters, including: (1) Input layer → Hidden layer weight matrix: (28L parameters, controlling the mapping of input features to the hidden layer); (2) Hidden layer bias vector: (L parameters, adjusting the output threshold of hidden layer nodes); (3) Hidden layer → Output layer weight vector: (L parameters control the mapping of hidden layer features to the output layer, which are randomly initialized in the original ELM algorithm.)

[0117] Step S302: Set the parameters of the improved genetic algorithm.

[0118] Specifically, the population size is Npop=80 (each individual corresponds to one set of ELM parameters, the size is adapted to the parameter dimension, and the breadth of optimization is ensured).

[0119] Maximum number of iterations: Gmax=100 (balancing optimization accuracy and computational efficiency to meet the real-time control requirements of water supply plants).

[0120] Elite individual ratio: 5% (i.e., 4 elite individuals are directly retained to the next generation to avoid the loss of excellent parameter combinations and solve the problem of premature convergence in traditional GA).

[0121] Adaptive crossover probability (dynamically adjusted based on individual fitness to preserve superior genes):

[0122] .

[0123] Where P cmax =0.9, P cmin =0.4 (crossover probability range, ensuring a balance between population diversity and convergence speed).

[0124] f: Individual fitness value, favg: Population average fitness value, fmax: Population maximum fitness value. Among these, the smaller the fitness value, the better the parameter combination.

[0125] Adaptive mutation probability (dynamically adjusted based on individual fitness to increase the probability of improvement in inferior individuals):

[0126] .

[0127] Pm max =0.1, Pm min =0.001 (mutation probability range, to avoid excessive mutations that could damage high-quality genes).

[0128] The formula for calculating the mean squared error (MSE), which measures the deviation between the predicted and actual values, is as follows: Where n: sample size (training set n1=128, accounting for 70% of historical data; test set n2=55, accounting for 30% of historical data); yi is the actual value of coagulant dosage. is the predicted value of the BP neural network, and n is the number of samples.

[0129] Fitness function (combining training and test set errors to avoid model overfitting): MSE train Mean Squared Error of Training Set (MSE) test : Mean squared error of the test set; the smaller the fitness function value, the higher the prediction accuracy and the stronger the generalization ability of the ELM parameter combination.

[0130] Step S303: Globally optimize the weights and bias parameters of the extreme learning machine model by improving the genetic algorithm.

[0131] Step B1, Population Initialization:

[0132] The range of values ​​for each individual parameter is as follows: input layer → hidden layer weight W∈[-3, 3], hidden layer bias b∈[-3, 3], hidden layer → output layer weight β∈[-3, 3] (adapting to the characteristics of ELM parameter values ​​to ensure reasonable mapping relationships); 80 individuals are randomly generated, each encoded as a 29L+1 dimensional real vector L∈[10, 50], determined according to the number of hidden layer nodes), corresponding to a complete set of ELM parameters.

[0133] Step B2, Fitness Calculation:

[0134] Individual decoding: Decode the real number vector of each individual into the W, b, and β parameters of the ELM.

[0135] ELM prediction: Input the 28-dimensional normalized feature vector output from step S3-1 into the ELM, and calculate the hidden layer output H and the predicted output. .

[0136] Hidden layer output: H = σ(XW + b×1) n×1 )

[0137] Where, X∈R n×28 Let W be the input feature matrix, and W ∈ R. 28×L (L is the number of hidden layer nodes); b∈R 1×L (Hidden layer bias), requires 1 n×1 (Including 1 vector) is expanded to n×L dimensions; σ(·) is the Sigmoid activation function.

[0138] Predicted output: =Hβ.

[0139] Error calculation: Substitute the MSE formula to calculate the error between the training set and the test set, and then obtain the f value of each individual through the fitness function.

[0140] Step B3, Elite Retention: Sort all individuals in the population in ascending order of f value, select the top 4 elite individuals (with the smallest f value), and directly copy them to the next generation of the population to ensure that high-quality parameter combinations are not lost.

[0141] Step B4, selection operation (roulette wheel selection method, to increase the probability of high-quality individuals being inherited).

[0142] Probability of an individual being selected: (The smaller fi is, the larger pi is, and the higher the probability that a high-quality individual will be selected to participate in reproduction).

[0143] Based on probability pi, 76 individuals are randomly selected from the non-elite individuals (76 individuals) and combined with the elite individuals (4 individuals) to form a new generation population (80 individuals in total).

[0144] Step B5, cross operation (arithmetic cross, to achieve parameter combination optimization).

[0145] Non-elite individuals in the new generation population randomly select crossover pairs (i, j) according to the adaptive crossover probability \(P_c\).

[0146] Generate new individuals: .

[0147] x i x j : The parameter vector of the cross pair The new individual parameter vector generated after crossover ensures the diversity and rationality of parameter combinations.

[0148] Step B6, mutation operation (Gaussian mutation, to avoid the population getting trapped in local optima).

[0149] For non-elite individuals after crossover, the mutation position d is randomly selected according to the adaptive mutation probability Pm; parameter update: xm(d) = x(d) + N(0, 0.1).

[0150] x(d): The parameter value at position d in the individual before the mutation.

[0151] N(0, 0.1): A random number that follows a normal distribution with a mean of 0 and a standard deviation of 0.1, ensuring that the variation range is moderate, avoiding the destruction of high-quality parameters, and introducing new gene diversity.

[0152] Boundary handling: If the mutated parameter xm(d) exceeds the range of [-3, 3], it is forcibly pulled back to the boundary value (i.e., xm(d) = -3 or xm(d) = 3) to ensure that the ELM parameter value is reasonable.

[0153] Step B7, convergence judgment.

[0154] If the current iteration number (G=Gmax) (i.e., reaching 100 generations), or the change in the minimum fitness value of the population over 10 consecutive generations is less than 10. -7 If the iteration stops, the globally optimal individual is output.

[0155] After decoding the globally optimal individual, the optimal parameter combination W* (optimal weight matrix from input layer to hidden layer, optimal bias vector from hidden layer, optimal weight vector from hidden layer to output layer) is obtained and used for subsequent ELM model training and prediction.

[0156] Step S304: Train the extreme learning machine model based on the optimized parameters.

[0157] The input data consists of 28-dimensional normalized feature vectors (assuming a total of 128 training sets are required).

[0158] Training process: Substitute the optimal parameters: Substitute the improved GA outputs W*, b*, β* into the ELM model.

[0159] Calculate the hidden layer output: Based on the Sigmoid activation function, calculate the hidden layer output matrix H* using the input feature matrix and optimal weights and biases. .

[0160] Among them, X train ∈R128×28: Input feature matrix of the training set; σ(): Sigmoid activation function, This enables nonlinear mapping of input features.

[0161] Verify training error: Calculate the predicted output on the training set. Compared with the true value y train The error is used to verify whether the model meets the accuracy requirements: ; .

[0162] If Error trainIf the error is less than 0.0001, based on historical operation and maintenance data, the corresponding coagulant dosage error is ≤0.1mg / L, and the effluent turbidity is <1.0NTU, then the training is complete; otherwise, return to the previous step and re-execute the improved genetic algorithm for optimization until the error requirement is met.

[0163] Step S305: Input the feature vector into the trained extreme learning machine model to obtain the predicted value of coagulant dosage.

[0164] Based on the 28-dimensional normalized feature vectors of the above test set (55 groups in total, accounting for 30% of historical data), calculate the hidden layer output of the test set: ; where X test ∈R55×28: Input feature matrix for the test set.

[0165] Output prediction results: ;in, : Normalized predicted value of coagulant dosage (mapped to the interval [0, 1]);

[0166] Deprecated values ​​denormalized (restored to the scale of actual dosage): Among them, A max A min : Maximum and minimum dosage of coagulant (unit: mg / L); The final predicted value of coagulant dosage is directly connected to the variable frequency metering pump execution module of the original step Y1 (coagulant dosage control).

[0167] Model overall performance evaluation:

[0168] Input the feature vectors of the adaptive modality number optimization algorithm from the test set into the trained model, and output the predicted value of coagulant dosage. Evaluate the model performance using the MAE, RMSE, MAPE, and R2 metrics in the file. Calculation formula: , , , .in, This represents the average of the actual amounts of coagulant added.

[0169] Example 4

[0170] Corresponding to the above method embodiments, this invention provides a sedimentation tank flocculation and sedimentation control device. Figure 4 This is a schematic diagram of a sedimentation tank flocculation and sedimentation control device provided in an embodiment of the present invention, as shown below. Figure 4 As shown, the sedimentation tank flocculation and sedimentation control device may include:

[0171] The raw water quality target parameter acquisition module 401 is used to acquire the raw water quality target parameters.

[0172] The feature vector acquisition module 402 is used to obtain feature vectors by performing feature extraction and noise reduction processing based on the target parameters of raw water quality using an adaptive mode decomposition algorithm.

[0173] The feature vector processing module 403 is used to process the feature vector using an improved genetic algorithm-optimized extreme learning machine model to obtain the predicted value of coagulant dosage.

[0174] The operating parameter monitoring module 404 is used to add coagulant to the raw water based on the predicted coagulant dosage and to monitor the operating parameters of the sedimentation tank.

[0175] The dry sludge quantity determination module 405 is used to determine the amount of dry sludge in the sedimentation tank based on the operating parameters and the dry sludge quantity model.

[0176] The sludge discharge module 406 is used to discharge sludge if the sludge discharge conditions are met based on the dry sludge volume and a pre-set sludge level threshold.

[0177] The coagulant dosage adjustment module 407 is used to monitor the turbidity of the sedimentation tank effluent in real time during the sludge discharge process, and dynamically adjust the coagulant dosage based on the effluent turbidity using an adaptive mode decomposition algorithm.

[0178] The sedimentation tank flocculation and sedimentation control device provided in this embodiment of the invention can obtain target parameters of raw water quality, extract features and reduce noise based on these parameters using an adaptive mode decomposition algorithm to obtain feature vectors, process these feature vectors using an optimized extreme learning machine model with an improved genetic algorithm to obtain predicted values ​​for coagulant dosage, add coagulant to the raw water based on these predicted values, and monitor the operating parameters of the sedimentation tank. Based on these operating parameters, the amount of dry sludge in the sedimentation tank is determined using a dry sludge quantity model. If the dry sludge quantity and a pre-set sludge level threshold indicate that the sludge discharge conditions are met, sludge discharge is performed. During the sludge discharge process, the turbidity of the effluent from the sedimentation tank is monitored in real time, and the coagulant dosage is dynamically adjusted based on the effluent turbidity using an adaptive mode decomposition algorithm. This method improves the accuracy of coagulant dosage and process synergy, and enables on-demand sludge discharge from the sedimentation tank.

[0179] In some embodiments, the raw water quality target parameter acquisition module is further used to collect raw water quality parameters; normalize the raw water quality parameters to obtain raw water quality parameter mapping values; and use Pearson correlation analysis to screen out parameters that are significantly correlated with the amount of coagulant added as raw water quality target parameters based on the raw water quality parameter mapping values.

[0180] In some embodiments, the feature vector acquisition module is further configured to set the search range of the number of modes and perform multimodal decomposition on the time series data of water quality parameters; iteratively update the modal components, center frequencies and Lagrange multipliers until the convergence condition is met; construct a comprehensive evaluation index based on the average sample entropy and average kurtosis, select the optimal number of modes; retain the key modal components and splice them in order to form a feature vector, and normalize the feature vector.

[0181] In some embodiments, the feature vector processing module is further configured to define the architecture of the extreme learning machine model; set the parameters of the improved genetic algorithm; globally optimize the weights and bias parameters of the extreme learning machine model using the improved genetic algorithm; train the extreme learning machine model based on the optimized parameters; and input the feature vectors into the trained extreme learning machine model to obtain the predicted value of the coagulant dosage.

[0182] In some embodiments, the sludge discharge module is further configured to determine the total amount of dry sludge in the sedimentation tank through a dry sludge volume model; when the sludge level of the total dry sludge volume meets the sludge level threshold, or when the dry sludge volume accumulates to a preset proportion of the effective volume of the sedimentation tank, it is determined that the sludge discharge conditions are met.

[0183] In some embodiments, the sludge discharge module is further configured to sequentially open the sludge discharge valves according to a first-in-first-out strategy; monitor the sludge discharge concentration in real time and dynamically adjust the sludge discharge time based on the sludge discharge concentration; and close the sludge discharge valves when the sludge discharge concentration drops to a preset concentration threshold or the sludge level drops to a zero threshold.

[0184] In some embodiments, the coagulant dosage adjustment module is further configured to increase the weight of the turbidity parameter in the adaptive mode decomposition algorithm when the effluent turbidity exceeds a preset turbidity threshold; and redetermine the coagulant dosage through the adaptive mode decomposition algorithm.

[0185] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0186] Example 5

[0187] This invention also provides an electronic device for operating the above-described sedimentation tank flocculation and sedimentation control method; see [link to previous document]. Figure 5 The diagram shows the structure of an electronic device, which includes a memory 500 and a processor 501. The memory 500 stores one or more computer instructions, which are executed by the processor 501 to implement the above-mentioned flocculation and sedimentation control method in the sedimentation tank.

[0188] Furthermore, Figure 5The electronic device shown also includes a bus 502 and a communication interface 503. The processor 501, the communication interface 503 and the memory 500 are connected via the bus 502.

[0189] The memory 500 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 503 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 502 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0190] Processor 501 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 501 or by instructions in software form. Processor 501 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a readily available storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 500, and processor 501 reads information from memory 500 and, in conjunction with its hardware, completes the steps of the method described in the foregoing embodiments.

[0191] This invention also provides a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are called and executed by a processor, they cause the processor to implement the above-described sedimentation tank flocculation and sedimentation control method. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0192] The computer program product for controlling flocculation and sedimentation in a sedimentation tank provided in this embodiment of the invention includes a computer-readable storage medium storing non-volatile program code executable by a processor. The instructions included in the program code can be used to execute the methods in the preceding method embodiments. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0193] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0194] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0195] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0196] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0197] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0198] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for controlling flocculation and sedimentation in a sedimentation tank, characterized in that, The method includes: Obtain the target parameters for raw water quality; Based on the raw water quality target parameters, feature vectors are obtained by feature extraction and noise reduction using an adaptive mode decomposition algorithm; The feature vector is processed using an optimized extreme learning machine model based on an improved genetic algorithm to obtain the predicted value of coagulant dosage; Coagulant is added to the raw water based on the predicted coagulant dosage, and the operating parameters of the sedimentation tank are monitored. Based on the operating parameters, the amount of dry sludge in the sedimentation tank is determined using a dry sludge volume model. If the conditions for sludge discharge are met based on the amount of dry sludge and the preset sludge level threshold, then sludge discharge is performed. During the sludge discharge process, the turbidity of the effluent from the sedimentation tank is monitored in real time, and the amount of coagulant added is dynamically adjusted based on the effluent turbidity using an adaptive mode decomposition algorithm. The formula for calculating the dry mud quantity is: TDS total=Q[T×E1+A×E2×K] pH +(COD×E3+NH4 + ×E4)C×K org ]×10 -6 Where TDS is the total dry sludge volume, t / h; Q is the influent volume, m³ / h. 3 / h; T is the raw water turbidity, NTU; E1 is the conversion factor between turbidity and SS; A is the aluminum salt coagulant injection rate, mg / L; E2 is the conversion factor between Al2O3 and Al(OH)3, K pH E3 represents the conversion factor for the dry sludge amount of COD organic matter complexes; E4 represents the conversion factor for the dry sludge amount of ammonia nitrogen reaction products; K represents the correction factor. org This indicates the correction factor for the influence of organic matter; For the amount of dry mud in the foundation The formula for dynamic correction is: ;in, , K turb This represents the turbidity correction factor for the effluent. T out T represents the measured value of turbidity in the effluent from the sedimentation tank. out,set Indicates the set value for effluent turbidity; K moist P represents the sludge moisture content correction factor. 1,act P represents the actual moisture content of the sludge. 1,des This indicates the design moisture content of the sludge.

2. The method according to claim 1, characterized in that, The acquisition of raw water quality target parameters includes: Collect raw water quality parameters; The raw water quality parameters are normalized to obtain the raw water quality parameter mapping values; Based on the raw water quality parameter mapping values, Pearson correlation analysis was used to screen out parameters that were significantly correlated with the coagulant dosage as raw water quality target parameters.

3. The method according to claim 1, characterized in that, The feature vector obtained by feature extraction and noise reduction based on the raw water quality target parameters using an adaptive mode decomposition algorithm includes: Define the search range for the number of modes and perform multimodal decomposition on the time series data of water quality parameters; The modal components, center frequency, and Lagrange multiplier are updated iteratively until the convergence condition is met. A comprehensive evaluation index is constructed based on average sample entropy and average kurtosis, and the optimal number of modes is selected. Key modal components are retained and concatenated in sequence to form feature vectors, and the feature vectors are then normalized.

4. The method according to claim 1, characterized in that, The process of using an optimized extreme learning machine model based on an improved genetic algorithm to process the feature vector to obtain the predicted coagulant dosage includes: Define the architecture of the Extreme Learning Machine model; Set the parameters for the improved genetic algorithm; The weights and bias parameters of the extreme learning machine model are globally optimized by improving the genetic algorithm; The extreme learning machine model is trained based on the optimized parameters; The feature vector is input into the trained extreme learning machine model to obtain the predicted value of coagulant dosage.

5. The method according to claim 1, characterized in that, The determination of whether the sludge discharge conditions are met based on the amount of dry sludge and a pre-set sludge level threshold includes: The total amount of dry sludge in the sedimentation tank was determined using a dry sludge volume model. When the total dry sludge volume meets the sludge level threshold, or when the dry sludge volume accumulates to a preset proportion of the effective volume of the sedimentation tank, the sludge discharge conditions are determined to be met.

6. The method according to claim 1, characterized in that, The process of removing sludge includes: Open the mud discharge valves sequentially according to the first-in-first-out strategy; The sludge concentration is monitored in real time, and the sludge discharge time is dynamically adjusted based on the sludge concentration. When the sludge concentration decreases to a preset concentration threshold or the sludge level drops to the zero threshold, the sludge discharge valve is closed.

7. The method according to claim 1, characterized in that, The step of dynamically adjusting the coagulant dosage based on the effluent turbidity using an adaptive mode decomposition algorithm includes: When the turbidity of the effluent exceeds the preset turbidity threshold, the weight of the turbidity parameter in the adaptive mode decomposition algorithm is increased. The dosage of coagulant is re-determined using the adaptive mode decomposition algorithm.

8. A flocculation and sedimentation control device for a sedimentation tank, characterized in that, The apparatus for implementing the sedimentation tank flocculation and sedimentation control method according to any one of claims 1 to 7, the apparatus comprising: The raw water quality target parameter acquisition module is used to acquire raw water quality target parameters; The feature vector acquisition module is used to obtain feature vectors by performing feature extraction and noise reduction processing based on the target parameters of the raw water quality using an adaptive mode decomposition algorithm. The feature vector processing module is used to process the feature vector using an improved genetic algorithm-optimized extreme learning machine model to obtain the predicted value of coagulant dosage. The operating parameter monitoring module is used to add coagulant to the raw water based on the predicted coagulant dosage and monitor the operating parameters of the sedimentation tank. The dry sludge quantity determination module is used to determine the quantity of dry sludge in the sedimentation tank based on the operating parameters and a dry sludge quantity model. The sludge discharge module is used to discharge sludge if it is determined, based on the amount of dry sludge and a preset sludge level threshold, that the sludge discharge conditions are met. The coagulant dosage adjustment module is used to monitor the turbidity of the effluent from the sedimentation tank in real time during the sludge discharge process, and dynamically adjust the coagulant dosage based on the effluent turbidity using an adaptive mode decomposition algorithm.

9. An electronic device, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions that can be executed by the processor, the processor executing the computer-executable instructions to implement the sedimentation tank flocculation and sedimentation control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the sedimentation tank flocculation and sedimentation control method according to any one of claims 1 to 7.

Images