Big data optimized fiber dyeing wastewater treatment and recovery method and system

Abstract

The invention provides a fiber dyeing wastewater treatment and recovery method and system optimized by big data, and relates to the technical field of wastewater treatment.The method comprises the steps that a plurality of wastewater treatment and recovery branches are arranged; performing recoverable sludge resource activity matching analysis on the current fiber dyeing wastewater in the wastewater treatment tank according to a plurality of preset activity intervals to obtain a matched wastewater treatment recovery branch; and acquiring a plurality of treatment process nodes and a plurality of groups of treatment control parameters, performing activity constraint optimization on the plurality of groups of treatment control parameters according to the matched activity-control parameter constraint, outputting a plurality of groups of optimized treatment control parameters, and matching the wastewater treatment and recovery branch to execute fiber dyeing wastewater treatment and recovery operation. The technical problem that in the prior art, due to the fact that the water quality of the fiber dyeing wastewater fluctuates violently and a traditional treatment process is poor in adaptability, the wastewater treatment and recovery efficiency is low is solved, and the wastewater treatment and recovery efficiency is improved by constructing a plurality of wastewater treatment and recovery branches.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wastewater treatment, in particular to a fiber dyeing wastewater treatment and recovery method and system optimized by big data. BACKGROUND

[0002] The fixed combination process of pretreatment, biochemical treatment and advanced treatment is generally used in the field of fiber dyeing wastewater treatment. Due to the great water quality fluctuation of fiber dyeing wastewater, it usually contains various dyes, chemical additives, salts and other complex components. The water quality and pollutant activity of the wastewater fluctuate sharply. The traditional fixed process flow and operating parameters are difficult to dynamically adapt to such changes. Since the traditional treatment process usually uses fixed or semi-fixed process parameters, the anti-shock load capacity is weak, the running state is unstable, and the problems of unqualified treatment effect or out-of-control operation cost often occur. Especially in the face of high-concentration and difficult-to-degrade wastewater, long-term operation under non-optimal conditions often has to maintain excessive reagent addition and energy consumption to ensure that the effluent meets the standard in the most unfavorable case, resulting in a significant decrease in wastewater treatment and recovery efficiency.

[0003] In summary, the existing technology has the technical problem of low wastewater treatment and recovery efficiency due to the sharp fluctuation of fiber dyeing wastewater quality and the poor adaptability of traditional treatment process. SUMMARY

[0004] The purpose of the present application is to provide a fiber dyeing wastewater treatment and recovery method and system optimized by big data, to solve the technical problem of low wastewater treatment and recovery efficiency due to the sharp fluctuation of fiber dyeing wastewater quality and the poor adaptability of traditional treatment process in the prior art.

[0005] In order to achieve the above-mentioned purpose, the present application provides a fiber dyeing wastewater treatment and recovery method and system optimized by big data.

[0006] In a first aspect, the present application provides a fiber dyeing wastewater treatment and recovery method optimized by big data, which is realized by a fiber dyeing wastewater treatment and recovery system optimized by big data, wherein the fiber dyeing wastewater treatment and recovery method optimized by big data comprises: setting a plurality of wastewater treatment and recovery branches, wherein the plurality of wastewater treatment and recovery branches comprise a plurality of preset activity intervals and a plurality of activity-control parameter constraints in a mapping configuration; performing a recoverable sludge resource activity matching analysis on current fiber dyeing wastewater in a wastewater treatment tank according to the plurality of preset activity intervals to obtain a matching wastewater treatment and recovery branch; wherein a plurality of treatment process nodes and a plurality of sets of treatment control parameters of the matching wastewater treatment and recovery branch are obtained, the plurality of sets of treatment control parameters are subjected to activity constraint optimization according to the matched activity-control parameter constraints, a plurality of sets of optimized treatment control parameters are output, and the matching wastewater treatment and recovery branch performs fiber dyeing wastewater treatment and recovery operations by using the plurality of sets of optimized treatment control parameters.

[0007] Optionally, the plurality of activity-control parameter constraints correspond to the plurality of preset activity intervals, and each activity-control parameter constraint comprises a control parameter constraint corresponding to each treatment process node; wherein an activity influence control parameter of each treatment process node is identified, and a control parameter constraint of each treatment process node is constructed based on an activity influence threshold of the activity influence control parameter.

[0008] Optionally, a recovery key component of a recoverable sludge resource of the current fiber dyeing wastewater is extracted; the wastewater treatment tank comprises a multi-parameter sensor array, a plurality of sets of multi-parameter real-time sensing data of the recovery key component are identified by the multi-parameter sensor array respectively; a recovery branch matching model is constructed according to the plurality of preset activity intervals, an activity absorption feature vector of the plurality of sets of multi-parameter real-time sensing data is extracted by using the recovery branch matching model, the activity absorption feature vector is analyzed, and a plurality of recovery branch matching probabilities are obtained; and a matching wastewater treatment and recovery branch corresponding to a first recovery branch matching probability is selected from the plurality of wastewater treatment and recovery branches according to the plurality of recovery branch matching probabilities.

[0009] Optionally, a plurality of sets of multi-parameter real-time sensing data of the recovery key component are identified by the multi-parameter sensor array respectively, and the plurality of sets of multi-parameter real-time sensing data at least comprise a dye-related sensing data set, a salt-related sensing data set, an organic matter-related sensing data set, a heavy metal-related sensing data set, and a microorganism activity sensing data set; the plurality of sets of multi-parameter real-time sensing data are subjected to standardization processing, and a plurality of sets of processed multi-parameter real-time sensing data are output.

[0010] Optionally, the multiple preset activity intervals are interval vector quantization coded to obtain an activity interval vector set; a recyclable sludge resource sample of the fiber dyeing wastewater is collected; an activity absorption feature vector is extracted from the recyclable sludge resource sample, and an activity absorption feature vector sample is output, the activity absorption feature vector including a known activity interval category; a recycling branch matching model is similarity fitting trained according to the activity absorption feature vector sample, the known activity interval category is used as a supervised label, until a label prediction accuracy rate of a recycling branch matching probability sample output by the recycling branch matching model is greater than a preset threshold, and the recycling branch matching model is downloaded.

[0011] Optionally, the activity attenuation coefficient is introduced to perform time sequence attenuation analysis on the multiple recycling branch matching probabilities, to obtain multiple recycling branch matching probability curves; and the switching time sequence nodes of the multiple recycling branch matching probability curves are analyzed, and the corresponding matching wastewater treatment recycling branch is obtained again according to the switching time sequence nodes.

[0012] Optionally, the switching time sequence node is a time sequence node representing a change of the recycling branch corresponding to the first recycling branch matching probability in the multiple recycling branch matching probability curves.

[0013] Optionally, the activity influence control parameter and the non-activity influence control parameter in the multiple groups of treatment control parameters are identified; the activity influence control parameter is optimized according to the matched activity-control parameter constraint, and the optimized activity influence control parameter is output; the non-activity influence control parameter is cooperatively optimized according to the optimized activity influence control parameter, and the optimized non-activity influence control parameter is output; and the multiple groups of optimized treatment control parameters are output according to the optimized activity influence control parameter and the optimized non-activity influence control parameter.

[0014] Optionally, a total amount of the current fiber dyeing wastewater and a wastewater treatment component of the matching wastewater treatment recycling branch are obtained; a treatment time length is predicted according to the wastewater total amount and the wastewater treatment component, and a predicted treatment time length is obtained; and the activity attenuation coefficient is introduced to perform time sequence attenuation analysis on the multiple recycling branch matching probabilities according to the predicted treatment time length.

[0015] Secondly, this application also provides a fiber dyeing wastewater treatment and recycling system optimized using big data, used to execute the fiber dyeing wastewater treatment and recycling method optimized using big data as described in the first aspect. The system includes: a branch setting module for setting multiple wastewater treatment and recycling branches, each branch including multiple preset activity ranges and multiple activity-control parameter constraints; an activity matching analysis module for performing recyclable sludge resource activity matching analysis on the current fiber dyeing wastewater in the wastewater treatment tank based on the multiple preset activity ranges, to obtain a matching wastewater treatment and recycling branch; and an activity constraint optimization module for acquiring multiple process nodes and multiple sets of treatment control parameters of the matching wastewater treatment and recycling branch, optimizing the multiple sets of treatment control parameters according to the matched activity-control parameter constraints, and outputting multiple sets of optimized treatment control parameters. The matching wastewater treatment and recycling branch performs fiber dyeing wastewater treatment and recycling operations using the multiple sets of optimized treatment control parameters.

[0016] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0017] By setting up multiple wastewater treatment and recycling branches, each branch includes multiple preset activity ranges and multiple activity-control parameter constraints. Based on these preset activity ranges, a recyclable sludge resource activity matching analysis is performed on the current fiber dyeing wastewater in the wastewater treatment tank to obtain a matching wastewater treatment and recycling branch. Specifically, multiple treatment process nodes and multiple sets of treatment control parameters are acquired for each matching wastewater treatment and recycling branch. These parameters are then optimized according to the matched activity-control parameter constraints, resulting in multiple sets of optimized treatment control parameters. The matching wastewater treatment and recycling branch then performs the fiber dyeing wastewater treatment and recycling operation based on these optimized treatment control parameters. In other words, by constructing multiple wastewater treatment and recycling branches, combining multiple preset activity ranges and multiple activity-control parameter constraints, and selecting the most suitable treatment path based on real-time changes in water quality in the wastewater treatment tank, the treatment control parameters are dynamically optimized, thereby improving wastewater treatment and recycling efficiency.

[0018] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the process for the fiber dyeing wastewater treatment and recycling method optimized by big data according to this application.

[0021] Figure 2 This is a schematic diagram of the fiber dyeing wastewater treatment and recycling system optimized using big data according to this application.

[0022] Explanation of reference numerals in the attached diagram: Branch setting module 11, Activity matching analysis module 12, Activity constraint optimization module 13. Detailed Implementation

[0023] This application provides a method and system for treating and recycling fiber dyeing wastewater optimized using big data. This addresses the technical problem of low wastewater treatment and recycling efficiency in existing technologies due to the drastic fluctuations in wastewater quality and the poor adaptability of traditional treatment processes. By constructing multiple wastewater treatment and recycling pathways, combined with multiple preset activity ranges and multiple activity-control parameter constraints, the most suitable treatment path is selected based on real-time changes in water quality within the wastewater treatment tank. The treatment control parameters are dynamically optimized, thereby improving wastewater treatment and recycling efficiency.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a method for treating and recycling fiber dyeing wastewater optimized using big data. The method is applied to a fiber dyeing wastewater treatment and recycling system optimized using big data, and specifically includes the following steps:

[0026] Multiple wastewater treatment and recycling branches are set up, and the multiple wastewater treatment and recycling branches include multiple preset activity ranges and multiple activity-control parameter constraints with mapping configuration.

[0027] Furthermore, this application also includes the following steps: the plurality of activity-control parameter constraints correspond to the plurality of preset activity ranges, and each activity-control parameter constraint includes a control parameter constraint corresponding to each processing node; wherein, by identifying the activity-influence control parameter of each processing node, the control parameter constraint of each processing node is constructed based on the activity-influence threshold of the activity-influence control parameter.

[0028] Specifically, multiple wastewater treatment and recycling branches are set up to form a flexible treatment framework. Each wastewater treatment and recycling branch corresponds to different wastewater characteristics and is associated with multiple preset activity ranges, which are determined by presetting and classifying the sludge activity state of different wastewater treatment processes. The treatment process of each wastewater treatment and recycling branch has corresponding activity-control parameter constraints, including control parameters related to each treatment process node.

[0029] Each treatment branch and all its process nodes are analyzed individually. For each node, the control parameters affecting activity are identified. For example, in the aerobic biological treatment node, dissolved oxygen concentration, sludge return ratio, and hydraulic retention time are key parameters affecting activity. Control parameter constraints are constructed based on activity impact thresholds. By mining historical big data from long-term stable operation, the optimal operating range for each activity-influencing parameter under different influent activity states is determined. For example, when treating medium-active wastewater, maintaining the dissolved oxygen concentration in the aerobic tank at 3.0-4.0 mg / L ensures treatment efficiency while avoiding excessive aeration and energy waste; 3.0-4.0 is a specific constraint for this parameter within this activity range.

[0030] The activity impact control parameter refers to the operational variable that directly determines the core purification efficiency of wastewater at a specific point in the treatment process. The activity impact threshold is an optimal or permissible range of values ​​determined for a particular activity impact control parameter to achieve the expected treatment effect. For example, experiments may show that for wastewater in a certain activity range, sedimentation is ineffective when the PAC dosage is below 300 mg / L, while costs increase dramatically and sludge production is excessive when it exceeds 600 mg / L. Therefore, its activity impact threshold can be set at 300-600 mg / L.

[0031] All the constraints for each node and parameter are integrated to form an activity-control parameter constraint that is fully bound to each preset activity range and its corresponding wastewater treatment and recovery branch. Multiple activity-control parameter constraints correspond to multiple preset activity ranges, and each activity-control parameter constraint includes control parameter constraints corresponding to each treatment process node.

[0032] By setting up multiple wastewater treatment and recovery branches, and dynamically adjusting the recovery branches and control parameters according to changes in water quality, the wastewater treatment efficiency can be greatly improved, allowing for flexible responses to different water quality conditions, increasing dye removal rate, and ensuring that salt concentration meets discharge standards, thus achieving a high water recovery rate.

[0033] Based on the multiple preset activity ranges, a recyclable sludge resource activity matching analysis is performed on the current fiber dyeing wastewater in the wastewater treatment tank to obtain a matching wastewater treatment and recycling branch.

[0034] Furthermore, this application also includes the following steps: extracting the key components for recycling the recyclable sludge resources of the current fiber dyeing wastewater; the wastewater treatment tank includes a multi-parameter sensor array, and the multi-parameter real-time sensing dataset of the key components for recycling is identified by the multi-parameter sensor array; a recycling branch matching model is constructed based on the multiple preset activity intervals, and the active absorption feature vector of the multi-parameter real-time sensing dataset is extracted using the recycling branch matching model; the active absorption feature vector is analyzed to obtain multiple recycling branch matching probabilities; and a matching wastewater treatment recycling branch corresponding to the first recycling branch matching probability is selected from the multiple wastewater treatment recycling branches according to the multiple recycling branch matching probabilities.

[0035] Specifically, the fiber dyeing wastewater in the wastewater treatment pond contains various pollutants. The key components for recyclable sludge resources are extracted from the current fiber dyeing wastewater in the wastewater treatment pond. These components include recyclable substances contained in the fiber dyeing wastewater, such as deposited dyes, undegraded chemical auxiliaries, salts, and heavy metals. This does not refer to the sludge itself, but rather to identifying these key substances at the source of wastewater treatment from a holistic resource management perspective.

[0036] A multi-parameter sensor array installed in the wastewater treatment tank is activated, acting like a sophisticated sensory organ to synchronously or rapidly measure the concentration or related parameters of these key components in the wastewater, generating a multi-parameter real-time sensing dataset. This standardized dataset is then input into a recycling branch matching model for deep feature learning and extraction. This involves extracting the most representative active absorption feature vectors from the seemingly complex data, removing noise and redundancy, and focusing on the core issues. Based on the refined feature vectors, the recycling branch matching model analyzes and compares them within its complex mathematical space, calculating the matching degree between the current feature vector and each wastewater treatment recycling branch representing a preset activity range. This matching degree is then converted into specific matching probabilities for multiple recycling branches.

[0037] For example, a multi-parameter sensor array completes a comprehensive detection within 5 minutes, generating a raw multi-parameter real-time sensing dataset: extremely high absorbance of the indigo characteristic peak, with a color intensity 800 times higher; conductivity of 25000 μS / cm due to the extensive use of sodium chloride, and chloride ion concentration of 5200 mg / L; organic matter COD 3200 mg / L, BOD 640 mg / L, B / C ratio of 0.20; pH 9.8, and total suspended solids 850 mg / L. After standardization, this data is input into a deployed neural network model. A 128-dimensional hidden layer within the recycling branch matching model performs a nonlinear transformation on the data, ultimately extracting a representative 10-dimensional active absorption feature vector. The analysis layer of the recycling branch matching model processes this feature vector and outputs the final recycling branch matching probability. For example, wastewater treatment and recycling branch 1 is coagulation + aerobic biological treatment with a matching probability of 2%. It is judged that high salinity and low B / C ratio will severely inhibit conventional biological treatment. Wastewater treatment and recycling branch 2 is coagulation + Fenton oxidation + biological contact oxidation with a matching probability of 23%. It is judged that oxidation can degrade some organic matter and improve biodegradability, but high salinity is still a potential risk for subsequent biological treatment units. Wastewater treatment and recycling branch 3 is coagulation + membrane filtration + high-pressure reverse osmosis with a matching probability of 75%. It is judged that high salinity and recalcitrant COD are the core contradictions, and the combination of physicochemical and membrane methods is currently the most reliable path to ensure compliance and reuse.

[0038] Multiple recycling branch matching probabilities are scanned, and the highest-valued first recycling branch matching probability is selected. The matching wastewater treatment recycling branch corresponding to the first recycling branch matching probability is then locked. By introducing a multi-parameter sensor array and a recycling branch matching model, recyclable sludge resources in wastewater are accurately identified, and the optimal treatment path is intelligently selected based on water quality data. This significantly improves wastewater treatment efficiency and resource recovery rate, ensuring that all pollutants in the wastewater are effectively removed.

[0039] Furthermore, this application also includes the following steps: identifying the multi-parameter real-time sensing datasets of the key components to be recovered through the multi-parameter sensor array, wherein the multi-parameter real-time sensing datasets include at least dye-related sensing datasets, salt-related sensing datasets, organic matter-related sensing datasets, heavy metal-related sensing datasets, and microbial activity sensing datasets; standardizing the multi-parameter real-time sensing datasets and outputting the processed multi-parameter real-time sensing datasets.

[0040] Specifically, a multi-parameter sensor array located in the wastewater treatment tank simultaneously monitors multiple important parameters in the wastewater, such as pH, temperature, dye concentration, salt concentration, organic matter concentration, and heavy metal content. The multi-parameter sensor array consists of multiple sensors, typically including sensors for measuring different pollutants, such as detecting dyes, salts, organic matter, heavy metals, and microbial activity. For example, optical sensors scan spectra to capture dye-related information; conductivity sensors and ion electrodes analyze salt composition; COD / TOC analyzers and ultraviolet sensors resolve organic matter load; heavy metal detection units capture trace heavy metal signals; and rapid BOD detectors or respirometers detect the potential for microbial activity in the wastewater. All these sensors sample synchronously or sequentially at a high frequency, such as every 5-15 minutes, generating a raw, multi-dimensional, multi-parameter real-time sensor dataset.

[0041] Dye-related sensor datasets are collected using optical sensors to reflect the types and concentrations of dyes in wastewater. These datasets include absorbance values, chromaticity values, or estimated concentrations of characteristic dyes obtained through spectral fitting at specific wavelengths. For example, for reactive brilliant blue dyes, the absorbance near a wavelength of 620 nm is continuously monitored to serve as characterization data for the concentration of this type of dye.

[0042] Salt-related sensor datasets are used to reflect the total dissolved inorganic salts and specific ion concentrations in wastewater, including conductivity and specific ion concentrations such as chloride, sulfate, and sodium ions, as measured by ion-selective electrodes or online ion chromatography.

[0043] Organic matter-related sensor datasets are comprehensive datasets used to characterize the total amount and composition of organic pollutants in wastewater. They include online measurements of chemical oxygen demand (COD) or total organic carbon (TOC), as well as ultraviolet absorbance that can be measured at a wavelength of 254 nm via ultraviolet absorption spectroscopy, which is correlated with the concentration of certain aromatic organic compounds.

[0044] Heavy metal-related sensor datasets are collections of monitoring data on the concentration of trace heavy metal ions with biotoxicity in wastewater, including the concentrations of metal ions contained in common dyeing and printing auxiliaries such as copper, chromium, zinc, and nickel.

[0045] Microbial activity sensor datasets are used to indirectly and rapidly assess the biodegradability and microbial inhibition of wastewater. Instead of directly measuring microorganisms, they measure parameters that are strongly correlated with microbial activity through sensors, such as rapid estimates of biological oxygen demand (BOD), respiration rate, or the percentage inhibition rate output by a biotoxicity monitor.

[0046] Multi-parameter real-time sensing datasets are typically heterogeneous, meaning that each parameter has different units, magnitudes, or data distributions. The units and numerical ranges of different indicators vary greatly. Directly inputting these into the retrieval branch matching model can lead to the algorithm being dominated by features with large numerical values, while neglecting smaller but crucial features. Standardizing multi-parameter real-time sensing datasets unifies the raw data from different sensors, which have different dimensions and numerical ranges, to the same scale or standard normal distribution through mathematical transformation. The most common method is Z-score standardization, which involves subtracting the average of its historical data for each data point and then dividing by the standard deviation, resulting in a mean of 0 and a standard deviation of 1. This eliminates dimensional differences and prevents features with large numerical ranges from occupying inappropriate weights in subsequent models.

[0047] After standardization, a multi-parameter real-time sensor dataset is obtained. For example, in the wastewater treatment process of a textile factory, assume the dye concentration in the wastewater is 500 mg / L, the salt concentration is 2800 mg / L, the organic matter concentration is 150 mg / L, the heavy metal concentration is 0.3 mg / L, and the microbial activity is 0.85. These data are collected by a multi-parameter sensor array and, after standardization, the following standardized data are obtained: standardized dye concentration is 0.65, standardized salt concentration is 0.82, standardized organic matter concentration is 0.45, standardized heavy metal concentration is 0.12, and standardized microbial activity is 0.8.

[0048] Through standardization, the data of various parameters in the wastewater are unified to the same scale, eliminating unit differences that may be introduced by different sensors, enabling direct comparison and fusion of data. By accurately acquiring real-time wastewater data and standardizing it, and adjusting the treatment strategy according to real-time changes, the efficiency and recovery effect of wastewater treatment are significantly improved.

[0049] Furthermore, this application also includes the following steps: performing interval vectorization encoding on the multiple preset active intervals to obtain an active interval vector set; collecting recyclable sludge resource samples of fiber dyeing wastewater; extracting active absorption feature vectors from the recyclable sludge resource samples to output active absorption feature vector samples, wherein the active absorption feature vectors include known active interval categories; training the recycling branch matching model on similarity fitting according to the active absorption feature vector samples, using the known active interval categories as supervision labels, until the label prediction accuracy of the recycling branch matching probability samples output by the recycling branch matching model is greater than a preset threshold, and downloading the recycling branch matching model.

[0050] Specifically, multiple preset active regions are vectorized and encoded into mathematical vectors to form an active region vector set. Interval vectorization encoding refers to converting preset active regions described by text or rules into digital vector forms that can be directly processed and calculated by a computer. For example, three active regions are encoded into three-dimensional one-hot vectors: the high-activity region is encoded as [1,0,0], the medium-activity region is encoded as [0,1,0], and the low-activity region is encoded as [0,0,1].

[0051] Recyclable sludge resource samples of fiber dyeing wastewater were collected from actual production sites or pilot plants. Each sample included multiple characteristics, such as dye concentration, salt concentration, and pH value. Each sludge sample was assigned to a different activity range based on the type and concentration of pollutants it treated. A corresponding active absorption feature vector was extracted for each sludge sample according to its activity range. The active absorption feature vector extraction of the recyclable sludge resource samples yielded an active absorption feature vector sample, representing the sample's absorption and degradation capacity for different pollutants during the treatment process. The active absorption feature vector included known activity range categories, i.e., predefined activity range labels, with each range corresponding to different treatment capacities of the sludge or different concentration ranges of pollutants in the wastewater.

[0052] The training data includes the active absorption feature vector and the corresponding known activity range category for each sludge sample. The active absorption feature vector of each sample describes the sludge sample's ability to degrade pollutants. The supervision label is the known activity range category for each sample. A supervised learning model suitable for the recycling branch matching task is selected, such as a support vector machine (SVM) model. The input is the active absorption feature vector, and the output is the corresponding recycling branch matching probability. The model parameters are initialized, such as the kernel function parameters in the SVM model or the weights in the neural network. The active absorption feature vector of each sample is input into the model, and the model calculates the similarity between the sample features and each category of recycling branches. The loss function is obtained by calculating the difference between the model's output, i.e., the recycling branch matching probability, and the actual label. The model parameters are gradually adjusted using gradient descent or other optimization algorithms to minimize the loss function. Cosine similarity is used to measure the similarity between the input feature vector and the recycling branch feature vector; the SVM model will make predictions based on these similarities. During training, cross-validation is used to evaluate the model's performance. Typically, the dataset is divided into a training set and a validation set; the training set is used for model training, and the validation set is used to evaluate the model's prediction accuracy. After each training round, the model's prediction accuracy on the validation set is calculated to determine if the model meets a preset threshold. For example, if a 90% accuracy threshold is set, the model needs to achieve an accuracy exceeding 90% on the validation set to be considered training complete. The label prediction accuracy for retrieval branch matching is calculated based on the predictions on the validation set. Assuming the model's accuracy on the validation set is 85%, if the preset threshold is 90%, training needs to continue, and the model's hyperparameters need to be adjusted. If the accuracy meets the preset threshold, training ends, and the trained model is downloaded. If not, retraining is performed by adjusting model hyperparameters, such as the learning rate and regularization parameters, until the desired accuracy is achieved.

[0053] Once the model training is complete and the accuracy meets the preset threshold, the branch matching model is downloaded and recycled. The preset threshold is a minimum performance threshold set when the model is trained and evaluated. It usually refers to the percentage of samples whose predicted class matches the true class on an independent test dataset that the model must achieve. For example, if the preset threshold is set to 95%, the model is considered to have passed training and can be used only when its accuracy exceeds 95%.

[0054] By training a recycling branch matching model, the most suitable recycling branch is automatically selected based on real-time data from the wastewater treatment process. Selecting the most appropriate recycling branch avoids overtreatment or undertreatment, reducing waste of energy and chemical reagents.

[0055] Furthermore, this application also includes the following steps: performing an activity decay analysis on the current fiber dyeing wastewater in the wastewater treatment pond to obtain an activity decay coefficient; introducing the activity decay coefficient to perform a time-series decay analysis on the matching probability of the multiple recycling branches to obtain multiple recycling branch matching probability curves; analyzing the switching time sequence nodes of the multiple recycling branch matching probability curves, and re-obtaining the corresponding matching wastewater treatment recycling branches according to the switching time sequence nodes.

[0056] Furthermore, this application also includes the following step: the switching timing node is the timing node when the recycling branch corresponding to the matching probability of the first recycling branch changes in the multiple recycling branch matching probability curves.

[0057] Specifically, an activity decay analysis of recyclable sludge resources was conducted on the current fiber dyeing wastewater in the wastewater treatment pond. This involved quantitatively assessing the pattern of decrease in the concentration of pollutants or overall pollution indicators of the wastewater over time or during the treatment process. Several indicators strongly correlated with overall treatment efficiency were selected from the currently identified key recyclable components of the wastewater as direct characterizations of activity.

[0058] Once the wastewater enters the treatment tank, a high-frequency monitoring mode is activated. Using corresponding sensors in a multi-parameter sensor array, selected activity characterization indicators are continuously measured at short time intervals for a preset observation window, obtaining the initial data sequence of each indicator's decline over time. Based on the wastewater type and historical knowledge base, the best-matching decay kinetic model is invoked. For most biochemically or chemically oxidizable organic pollutants, an existing first-order reaction kinetic model is used. The collected time-concentration data is substituted into the model, and nonlinear regression or linearized least squares method is used for curve fitting to calculate the activity decay coefficient that best fits the measured data to the model. The activity decay coefficient is a parameter that measures the rate of activity decay of wastewater treatment materials during treatment, representing the percentage or proportion of activity decay per unit time. For example, if the activity decay coefficient is 0.1, it means that the sludge's decontamination capacity will decrease by 10% every hour.

[0059] By introducing the calculated activity decay coefficient and decaying the matching probabilities of multiple recovery branches, a matching probability curve for each recovery branch is obtained. As time progresses, the matching probability of the recovery branches gradually decreases, reflecting the gradual weakening of the processing effect. For each simulated time point, based on the predicted activity state, the logic of the recovery branch matching model is invoked again to calculate the matching probability of each branch at that time. Through continuous calculation, a sequence of matching probabilities of each branch changing over time can be obtained, i.e., multiple recovery branch matching probability curves. The recovery branch matching probability curve is a function graph of the matching probability of each preset processing branch changing over time. Through time-series decay analysis, multiple curves can be plotted, each curve representing the expected matching probability of a branch at various future time points, intuitively demonstrating how the optimality of the processing path shifts over time.

[0060] Simultaneously scan multiple recovery branch matching probability curves to find the first recovery branch matching probability corresponding to the highest probability. The time point when the recovery branch corresponding to the first recovery branch matching probability changes is the switching time point. Specifically, when the branch with the highest probability, i.e., the branch corresponding to the first recovery branch matching probability, is surpassed by the probability of another branch, the time point corresponding to the intersection of the two curves marks the result of one processing stage and the beginning of another better stage.

[0061] The corresponding matching wastewater treatment and recycling branches are replanned according to the switching time sequence nodes. The final decision output is not a single branch, but a sequence of treatment paths with time sequence switching points. As the activity of sludge in the wastewater gradually decreases, the wastewater treatment strategy is adjusted in a timely manner, such as adding new sludge or replacing treatment branches, to ensure the continuity of wastewater treatment effect.

[0062] Specifically, multiple process nodes and multiple sets of treatment control parameters are obtained for the matched wastewater treatment and recycling branch. The multiple sets of treatment control parameters are optimized according to the matched activity-control parameter constraints, and multiple sets of optimized treatment control parameters are output. The matched wastewater treatment and recycling branch performs fiber dyeing wastewater treatment and recycling operations according to the multiple sets of optimized treatment control parameters.

[0063] Furthermore, this application also includes the following steps: identifying the activity-affecting control parameters and inactivity-affecting control parameters among the multiple sets of processing control parameters; optimizing the activity-affecting control parameters according to the matched activity-control parameter constraints, and outputting the optimized activity-affecting control parameters; co-optimizing the inactivity-affecting control parameters based on the optimized activity-affecting control parameters, and outputting the optimized inactivity-affecting control parameters; and outputting multiple sets of optimized processing control parameters based on the optimized activity-affecting control parameters and the optimized inactivity-affecting control parameters.

[0064] Specifically, based on the multiple treatment process nodes of the currently activated matching wastewater treatment and recovery branch, all adjustable sets of treatment control parameters are listed for each node. A treatment process node is an independent process unit or operational stage that constitutes a complete wastewater treatment and recovery branch. Each node undertakes a specific treatment function and has an independent control objective. Treatment control parameters are operable variables that need to be set or adjusted at each treatment process node to control its operating status and achieve its treatment function, and are used to direct the actions of the field equipment.

[0065] By analyzing the roles and impacts of different treatment process nodes, we identified active and inactive control parameters among multiple sets of treatment control parameters. The classification criterion is whether the parameter directly participates in and significantly affects the chemical / physical reactions that transform or separate pollutants. Active control parameters are key variables that directly determine the removal efficiency and treatment cost of the core pollutants at that node, involving pollutant degradation in wastewater, sludge activity, and reaction rates. Inactive control parameters are auxiliary control variables that do not directly determine the removal effect of core pollutants, but are crucial for ensuring the stable, efficient, and safe operation of the node. These include equipment operating status, flow rate, and pump speed. These parameters mainly affect wastewater flow or equipment operating efficiency, but have a relatively small direct impact on the treatment effect.

[0066] Based on the specific characteristics of the current influent, and within the range specified by the matched activity-control parameter constraints, an optimization algorithm is used to find the optimal values ​​of a set of activity-influence control parameters, outputting an optimized set of activity-influence control parameters. The objective function of the optimization is typically to minimize the cost of this step while ensuring the expected treatment effect.

[0067] Based on the optimized active effect control parameters, and using these as known conditions, the optimization of inactive effect control parameters is initiated. Auxiliary parameters are then calculated and set based on the optimized active effect control parameters to optimize overall system operation. For example, the optimal stirring intensity is calculated based on the optimal dosage; and the most energy-efficient internal circulation flow rate is calculated based on the optimal biochemical load.

[0068] Combining all optimized control parameters, namely optimized active and inactive influence control parameters, multiple sets of optimized processing control parameters are output and sent to various field actuators, such as frequency converters, metering pumps, and valves, to drive the entire processing branch to operate according to the optimal settings.

[0069] For example, consider the ozone catalytic oxidation tower node in branch B, which treats wastewater in the intermediate-active range. Parameter identification: The active influence control parameters are ozone dosing acceleration rate (A) and reaction pH value (B); the inactive influence control parameters are the in-tower circulation pump frequency (C), which affects the contact time and uniformity of ozone and water. Constraints for the intermediate-active range are retrieved from the knowledge base: constraint A is 0.9-1.2 kg-O3 / kg-COD, constraint B is 8.0-9.0, constraint C is 30-50 Hz, and equipment safety and mixing requirements are considered. With the current real-time COD of the incoming water at 1600 mg / L and a target removal rate of 85%, simulation is performed using the built-in ozone oxidation kinetic model. Optimization is sought within the constraints, and the optimal active influence parameters are calculated as follows: A_opt = 1.05 kg-O3 / kg-COD, i.e., the actual dosing acceleration rate is 1.68 g-O3 / L·h, and B_opt = 8.5. Based on the optimized A_opt (ozone dosing acceleration rate) and B_opt (pH), the fluid dynamics and mass transfer model is invoked. At a dosing intensity of A_opt = 1.05, the optimal frequency for the circulating pump is 42Hz to achieve the best ozone mass transfer efficiency and avoid short-circuiting. Below 40Hz, mixing is uneven; above 45Hz, energy consumption increases while mass transfer gain is negligible; C_opt = 42Hz. The final optimized parameters are output as follows: the multiple optimized control parameters for this node are: ozone dosing acceleration rate 1.68 g-O3 / L·h, pH setpoint 8.5, and circulating pump frequency 42Hz. The control system will then precisely adjust the ozone generator, alkali pump, and circulating pump frequency converter accordingly.

[0070] Optimized control parameters significantly improve the removal efficiency of pollutants in wastewater. By comprehensively optimizing both active and inactive control parameters, the stable operation of the wastewater treatment system is ensured, avoiding fluctuations or failures caused by equipment instability or unreasonable control parameters.

[0071] Furthermore, this application also includes the following steps: obtaining the current total amount of fiber dyeing wastewater and the wastewater treatment component of the matched wastewater treatment and recycling branch; predicting the treatment time based on the total amount of wastewater and the wastewater treatment component to obtain the predicted treatment time; and conducting a time-series decay analysis on the matching probability of the multiple recycling branches based on the predicted treatment time by introducing an activity decay coefficient.

[0072] Specifically, the system reads the current total volume of fiber dyeing wastewater from the field instruments. Simultaneously, it retrieves the rated treatment capacity (i.e., the wastewater treatment capacity) of the currently matched wastewater treatment and recovery branch from the process knowledge base. The current total volume of fiber dyeing wastewater refers to the total volume of wastewater awaiting treatment or currently accumulated at the inlet of the wastewater treatment system. The wastewater treatment capacity of the matched wastewater treatment and recovery branch is the maximum designed or safe operating flow rate of the branch selected as the optimal path by the system. This represents the amount of water that the process route can stably treat and ensure effluent compliance within a unit of time, reflecting the upper limit of the branch's treatment capacity.

[0073] The treatment time is predicted based on the total amount of fiber dyeing wastewater and the amount of wastewater to be treated. The predicted treatment time represents the ideal time required to treat all the current wastewater. Typically, the predicted treatment time is calculated as: Total wastewater volume / Wastewater to be treated volume. For example, the level gauge shows the effective volume of the collection tank is 250 m³. 3 That is, the current total amount of fiber dyeing wastewater is 250m³. 3 The initially matched branch is C, an advanced treatment route that includes advanced oxidation and reverse osmosis. This branch treats 40m³ of wastewater. 3 / h. Therefore, the prediction processing time is 6.25h.

[0074] The activity decay coefficient obtained from previous real-time data analysis is introduced and combined with the predicted treatment duration to conduct future-oriented time-series simulation analysis. Starting from the currently measured initial activity of the wastewater, a predictive model of wastewater activity changing with treatment time is constructed using the activity decay coefficient, such as the exponential decay model A(t)=A0*e -kt Where k is the activity decay coefficient, t is the treatment time, A0 is the initial activity of the wastewater, and A(t) is the activity state of the wastewater at time t. The predicted treatment time is divided into several small time steps. For each future simulated time point, the activity state of the wastewater at that time is predicted using an exponential decay model. For each simulated time point, the predicted activity state is used as a virtual future influent characteristic. The core logic of the recycling branch matching model is called again to calculate the matching probability of each preset treatment branch under this assumed state, resulting in a sequence of matching probabilities of each branch changing over time. Connecting the sequences of matching probabilities of each branch changing over time yields multiple recycling branch matching probability curves, which visually demonstrate how the optimality of each process route increases or decreases over time throughout the entire treatment cycle.

[0075] Analyzing these probability curves, we pay particular attention to whether the branch corresponding to the matching probability of the first recovery branch changes over time, i.e., the highest probability. If a switching timing node is found, i.e., the moment when the highest probability changes hands, the strategy will be dynamically adjusted to generate a phased processing plan, rather than simply executing a single initial branch.

[0076] By predicting the time required for wastewater treatment, reasonable resources and time are allocated to each recovery branch to avoid excessively long or short treatment times, ensuring timely completion of wastewater treatment. Introducing an activity decay coefficient allows for dynamic adjustment of the matching probability and treatment strategy of the recovery branches based on real-time changes during the wastewater treatment process, ensuring high efficiency even during activity decay.

[0077] In summary, based on the identified matching wastewater treatment and recycling branch, the complete configuration information of the branch, namely multiple treatment process nodes and multiple sets of treatment control parameters, is retrieved from its extensive process knowledge base. In other words, the branch is structured into a structured control task list, which identifies which treatment process nodes are configured in what order or logic, and which key sets of treatment control parameters for each node need to be adjusted.

[0078] Using real-time characteristic data of the current wastewater as input, and invoking the activity-control parameter constraints corresponding to the branch and the current wastewater activity range as boundary conditions for the optimization problem, multiple sets of optimized treatment control parameters are obtained by optimizing the activity-affecting control parameters and co-optimizing the inactivity-affecting control parameters. These optimized treatment control parameters are then checked for rationality, such as whether they are within the safe operating range of the equipment. These numerical parameters are then compiled into equipment control instructions that the underlying control system can recognize and execute. The compiled control instruction set is synchronously sent to the field controllers of each treatment process node on the wastewater treatment and recycling branch. The field controllers drive the actuators to operate precisely according to the optimized parameters. The entire branch then starts, beginning efficient and economical treatment and recycling of the current batch of fiber dyeing wastewater. The operating effect is continuously monitored, and fine-tuning is made based on feedback to ensure that the operation always proceeds along the optimal trajectory. Each process unit operates under its customized optimal conditions, working synergistically to ensure that the final effluent consistently meets the highest standards. Through a global optimization algorithm, the lowest-cost operating point that meets the treatment requirements is found, reducing wastewater treatment costs. Optimized control parameters can improve the stability of the wastewater treatment process and avoid treatment fluctuations or failures caused by improper control parameters.

[0079] In summary, the fiber dyeing wastewater treatment and recycling method optimized by big data provided in this application has the following technical effects:

[0080] By setting up multiple wastewater treatment and recycling branches, each branch includes multiple preset activity ranges and multiple activity-control parameter constraints. Based on these preset activity ranges, a recyclable sludge resource activity matching analysis is performed on the current fiber dyeing wastewater in the wastewater treatment tank to obtain a matching wastewater treatment and recycling branch. Specifically, multiple treatment process nodes and multiple sets of treatment control parameters are acquired for each matching wastewater treatment and recycling branch. These parameters are then optimized according to the matched activity-control parameter constraints, resulting in multiple sets of optimized treatment control parameters. The matching wastewater treatment and recycling branch then performs the fiber dyeing wastewater treatment and recycling operation based on these optimized treatment control parameters. In other words, by constructing multiple wastewater treatment and recycling branches, combining multiple preset activity ranges and multiple activity-control parameter constraints, and selecting the most suitable treatment path based on real-time changes in water quality in the wastewater treatment tank, the treatment control parameters are dynamically optimized, thereby improving wastewater treatment and recycling efficiency.

[0081] Example 2: Based on the same inventive concept as the fiber dyeing wastewater treatment and recycling method optimized using big data in Example 1, this application also provides a fiber dyeing wastewater treatment and recycling system optimized using big data. Please refer to the appendix. Figure 2 The fiber dyeing wastewater treatment and recycling system optimized using big data includes:

[0082] Branch setting module 11 is used to set multiple wastewater treatment and recycling branches, wherein the multiple wastewater treatment and recycling branches include multiple preset activity ranges and multiple activity-control parameter constraints configured by mapping; Activity matching analysis module 12 is used to perform recyclable sludge resource activity matching analysis on the current fiber dyeing wastewater in the wastewater treatment tank according to the multiple preset activity ranges to obtain matching wastewater treatment and recycling branches; Activity constraint optimization module 13 is used to obtain multiple treatment process nodes and multiple sets of treatment control parameters of the matching wastewater treatment and recycling branches, optimize the multiple sets of treatment control parameters according to the matching activity-control parameter constraints, output multiple sets of optimized treatment control parameters, and the matching wastewater treatment and recycling branches perform fiber dyeing wastewater treatment and recycling operations according to the multiple sets of optimized treatment control parameters.

[0083] Furthermore, the branch setting module 11 in the fiber dyeing wastewater treatment and recycling system optimized by big data is also used for: the plurality of activity-control parameter constraints correspond to the plurality of preset activity ranges, and each activity-control parameter constraint includes control parameter constraints corresponding to each treatment process node; wherein, by identifying the activity-influence control parameters of each treatment process node, control parameter constraints for each treatment process node are constructed based on the activity-influence threshold of the activity-influence control parameters.

[0084] Furthermore, the activity matching analysis module 12 in the fiber dyeing wastewater treatment and recycling system optimized by big data is also used for: extracting the key components for recycling the recyclable sludge resources of the current fiber dyeing wastewater; the wastewater treatment tank includes a multi-parameter sensor array, which identifies the multi-parameter real-time sensing datasets of the key components for recycling; constructing a recycling branch matching model based on the multiple preset activity intervals, extracting the activity absorption feature vector of the multi-parameter real-time sensing dataset using the recycling branch matching model, analyzing the activity absorption feature vector, and obtaining multiple recycling branch matching probabilities; and selecting the matching wastewater treatment and recycling branch corresponding to the first recycling branch matching probability from the multiple wastewater treatment and recycling branches according to the multiple recycling branch matching probabilities.

[0085] Furthermore, the activity matching analysis module 12 in the fiber dyeing wastewater treatment and recycling system optimized by big data is also used to: identify the multi-parameter real-time sensing datasets of the key components to be recycled through the multi-parameter sensor array, wherein the multi-parameter real-time sensing datasets include at least dye-related sensing datasets, salt-related sensing datasets, organic matter-related sensing datasets, heavy metal-related sensing datasets, and microbial activity sensing datasets; standardize the multi-parameter real-time sensing datasets and output the processed multi-parameter real-time sensing datasets.

[0086] Furthermore, the activity matching analysis module 12 in the fiber dyeing wastewater treatment and recycling system optimized by big data is also used for: performing interval vectorization encoding on the multiple preset activity intervals to obtain an activity interval vector set; collecting recyclable sludge resource samples of fiber dyeing wastewater; extracting activity absorption feature vectors from the recyclable sludge resource samples and outputting activity absorption feature vector samples, wherein the activity absorption feature vectors include known activity interval categories; training the recycling branch matching model on similarity fitting according to the activity absorption feature vector samples, using the known activity interval categories as supervision labels, until the label prediction accuracy of the recycling branch matching probability samples output by the recycling branch matching model is greater than a preset threshold, and downloading the recycling branch matching model.

[0087] Furthermore, the activity matching analysis module 12 in the fiber dyeing wastewater treatment and recycling system optimized by big data is also used for: performing activity decay analysis on the recyclable sludge resources of the current fiber dyeing wastewater in the wastewater treatment tank to obtain the activity decay coefficient; introducing the activity decay coefficient to perform time-series decay analysis on the matching probability of the multiple recycling branches to obtain multiple recycling branch matching probability curves; analyzing the switching time sequence nodes of the multiple recycling branch matching probability curves, and re-obtaining the corresponding matching wastewater treatment and recycling branches according to the switching time sequence nodes.

[0088] Furthermore, the activity matching analysis module 12 in the fiber dyeing wastewater treatment and recycling system optimized by big data is also used for: the switching timing node is the timing node when the recycling branch corresponding to the matching probability of the first recycling branch in the multiple recycling branch matching probability curves changes.

[0089] Furthermore, the activity matching analysis module 12 in the fiber dyeing wastewater treatment and recycling system optimized using big data is also used to: identify the activity-affecting control parameters and inactivity-affecting control parameters among the multiple sets of treatment control parameters; optimize the activity-affecting control parameters according to the matched activity-control parameter constraints, and output the optimized activity-affecting control parameters; perform collaborative optimization of the inactivity-affecting control parameters based on the optimized activity-affecting control parameters, and output the optimized inactivity-affecting control parameters; and output multiple sets of optimized treatment control parameters based on the optimized activity-affecting control parameters and the optimized inactivity-affecting control parameters.

[0090] Furthermore, the activity constraint optimization module 13 in the fiber dyeing wastewater treatment and recycling system optimized by big data is also used to: obtain the current total amount of fiber dyeing wastewater and the wastewater treatment component of the matching wastewater treatment and recycling branch; predict the treatment time based on the total amount of wastewater and the wastewater treatment component to obtain the predicted treatment time; and, based on the predicted treatment time, introduce an activity decay coefficient to perform time-series decay analysis on the matching probability of the multiple recycling branches.

[0091] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The fiber dyeing wastewater treatment and recycling method and specific examples optimized by big data in the aforementioned embodiment one are also applicable to the fiber dyeing wastewater treatment and recycling system optimized by big data in this embodiment. Through the foregoing detailed description of the fiber dyeing wastewater treatment and recycling method optimized by big data, those skilled in the art can clearly understand the fiber dyeing wastewater treatment and recycling system optimized by big data in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0092] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0093] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A fiber dyeing wastewater treatment and recycling method optimized using big data, characterized in that, include: Multiple wastewater treatment and recycling branches are set up, and the multiple wastewater treatment and recycling branches include multiple preset activity ranges and multiple activity-control parameter constraints with mapping configuration; Based on the multiple preset activity ranges, a recyclable sludge resource activity matching analysis is performed on the current fiber dyeing wastewater in the wastewater treatment tank to obtain a matching wastewater treatment and recycling branch. Specifically, multiple process nodes and multiple sets of treatment control parameters are obtained for the matched wastewater treatment and recycling branch. The multiple sets of treatment control parameters are optimized according to the matched activity-control parameter constraints, and multiple sets of optimized treatment control parameters are output. The matched wastewater treatment and recycling branch performs fiber dyeing wastewater treatment and recycling operations according to the multiple sets of optimized treatment control parameters.

2. The fiber dyeing wastewater treatment and recycling method optimized using big data as described in claim 1, characterized in that, Based on the multiple preset activity ranges, a recyclable sludge resource activity matching analysis is performed on the current fiber dyeing wastewater in the wastewater treatment tank to obtain a matching wastewater treatment and recycling branch. The method includes: Extracting key components for the recovery of recyclable sludge resources from current fiber dyeing wastewater; The wastewater treatment tank includes a multi-parameter sensor array, which identifies the multi-parameter real-time sensor datasets of the key components to be recovered. A retrieval branch matching model is constructed based on the multiple preset active regions. The active absorption feature vector of the multi-parameter real-time sensing dataset is extracted using the retrieval branch matching model. The active absorption feature vector is analyzed to obtain multiple retrieval branch matching probabilities. According to the matching probability of the multiple recycling branches, the matching wastewater treatment recycling branch corresponding to the first recycling branch matching probability is selected from the multiple wastewater treatment recycling branches.

3. The fiber dyeing wastewater treatment and recycling method optimized using big data as described in claim 2, characterized in that, The multi-parameter real-time sensing datasets of the key recovered components are identified by the multi-parameter sensor array. The multi-parameter real-time sensing datasets include at least dye-related sensing datasets, salt-related sensing datasets, organic matter-related sensing datasets, heavy metal-related sensing datasets, and microbial activity sensing datasets. The multi-parameter real-time sensing dataset is standardized, and the processed multi-parameter real-time sensing dataset is output.

4. The fiber dyeing wastewater treatment and recycling method optimized using big data as described in claim 3, characterized in that, The method for constructing a recycling branch matching model based on the multiple preset active regions and training the recycling branch matching model includes: The multiple preset active regions are encoded using interval vectorization to obtain an active region vector set; Collect recyclable sludge samples from fiber dyeing wastewater; The recyclable sludge resource sample is subjected to active absorption feature vector extraction, and the active absorption feature vector sample is output. The active absorption feature vector includes known active interval categories. The retrieval branch matching model is trained by similarity fitting based on the active absorption feature vector samples, with the known active interval category as the supervision label, until the label prediction accuracy of the retrieval branch matching probability samples output by the retrieval branch matching model is greater than a preset threshold, and then the retrieval branch matching model is downloaded.

5. The fiber dyeing wastewater treatment and recycling method optimized using big data as described in claim 2, characterized in that, The method further includes selecting a matching wastewater treatment recycling branch from the plurality of wastewater treatment recycling branches according to the matching probability of the first recycling branch, wherein the matching probability of the first recycling branch corresponds to the matching probability of the first recycling branch; An activity decay analysis was performed on the current fiber dyeing wastewater in the wastewater treatment pond to obtain the activity decay coefficient; An activity decay coefficient is introduced to perform time-series decay analysis on the matching probability of the multiple recovery branches, resulting in multiple recovery branch matching probability curves; Analyze the switching timing nodes of the matching probability curves of the multiple recycling branches, and re-obtain the corresponding matching wastewater treatment recycling branches according to the switching timing nodes.

6. The fiber dyeing wastewater treatment and recycling method optimized using big data as described in claim 5, characterized in that, The matching wastewater treatment and recycling branch performs fiber dyeing wastewater treatment and recycling operations, and the method further includes: Obtain the current total amount of fiber dyeing wastewater, and the wastewater treatment component of the matched wastewater treatment and recovery branch; Based on the total amount of wastewater and the amount of wastewater to be treated, the treatment time is predicted to obtain the predicted treatment time. Based on the predicted processing time, an activity decay coefficient is introduced to perform time-series decay analysis on the matching probability of the multiple recycling branches.

7. The fiber dyeing wastewater treatment and recycling method optimized using big data as described in claim 1, characterized in that, Configuring multiple activity-control parameter constraints can be achieved through methods including: The plurality of activity-control parameter constraints correspond to the plurality of preset activity ranges, and each activity-control parameter constraint includes control parameter constraints corresponding to each processing node; Specifically, by identifying the activity impact control parameters of each processing node, control parameter constraints for each processing node are constructed based on the activity impact threshold of the activity impact control parameters.

8. The fiber dyeing wastewater treatment and recycling method optimized using big data as described in claim 7, characterized in that, The activity constraint optimization of the multiple sets of treatment control parameters is performed according to the matched activity-control parameter constraints, and the method includes: Identify the active and inactive influence control parameters among the multiple sets of processing control parameters; The activity-influence control parameters are optimized according to the matching activity-control parameter constraints, and the optimized activity-influence control parameters are output. Based on the optimized active influence control parameters, the inactive influence control parameters are co-optimized, and the optimized inactive influence control parameters are output. Based on the optimized control parameters for the effects of activity and inactivity, multiple sets of optimized processing control parameters are output.

9. The fiber dyeing wastewater treatment and recycling method optimized using big data as described in claim 5, characterized in that, The switching timing node is the timing node in the multiple recovery branch matching probability curves when the recovery branch corresponding to the first recovery branch matching probability changes.

10. A fiber dyeing wastewater treatment and recycling system optimized using big data, characterized in that, The step of implementing the fiber dyeing wastewater treatment and recycling method optimized by big data according to any one of claims 1 to 9, wherein the fiber dyeing wastewater treatment and recycling system optimized by big data comprises: A branch setting module is used to set up multiple wastewater treatment and recycling branches, wherein the multiple wastewater treatment and recycling branches include multiple preset activity ranges and multiple activity-control parameter constraints configured by mapping; The activity matching analysis module is used to perform recyclable sludge resource activity matching analysis on the current fiber dyeing wastewater in the wastewater treatment tank according to the multiple preset activity ranges, and to obtain the matching wastewater treatment and recycling branch. The activity constraint optimization module is used to obtain multiple process nodes and multiple sets of treatment control parameters of the matched wastewater treatment and recovery branch, optimize the multiple sets of treatment control parameters according to the matched activity-control parameter constraints, and output multiple sets of optimized treatment control parameters. The matched wastewater treatment and recovery branch performs fiber dyeing wastewater treatment and recovery operations according to the multiple sets of optimized treatment control parameters.

Images