Pretreatment method for printing wastewater in carton packaging

Abstract

The invention relates to the technical field of printing wastewater treatment, in particular to a method for pretreating printing wastewater in carton packaging, which comprises the following steps of: collecting wastewater with different concentrations according to quality, analyzing characteristics, and ensuring that the wastewater enters a regulating tank to be homogenized and homogenized; an automatic intelligent decision processing technology is adopted according to water quality data obtained through quality-based collection and analysis; different kinds of wastewater are subjected to attacking physicochemical strengthening pretreatment; an intensified preprocessing step driven by an intelligent algorithm to complete removal and conversion tasks of chemical pollutants; and fusing the multi-dimensional indexes into a comprehensive evaluation score for visually evaluating the overall performance. According to the method, a full-process closed loop with intelligent decision making, accurate execution and dynamic evaluation is constructed, collaborative operation of pretreatment is dynamically optimized and commanded according to real-time water quality data, the pretreatment efficiency and economical efficiency are quantified, the removal efficiency is high, the operation cost is low, and the effluent quality is stable and controllable, and the method is suitable for popularization and application. And ideal conditions are provided for subsequent deep treatment.

Description

Technical Field

[0001] This invention relates to the field of printing wastewater treatment technology, and more particularly to a method for pretreatment of printing wastewater in cardboard packaging. Background Technology

[0002] In the production process of printed cartons in the cardboard packaging industry, water is used in many stages and the water consumption is large. Currently, each process in the carton printing production process requires tap water and chemical tanks. During operation, there are multiple overflow water washes after each chemical tank. The types of water are numerous, and there are significant differences in water quality and quantity. The composition of the wastewater is also different, so the treatment process is also quite complicated. Usually, corresponding treatment processes need to be adopted according to the characteristics of different wastewaters. After the production workshop is divided, the various streams of production wastewater will flow into different wastewater storage tanks and be sent to pretreatment facilities. The comprehensive wastewater treatment facility is responsible for the final treatment and discharge.

[0003] Existing technologies generally exhibit the characteristics of wastewater pretreatment methods, with poor coordination among various stages, relatively rough control over treatment intensity, and difficulty in dealing with the characteristics of printing wastewater with large fluctuations in water quality and quantity. Ultimately, this leads to insufficient system stability, high treatment costs, and inherent defects such as over-reliance on chemicals and high energy and material consumption. There is an urgent need to develop an efficient, economical, and rapid integrated pretreatment technology. Summary of the Invention

[0004] The purpose of this invention is to provide a method for pretreatment of printing wastewater in cardboard packaging in order to solve the problems in the prior art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for pretreatment of printing wastewater in cardboard packaging, comprising the following steps: Step S1: Collect and analyze the characteristics of printing wastewater separately, collect wastewater of different concentrations separately, and ensure that the wastewater entering the equalization tank is homogeneous and of uniform quantity. Step S2: The pretreatment technology selection automatically and intelligently decides on the treatment technology based on the water quality data obtained from the separate collection and real-time analysis. Step S3, Physicochemical Enhanced Pretreatment, receives instructions from the pretreatment technology steps and is responsible for the intensive treatment of wastewater with different concentrations and recalcitrant degradation. Step S4: Based on microbial enhanced pre-biochemical treatment, intelligent algorithms drive enhanced chemical pretreatment to remove and transform chemical pollutants in advance. Step S5, preprocessing effect evaluation, integrates indicators from multiple dimensions into a comprehensive evaluation score for intuitive evaluation of overall performance.

[0006] The beneficial effects of the technical solution provided by this invention include at least the following: This invention utilizes historical data to predict future water quality trends. Based on the prediction results, it proactively pre-homogenizes the water. Online sensors at the outlet of the equalization tank monitor the water quality after homogenization in real time and send the predicted data to the subsequent enhanced treatment unit in advance. The treatment unit can then pre-adjust the reagent dosage and formulation to achieve process synergistic optimization.

[0007] This invention constructs a random forest classification and regression model based on machine learning. Based on the input features, it predicts the most suitable treatment technology, including Fenton oxidation, electrochemical, and direct biochemical. If the Fenton path is selected, it predicts the optimal H2O2 dosage based on features such as COD, UV254, pH, and Fe2+.

[0008] The PID algorithm of this invention outputs a control signal in real time based on the magnitude, duration, and trend of the deviation, dynamically adjusting the output current of the DC power supply to achieve precise energy-saving control of power consumption, reduce operating costs, automatically reduce current under low load, and precisely increase current under high load, avoiding energy waste. All data is recorded and traceable, providing possibilities for subsequent process optimization, big data analysis, and remote monitoring.

[0009] The enhanced pretreatment step driven by the intelligent algorithm of this invention completes the task of removing and transforming chemical pollutants, and records process data more comprehensively, providing a solid foundation for evaluating the technical effectiveness, economic indicators and intelligent level of the final pretreatment effect evaluation step.

[0010] The pretreatment performance comprehensive score weight of this invention is not fixed, but dynamically adjusted according to operational goals and water quality characteristics. The dynamic weight is generated in real time by a small expert rule base or reinforcement learning model, which has a stronger anti-interference ability. Attached Figure Description

[0011] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 The diagram illustrates the method steps provided in this embodiment of the invention. Detailed Implementation

[0013] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for pretreating printing wastewater in cardboard packaging according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0014] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0015] The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0016] The following description, in conjunction with the accompanying drawings, details a specific scheme for a pretreatment method of printing wastewater in cardboard packaging provided by the present invention.

[0017] Example Please see Figure 1 The diagram illustrates a method flowchart for pretreatment of printing wastewater in cardboard packaging according to an embodiment of the present invention. The method includes the following steps: Step S1: Collect and analyze the characteristics of printing wastewater separately, collect wastewater of different concentrations separately, and ensure that the wastewater entering the equalization tank is homogeneous and of uniform quantity. Step S1 further includes the following sub-steps: S1-1, refined and classified collection: printing wastewater is divided into three levels according to concentration. The first category of high-concentration wastewater includes ink tray cleaning, ink tank cleaning, ink roller cleaning, waste ink, and dampening solution that is replaced regularly. The content of pollutants such as organic matter and volatile organic compounds is extremely high, mainly from the binders and additives in the ink. A special anti-corrosion sealed collection tank is set up next to the printing machine, with a conspicuous red label and QR code. The operator scans the code to record the discharge time, machine, ink type and other relevant information. The second category of medium-concentration wastewater includes periodic rinsing water for printing press cylinders and components, and plate washing wastewater. It mainly comes from the cleaning and maintenance process of printing plates in traditional offset printing. This type of wastewater contains developer, which is usually strongly alkaline, and may contain dissolved photosensitive resin layer components and trace heavy metals. Automatically opening and closing diversion weirs are installed in the workshop drainage ditches. During production or rinsing periods, the weirs guide the wastewater into the medium-concentration collection pipes. During rainfall or ground cleaning periods, the weirs are closed to prevent dilution. The third type of low-concentration wastewater comes from workshop floor cleaning water, equipment exterior wiping and cleaning, and cooling system wastewater. It is mixed with dripping ink, lubricating oil, dust, and contains mud and impurities from the ground. It flows directly into the equalization tank through a separate open channel or pipe. S1-2, multidimensional characteristic analysis and data management, enables online real-time monitoring. Each collection pipe and main collection pipe is equipped with an online multi-parameter water quality sensor array, integrated within a flow tank. This sensor array combines pH, conductivity, and turbidity sensors to rapidly monitor real-time acidity and alkalinity and calculate total dissolved solids (TDS) and suspended solids (SDS) concentrations. A miniaturized UV-Vis spectral water quality sensor provides continuous real-time monitoring. The sensor has a pre-established correlation model between UV absorbance at 254nm wavelength and chemical oxygen demand (COD) measured by standard methods. By measuring UV absorbance in real-time, the estimated COD value of the water sample can be quickly calculated. Simultaneously, by selecting specific wavelengths in the visible light range to monitor absorbance and establishing a model with standard colorimetric values, colorimetric values ​​can be calculated in real-time. The built-in processor processes the spectral data in real-time and directly outputs estimated COD and colorimetric values ​​through the algorithm model, achieving second-level water quality analysis. All sensor data is acquired via PLC and uploaded to the central SCADA control platform system. Wastewater samples of different categories are taken weekly and analyzed using standard methods. The laboratory is equipped with a COD rapid analyzer, a spectrophotometer (for colorimetry), and an ICP-MS. Laboratory data is used as the "true value" for calibration and the establishment of predictive models for online sensors. Establish a wastewater record database, including data items such as source machine, discharge time, ink type and color, online monitoring data, laboratory data, and estimated volume. Store and manage the data using an SQL database or time-series database, and create electronic records for different batches of high-concentration waste liquid and each batch of medium- and low-concentration wastewater.

[0018] The intelligent homogenization and equalization system establishes multiple treatment tanks and adopts a combination of perforated aeration pipes and submersible agitators to prevent sedimentation and avoid VOCs volatilization. The central control system calculates the weighted average water quality based on the wastewater records of each storage tank, simulates the possible water quality after mixing, and intelligently decides the opening sequence and duration of the pump valve linkage system based on the simulation results. It pumps wastewater of different concentrations into the equalization tank according to the preset ratio, controlling the fluctuation of influent water quality from the source. The system uses historical data to train a simple time series model to predict water quality trends over a future period. Based on the prediction results, it can start the mixing device or adjust the pumping scheme in advance to actively pre-homogenize the water. Online sensors at the outlet of the regulating tank monitor the water quality after homogenization in real time and send the data, especially the estimated COD data, to the subsequent enhanced treatment unit in advance. The treatment unit can then adjust the chemical dosing formula and dosage accordingly to achieve synergistic optimization throughout the entire process.

[0019] Step S2: Pretreatment technology selection: The pretreatment technology is automatically and intelligently determined based on the water quality data obtained from separate collection and real-time analysis. Step S2 further includes the following sub-steps: S2-1, the intelligent decision-making process, is based on real-time data and historical databases. It dynamically generates the optimal treatment plan through algorithm models, and builds a data-driven, rule-based and predictive model-based dynamic decision-making system to achieve precision, adaptability and cost reduction and efficiency improvement in the treatment process. It abandons the traditional "experience-based" or "one-size-fits-all" treatment methods, refines decision-making, determines the treatment route for different batches of wastewater, and provides early warning and control suggestions for subsequent microbial enhanced pretreatment steps. S2-2, based on industrial standard protocol data interfaces such as OPC UA and Modbus, acquires data from sensors and PLCs. The system receives data streams from the "separate collection and characteristic analysis and adjustment" unit in real time. The time-series database efficiently stores and queries water quality data with timestamps. The built-in data cleaning algorithm removes outliers caused by momentary sensor failures. S2-3: Extract key features for decision-making from raw data, perform detailed feature extraction and water quality assessment, and calculate key indicators in real time using the calculation module. BOD5 / COD ratio: B / C = k1 * UV254 + k2 * BOD Among them, UV254 represents the ultraviolet absorbance at a wavelength of 254 nanometers, k1 and k2 are coefficients obtained from regression of historical data, COD is chemical oxygen demand, which represents the amount of oxidant consumed by the oxidation of organic matter in water under the action of strong oxidants, representing the total amount of almost all organic matter in water, including biodegradable and non-biodegradable, BOD is biochemical oxygen demand, BOD5 is five-day biochemical oxygen demand, which represents the amount of dissolved oxygen consumed by microorganisms in decomposing organic matter in water within 5 days at 20℃, and the B / C ratio directly reflects the proportion of biodegradable organic matter in the total organic matter.

[0020] Based on the extracted features, innovative algorithms are used for intelligent technology selection and parameter optimization, and a built-in "if-then-else" rule base based on the experience of water treatment experts is incorporated. For example: If [COD>2000mg / L] AND [B / C<0.3], select the "Fenton oxidation" path; A machine learning-based random forest classification and regression model is constructed to predict the most suitable treatment technology based on input features, including Fenton oxidation, electrochemical, and direct biochemical methods. If the Fenton pathway is selected, the most suitable treatment technology is determined based on COD, UV254, pH, and Fe. 2+ Based on characteristics such as [list of characteristics], the optimal amount of H2O2 required can be predicted. Traditional chemical dosing calculations are mostly based on static formulas, which cannot adapt to dynamic changes in water quality. This design introduces an adaptive Fenton chemical dosing estimation model, using a feedback correction factor (λ) to dynamically adjust the dosing amount, giving the model self-learning capabilities.

[0021]

[0022] in, This represents the amount of hydrogen peroxide that needs to be added at the current time t. The dynamic correction factor at the current moment is represented by α, which represents the forgetting factor (0 < α < 1) and is used to balance the influence of historical experience and the latest data (usually set to 0.7-0.9). This indicates the COD removal rate achieved in the actual treatment of the previous batch of water. This means that the system predicts the COD removal rate of the previous batch of water based on the model, and then predicts the dosage and performs the treatment based on the initial model. After treatment, the actual effluent COD is analyzed, the actual removal rate is calculated, and the actual value is compared with the predicted value to calculate the efficiency ratio. This ratio is used to update the correction factor λ(t). If the actual effect is worse than the prediction, the dosage will be automatically reduced next time; if the effect is better, it will be maintained or slightly increased. The correction factor λ(t) will be continuously iterated and optimized, making the model prediction more and more accurate, ultimately achieving precise dosing on demand and avoiding waste.

[0023] Step S3, Physicochemical Enhanced Pretreatment: Receives instructions from the pretreatment technology steps and is responsible for tackling difficult-to-degrade wastewater of different concentrations. In step S3, instructions are received from the pretreatment technology step, which is responsible for the intensive treatment of high-concentration, recalcitrant wastewater. Compared with the traditional Fenton process, the electro-Fenton process generates H2O2 and Fe in situ through electrolysis. 2+ The reaction is more efficient, produces less sludge, and is easier to automate. First, wastewater from the equalization tank is fed into the inlet tank of the electro-Fenton reactor. Upon receiving instructions from the upper level, the intelligent pH controller automatically adds acid, precisely adjusting the pH to the optimal reaction range of 2.5-3.5. Based on data from the online COD meter, a small amount of external Fe is added as instructed. 2+ As a catalytic supplement; Wastewater flows into the Fenton electro-reactor, which contains an anode and a cathode. When direct current is applied, the cathode continuously reduces O2 to H2O2; the anodic oxidation produces Fe... 2+ It forms Fenton's reagent with H2O2, generating a large number of OH free radicals, which efficiently degrade organic matter. Using three-dimensional electrode technology, particulate conductive medium is filled between the anode and cathode to form countless micro-electrolysis units, which greatly increases the reaction area and efficiency. After the reaction, the wastewater enters the neutralization tank, where intelligent alkali dosing adjusts the pH to 7-8, and Fe... 2+ to become Fe 3+ The process forms ferric hydroxide flocs, which effectively adsorb, trap, and co-precipitate oxidized organic matter and colloids. The wastewater then enters an inclined plate sedimentation tank or a high-efficiency clarification tank for mud-water separation.

[0024] Based on changes in influent COD, the current density I of the electro-Fenton reactor is dynamically adjusted to minimize energy consumption while ensuring treatment effectiveness.

[0025] Among them, I actual K indicates direct control of power consumption. p k i k d These represent the proportional, integral, and derivative coefficients of the PID controller, respectively, and require tuning. e(t) = COD in (t)-COD setpoint This indicates the deviation between the current influent COD and the set target COD. This represents the integral of the deviation from time 0 to time t, which represents the cumulative total of all deviations throughout history. The derivative of the deviation represents the instantaneous rate of change of the deviation at the current moment. It is the slope of the deviation-time curve at that point, reflecting the speed and direction of the deviation change. The system monitors the influent COD(t) in real time and compares it with the set desired effluent COD. setpoint The deviation e(t) is calculated by comparison. Based on the magnitude, duration, and trend of the deviation, the PID algorithm outputs a control signal in real time to dynamically adjust the output current I of the DC power supply. actual This achieves precise energy-saving control of electricity consumption; Compared to traditional adjustment modes, this dynamic control strategy is extremely energy-efficient and reduces operating costs. It automatically reduces current under low load and precisely increases current under high load, avoiding energy waste and significantly reducing electricity costs, a major operating cost of advanced oxidation processes. Regardless of fluctuations in influent water quality and quantity, it can automatically resist interference and strive to keep the effluent COD within the acceptable range through dynamic adjustment, ensuring stable effluent quality and avoiding accidents caused by excessive effluent due to shock loads. It also avoids continuous full-load operation, reducing the burden on electrical equipment and helping to extend its service life. All data is recorded and traceable, providing possibilities for subsequent process optimization, big data analysis, and remote monitoring. The physicochemical unit outputs key performance indicators, including COD removal rate, color removal rate, power consumption, and iron sludge yield. Current density (I), voltage (U), pH / ORP curve, and reagent consumption process data are used to evaluate the efficiency and economy of the physicochemical process. These data are the core basis for evaluating the intelligence level and operational stability of the entire pretreatment system.

[0026] Step S4: Based on microbial enhanced pre-biochemical treatment, intelligent algorithms drive enhanced chemical pretreatment to remove and transform chemical pollutants in advance. Step S4 further includes the following sub-steps: S4-1, the effluent after physical and chemical treatment first enters the nutrient conditioning tank. The system automatically calculates and adds nitrogen and phosphorus sources based on the data from the online TOC / C / N analyzer, and precisely controls the ratio, for example, C:N:P=100:5:1; In S4-2, wastewater enters the hydrolysis acidification reactor, which is filled with specific compound bacterial agents and enzyme preparations. The reactor is equipped with a biological packing zone, where the combined packing forms a high-biomass biofilm. Under anoxic conditions, microorganisms and enzymes completely convert small-molecule organic matter and some recalcitrant substances into volatile fatty acids (VFAs). The reactor is equipped with a high-flow-rate internal circulation pump to return the sludge-water mixture to the inlet, maintaining a high biological concentration. A sludge-water separation zone is set at the rear end, where sludge is automatically returned to maintain a high sludge concentration in the reactor.

[0027] The concentration and composition of VFAs in the effluent from the hydrolysis acidification tank are monitored in real time, and the hydraulic retention time (HRT) and internal circulation ratio (R) are adjusted based on feedback to ensure that the system is in the most efficient acid production state. This is based on a microbial activity feedback control algorithm that monitors VFAs in real time.

[0028] in, Indicates the total VFAs concentration in the effluent. These represent the total organic carbon concentrations in the influent and effluent, respectively. This represents the VFAs yield index, which reflects the amount of VFAs converted per unit of TOC removal. A higher index indicates higher acid production efficiency and healthier operation of the system. The system calculates VFA-PI in real time and compares it with the set target value. If the measured value is lower than the target, it indicates a decrease in acid production efficiency, and the hydraulic calculation time should be adjusted accordingly.

[0029] in, This represents the calculated new hydraulic residence time. Indicates the current hydraulic residence time. The target yield index is determined using historical best performance data. This represents the yield index calculated through real-time monitoring. This represents the gain control coefficient. The algorithm automatically extends the HRT by reducing the influent flow rate, giving the microorganisms a longer reaction time; conversely, it appropriately shortens the HRT to improve treatment efficiency. At the same time, the internal circulation ratio is also adjusted in conjunction with the VFAs concentration. When the concentration is low, the circulation ratio is increased to improve mass transfer efficiency. Key performance indicators of microbial unit output include B / C ratio improvement, VFAs yield, hydrolysis acidification rate, VFAs concentration and composition changes, HRT adjustment log, nutrient salt addition process data. These data are used to evaluate the degree of improvement in system biodegradability and microbial activity, control log, adaptive adjustment records, and final optimized parameters. The enhanced pretreatment step driven by intelligent algorithms completes the task of removing and transforming chemical pollutants, and records process data more comprehensively, providing a solid foundation for evaluating the technical effectiveness, economic indicators and intelligence level of the final pretreatment effect assessment step.

[0030] Step S5: Preprocessing effect evaluation, which integrates multiple dimensions of indicators into a comprehensive evaluation score for intuitive evaluation of overall performance; Step S5 further includes the following sub-steps: S5-1, the system uses ETL tools to extract data from sensors, PLCs, and control systems of all preceding units, transforms and loads the data, and finally stores the data in the real-time database. During this process, data cleaning and timing alignment are performed to eliminate errors caused by different sensor data acquisition frequencies and delays. Information such as wastewater source, initial water quantity, water quality (COD), and color is obtained from the wastewater collection unit, and real-time energy consumption, H2O2, and Fe2+ are obtained from the physicochemical enhancement unit. 2+ Actual dosage, pH / ORP curve, and effluent COD data were obtained from the microbial pre-biochemical unit, including influent and effluent BOD, VFA concentration and composition, nutrient dosage, HRT, and internal circulation ratio. S5-2, based on the data after cleaning, calculate the treatment efficiency indicators, including total COD removal rate, total color removal rate, B / C ratio improvement rate, and VFAs yield; calculate the economic indicators, including: chemical consumption index, energy consumption index, and unit COD removal cost: (electricity cost + chemical cost + labor cost + equipment depreciation) / Δ total COD. The process stability index (SSI), standard deviation (σ), and coefficient of variation (CV) of the key parameters are calculated as follows: SSI = 1 / (σ CODout + σ pH + ...), the higher the value, the more stable it is; Ecotoxicity reduction indices were calculated, and the bioinhibition rates of influent and effluent were measured using an online microbial respiration rate meter. Toxicity reduction rate = (1 - effluent inhibition rate / influent inhibition rate) * 100%.

[0031] The above-mentioned multiple dimensions of indicators are integrated into a single, comprehensive score (PPI) for intuitive evaluation of overall performance and automatic diagnosis of problems. First, the indicators of different dimensions are normalized to the [0, 1] interval. For benefit-type indicators and cost-type indicators (such as unit cost), the following formula is used: Benefit-oriented: ; Cost-based: ; in, This represents the raw indicator value, which is the actual measured or calculated value of a certain indicator at a specific time t. For example: If the COD removal rate is 95%, then x = 95. The unit energy cost is 3.5 yuan / ton of water, therefore x = 3.5. If turbidity = 5 NTU, then x = 5. and These represent the maximum and minimum reference values ​​for the indicator, which fall within a pre-defined reasonable range. These values ​​can be based on historical operational data, such as the highest and lowest values ​​recorded for the indicator in the past year; theoretical values ​​or design standards, such as a theoretical maximum removal rate of 99%; or expert experience or industry regulations, such as a requirement for effluent turbidity to be below 5 NTU. The setting of these two values ​​directly determines the rationality and sensitivity of the scoring and serves as the benchmark for the model. This represents the normalized index value. The initial index value x is the result after dimensionless and standardized processing, and its value always lies between [0, 1]. The comprehensive scoring formula for preprocessing performance is:

[0032] in, This represents the preprocessing performance index at time t, with a score of 0-100, where higher is better. This represents the dynamic weight of the i-th indicator at time t. This represents the normalized value of the i-th index at time t. The weights are not fixed, but are dynamically adjusted according to operational goals and water quality characteristics: if the COD concentration of the influent increases sharply, the system automatically increases the weight of the "treatment efficiency" indicator; if electricity costs enter peak periods, the system automatically increases the weight of the "economic efficiency" indicator; if subsequent biological units fluctuate, the system automatically increases the weight of the "stability" indicator; the dynamic weights are generated in real time by a small expert rule base or reinforcement learning model.

[0033] In this way, a method for pretreating printing wastewater in cardboard packaging can be achieved.

[0034] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications 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 this application, and should all be included within the protection scope of this application.

Claims

1. A method for pretreatment of printing wastewater in cardboard packaging, characterized in that, The method includes: Step S1: Collect and analyze the characteristics of printing wastewater separately, collect wastewater of different concentrations separately, and ensure that the wastewater entering the equalization tank is homogeneous and of uniform quantity. Step S2: The pretreatment technology selection automatically and intelligently decides on the treatment technology based on the water quality data obtained from the separate collection and real-time analysis. Step S3, Physicochemical Enhanced Pretreatment, receives instructions from the pretreatment technology steps and is responsible for the intensive treatment of wastewater with different concentrations and recalcitrant degradation. Step S4: Based on microbial enhanced pre-biochemical treatment, intelligent algorithms drive enhanced chemical pretreatment to remove and transform chemical pollutants in advance. Step S5, preprocessing effect evaluation, integrates indicators from multiple dimensions into a comprehensive evaluation score for intuitive evaluation of overall performance.

2. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 1, characterized in that: Step S1 further includes the following sub-steps: S1-1, refined wastewater collection, printing wastewater is divided into three levels according to concentration. The first level, high-concentration wastewater, includes ink tray cleaning, ink tank cleaning, ink roller cleaning, waste ink, and dampening solution that is replaced periodically. The second level, medium-concentration wastewater, includes periodic rinsing water for printing press cylinders and components, and plate washing wastewater. The third level, low-concentration wastewater, comes from workshop floor cleaning water, external equipment wiping and cleaning, and cooling system wastewater, which is mixed with dripping ink, lubricating oil, dust, and contains mud and impurities from the ground. S1-2, multi-dimensional characteristic analysis and data management, enables online real-time monitoring. Each collection pipe and main collection pipe is equipped with an online multi-parameter water quality sensor array, integrated into a flow tank. The water quality sensor array integrates pH, conductivity, and turbidity sensors to quickly monitor real-time acidity and alkalinity and calculate total dissolved solids content and suspended solids concentration. Miniaturized UV-Vis spectral water quality sensors provide real-time continuous monitoring, and the built-in processor processes spectral data in real time. The algorithm model directly outputs COD estimates and color values. Sensor data is acquired by a PLC and uploaded to the central SCADA control platform system. S1-3. Sampling of wastewater of different categories is carried out regularly and standard method analysis is performed. The laboratory is equipped with a COD rapid analyzer, spectrophotometer, and ICP-MS. Laboratory data is used as the standard true value for calibration and establishment of online sensor prediction models. A wastewater archive database is established and stored and managed using an SQL database or time series database. Electronic archives are established for different batches of high-concentration waste liquid and each medium- and low-concentration wastewater.

3. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 2, characterized in that: In step S1, the intelligent homogenization and conditioning system establishes multiple treatment tanks to treat different types and batches of printing wastewater. It adopts a combination of perforated aeration pipes and submersible agitators to prevent sedimentation and avoid VOCs volatilization. The central control system calculates the weighted average water quality based on the wastewater records of each tank and simulates the possible water quality after mixing. Based on the simulation results, it intelligently decides the opening sequence and duration of the pump valve linkage system and pumps wastewater of different concentrations into the conditioning tank according to a preset ratio to control the fluctuation of the influent water quality from the source. The system uses historical data to train a simple time series model to predict future water quality trends. Based on the prediction results, it can start the stirring device or adjust the pumping scheme in advance to actively pre-homogenize. Online sensors at the outlet of the equalization tank monitor the water quality after homogenization in real time and send the estimated COD data to the subsequent enhanced treatment unit in advance. The treatment unit can then pre-adjust the reagent dosing formula and dosage to achieve process synergistic optimization.

4. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 1, characterized in that: Step S2 further includes the following sub-steps: S2-1, based on real-time data and historical databases, dynamically generates the optimal treatment plan, constructs a data-driven, rule-based and predictive model-based dynamic decision-making system, realizes the precision, adaptability and cost reduction and efficiency improvement of the treatment process, determines the treatment route of different batches of wastewater, and provides early warning and control suggestions for subsequent enhanced pretreatment steps. S2-2 acquires data from sensors and PLC. The system receives data streams from the separation collection and characteristic analysis adjustment steps in real time. The time-series database efficiently stores and queries time-stamped water quality data. The built-in data cleaning algorithm removes outliers caused by momentary sensor failures. S2-3: Extract key features for decision-making from raw data, perform detailed feature extraction and water quality assessment, and calculate the BOD5 / COD ratio in real time using the calculation module. B / C = k1 * UV254 + k2 * BOD; Where k1 and k2 are coefficients obtained through regression of historical data, COD is chemical oxygen demand, which represents the amount of oxidant consumed by the oxidation of organic matter in water under the action of strong oxidants, BOD is biochemical oxygen demand, BOD5 is five-day biochemical oxygen demand, which represents the amount of dissolved oxygen consumed by microorganisms in decomposing organic matter in water within 5 days at 20℃, and the B / C ratio directly reflects the proportion of biodegradable organic matter to total organic matter.

5. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 4, characterized in that: In step S2, based on the extracted features, an innovative algorithm is used to perform intelligent technology selection and parameter optimization, and a machine learning-based random forest classification and regression model is constructed to predict the most suitable processing technology based on the input features. Traditional chemical dosing calculations are mostly static formulas, which cannot adapt to dynamic changes in water quality. An adaptive Fenton chemical dosing estimation model is introduced, employing a feedback correction factor λ to dynamically adjust the dosing amount, giving the model self-learning capabilities. ； ； in, This represents the amount of hydrogen peroxide that needs to be added at the current time t. This represents the dynamic correction factor at the current moment, where α represents forgetting, used to balance the influence of historical experience and the latest data. This indicates the COD removal rate achieved in the actual treatment of the previous batch of water. This indicates the COD removal rate of the previous batch of water based on the model prediction. The system predicts the dosage based on the initial model and performs the treatment, analyzes the actual effluent COD, calculates the actual removal rate, compares the actual value with the predicted value, calculates the efficiency ratio, and uses this ratio to update the correction factor λ(t).

6. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 1, characterized in that: In step S3, instructions are received from the pretreatment technology step, which is responsible for the intensive treatment of high-concentration, recalcitrant wastewater. The electro-Fenton process generates H2O2 and Fe in situ through electrolysis. 2+ Wastewater from the equalization tank is fed into the inlet tank of the electro-Fenton reactor. Upon receiving instructions from the upper level, an intelligent pH controller automatically adds acid to precisely adjust the pH to the optimal reaction range. Based on data from the online COD meter, a small amount of external Fe is added as instructed. 2+ As a catalytic supplement; Wastewater flows into the Fenton electro-reactor, which contains an anode and a cathode. When direct current is applied, the cathode continuously reduces O2 to H2O2; the anodic oxidation produces Fe... 2+ It forms Fenton's reagent with H2O2, generating a large number of -OH free radicals, which efficiently degrade organic matter. It fills the space between the anode and cathode with particulate conductive medium, forming countless micro-electrolysis units, which greatly increases the reaction area and efficiency. After the reaction, the wastewater enters the neutralization tank, where alkali is added to adjust the pH, and Fe... 2+ to become Fe 3+ The process involves forming ferric hydroxide flocs, which effectively adsorb, trap, and co-precipitate oxidized organic matter and colloids. The wastewater then enters an inclined plate sedimentation tank or a high-efficiency clarification tank for mud-water separation. Based on changes in influent COD, the current density I of the electro-Fenton reactor is dynamically adjusted to minimize energy consumption while ensuring treatment effectiveness. ； Among them, I actual K indicates direct control of power consumption. p k i k d These represent the proportional, integral, and derivative coefficients of the PID controller, respectively, and require tuning. e(t) = COD in (t)-COD setpoint This indicates the deviation between the current influent COD and the set target COD. This represents the integral of the deviation from time 0 to time t, which represents the cumulative total of all deviations throughout history. The derivative of the deviation represents the instantaneous rate of change of the deviation at the current moment. It is the slope of the deviation-time curve at that point, reflecting the speed and direction of the deviation change. The system monitors the influent COD(t) in real time and compares it with the set desired effluent COD. setpoint The deviation e(t) is calculated by comparison. Based on the magnitude, duration, and trend of the deviation, the PID algorithm outputs a control signal in real time to dynamically adjust the output current I of the DC power supply. actual This achieves precise energy-saving control of electricity consumption.

7. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 1, characterized in that: Step S4 further includes the following sub-steps: S4-1, the effluent after physical and chemical treatment first enters the nutrient conditioning tank. The system automatically calculates and adds nitrogen and phosphorus sources based on the data from the online analyzer, and precisely controls the C, N and P ratio. In S4-2, wastewater enters the hydrolysis acidification reactor, which is filled with specific compound bacterial agents and enzyme preparations. The reactor is equipped with a biological packing zone, where the combined packing forms a high-biomass biofilm. Under anoxic conditions, microorganisms and enzymes completely convert small-molecule organic matter and some recalcitrant substances into volatile fatty acids. The reactor is equipped with a high-flow-rate internal circulation pump to return the mud-water mixture to the inlet to maintain a high biological concentration. A mud-water separation zone is set at the rear end, where sludge is automatically returned to maintain a high sludge concentration in the reactor.

8. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 7, characterized in that: The concentration and composition of VFAs in the effluent from the hydrolysis acidification tank are monitored in real time, and the hydraulic retention time (HRT) and internal circulation ratio (R) are adjusted based on feedback to ensure that the system is in the most efficient acid production state. This is based on a microbial activity feedback control algorithm that monitors VFAs in real time. ； in, Indicates the total VFAs concentration in the effluent. These represent the total organic carbon concentrations in the influent and effluent, respectively. This represents the VFAs yield index, which reflects the amount of VFAs converted per unit of TOC removal. A higher index indicates higher acid production efficiency and healthier operation of the system. The system calculates the VFA-PI in real time and compares it with the set target value. If the measured value is lower than the target, it indicates a decrease in acid production efficiency, and the hydraulic calculation time should be adjusted accordingly. ； in, This represents the calculated new hydraulic residence time. Indicates the current hydraulic residence time. The target yield index is determined using historical best performance data. This represents the yield index calculated through real-time monitoring. This represents the gain control coefficient. The algorithm automatically extends the HRT by reducing the influent flow rate, giving the microorganisms a longer reaction time; conversely, it appropriately shortens the HRT to improve treatment efficiency. At the same time, the internal circulation ratio is also adjusted in conjunction with the VFAs concentration. When the concentration is low, the circulation ratio is increased to improve mass transfer efficiency.

9. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 1, characterized in that: Step S5 further includes the following sub-steps: S5-1, the system uses ETL tools to extract data from sensors, PLCs, and control systems of all preceding units, transforms and loads the data, and finally stores the data in the real-time database. During this process, data cleaning and timing alignment are performed to eliminate errors caused by different sensor data acquisition frequencies and delays. Information such as wastewater source, initial water quantity, water quality (COD), and color is obtained from the wastewater collection unit, and real-time energy consumption, H2O2, and Fe2+ are obtained from the physicochemical enhancement unit. 2+ Actual dosage, pH / ORP curve, and effluent COD data were obtained from the microbial pre-biochemical unit, including influent and effluent BOD, VFA concentration and composition, nutrient dosage, HRT, and internal circulation ratio. S5-2, based on the cleaned data, calculate the treatment efficiency indicators, including total COD removal rate, total color removal rate, B / C ratio improvement rate, and VFAs yield; it is also necessary to calculate economic indicators, process stability indicators, and ecotoxicity reduction indicators.

10. The method for pretreatment of printing wastewater in cardboard packaging as described in claim 9, characterized in that: The above-mentioned multiple dimensions of indicators are integrated into a single comprehensive score, PPI, for intuitive evaluation of overall performance and automatic diagnosis of problems. First, the indicators of different dimensions are normalized to the [0, 1] interval. The comprehensive preprocessing performance scoring formula is as follows: ； in, This represents the preprocessing performance index at time t, with a score of 0-100, where higher is better. This represents the dynamic weight of the i-th indicator at time t. This represents the normalized value of the i-th indicator at time t. The weights are not fixed but are dynamically adjusted according to operational objectives and water quality characteristics.

Images