An online micro-reaction simulator-based old kiln water dosing pre-judgment intelligent control system and method

Abstract

The present application belongs to the technical field of industrial water treatment, and particularly relates to an intelligent control system and method for pre-judgment of old kiln water before dosing based on an online micro reaction simulator. A multi-parameter monitoring module collects the water inflow, pH value and iron ion concentration in real time. The micro reaction simulator performs simulation pre-reaction experiments on the real-time water sample diverted through a bypass pipeline, and obtains the dynamic response curve of the pH value and iron ion concentration of the water sample. The control module determines the optimal dosing amount according to the curve, and generates a feedforward dosing control instruction by weighting and correcting the real-time water inflow parameters, which is executed by an automatic dosing execution module. The effluent feedback correction module collects the actual effluent quality and feeds it back to the control module, which compensates and corrects the feedforward instruction according to the deviation between the effluent index and the target value. The single experiment duration of the micro reaction simulator is less than the hydraulic retention time of the old kiln water from the dosing point to the effluent monitoring point, so that the feedforward instruction is output before the corresponding water sample reaches the dosing point.

Description

Technical Field

[0001] This invention belongs to the field of industrial water treatment technology, specifically relating to an intelligent control system and method for predicting the pre-dosing of chemicals in old kiln water based on an online micro-reaction simulator. Background Technology

[0002] Old mine wastewater is typical mine wastewater accumulated in coal mine goaf areas. It is generally characterized by large fluctuations in water quality, unstable pH, and excessive iron ion content. The pH value of the influent often changes drastically within the acidic range, and the iron ion concentration can fluctuate by several to tens of times. Currently, most old mine wastewater treatment plants in China use manual, experience-based dosing or fixed-level quantitative dosing. Operators rely solely on experience to set fixed dosages. When the influent water quality is low, excessive dosing leads to waste; when the influent water quality exceeds standards, insufficient dosing results in substandard effluent. The few existing plants equipped with online monitoring still rely on effluent quality feedback correction for control logic. This means they wait for the chemicals to mix and react with the water in the main process pipeline and flow to the effluent monitoring point before adjusting the subsequent dosing based on measured effluent indicators. This is a typical lag control mode and cannot fundamentally solve the problem of excessive discharge during periods of water quality fluctuation.

[0003] There is a hydraulic residence time of several minutes to tens of minutes between the dosing point and the effluent monitoring point for wastewater from old kilns. When the pH or iron ion concentration of the effluent exceeds the standard, the substandard water has already flowed past the monitoring point. Feedback correction commands have no effect on the discharged wastewater, which is an inherent defect of the existing feedback control mode. In addition, traditional dosing systems lack the ability to sense the physical state of the dosing execution unit. Abnormalities such as blockage of dosing pipelines and inaccurate metering cannot be detected in time, resulting in a disconnect between commands and execution. Regulatory authorities lack remote monitoring means and can only rely on periodic on-site inspections. They cannot grasp the influent water quality, dosing conditions, effluent indicators and equipment operating status of each site in real time, resulting in incomplete and seriously lagging regulatory coverage. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent control system and method for predicting the addition of chemicals to old kiln water based on an online micro-reaction simulator, in order to solve the technical problems of pure lag caused by the hydraulic residence time in existing old kiln water chemical addition control systems that rely on effluent feedback correction, resulting in irreversible damage to the water exceeding the standard and a lack of supervision.

[0005] The present invention achieves the above objectives through the following technical solutions: Firstly, this invention proposes an intelligent control system for predicting the pre-dosing of chemicals in old kiln water based on an online micro-reaction simulator. This system is applied to old kiln water treatment scenarios, where the old kiln water is introduced into the treatment station via an inlet pipe, and the treated effluent is discharged via an outlet pipe. The system includes: A multi-parameter monitoring module is deployed in the inlet pipe to collect the instantaneous flow rate, pH value, and iron ion concentration of the inlet water in real time. A miniature reaction simulator is connected to the inlet pipe via a bypass pipeline to divert real-time water samples and conduct simulated pre-reaction experiments, so as to monitor the dynamic response curve of the pH value and iron ion concentration of the water sample in real time. The control module is used to determine the optimal dosage corresponding to the real-time water sample based on the dynamic response curve, and to correct the optimal dosage with the real-time influent parameters as weights, thereby generating a feedforward dosing control command. An automatic dosing execution module, deployed in the dosing pipeline, is used to respond to the feedforward dosing control command and execute the dosing of the agent; The water outlet feedback correction module is deployed in the water outlet pipe to collect the water outlet pH value and the water outlet iron ion concentration in real time and feed them back to the control module. The control module is also used to correct and compensate the feedforward dosing control command based on the deviation between the effluent pH value and the effluent iron ion concentration and the target value.

[0006] Furthermore, the micro-reaction simulator includes a micro-reaction vessel, a precision dosing unit, and an online detection unit; The micro-reaction vessel is connected to the inlet pipe via a bypass pipeline and has a built-in stirrer for containing the real-time water sample obtained from the diversion. The precision dosing unit is used to quantitatively add reagents into the micro-reaction vessel to simulate the main process dosing procedure; The online detection unit is used to monitor the dynamic response curves of pH value and iron ion concentration of the water sample in the micro-reaction vessel in real time.

[0007] Furthermore, the sampling frequency of the multi-parameter monitoring module is configured to match the fluctuation rate of the old kiln water quality in order to capture the dynamic changes in the influent water quality; the completion time of a single simulated pre-reaction experiment of the micro-reaction simulator is less than the hydraulic residence time of the old kiln water flowing from the dosing point to the effluent monitoring point, so that the feedforward dosing control command is output before the corresponding water sample arrives at the dosing point.

[0008] Furthermore, the control module determines the optimal dosage for the real-time water sample based on the dynamic response curve, specifically including: The control module extracts features from the dynamic response curve to obtain the steady-state pH value, steady-state iron ion concentration, and steady-state arrival time; the steady-state pH value and steady-state iron ion concentration are the values ​​corresponding to the changes in pH value and iron ion concentration in the dynamic response curve being less than preset thresholds in multiple consecutive sampling periods, respectively. The control module compares the steady-state pH value and steady-state iron ion concentration with preset target values ​​to determine the treatment effect under the current simulated dosage. If the treatment effect does not meet the standard, the control module adjusts the simulated dosage of the precision dosing unit and starts the next round of simulated pre-reaction experiment until the treatment effect meets the standard. If the processing effect meets the standard, the control module will use the current simulated dosage as the candidate optimal dosage.

[0009] Furthermore, the control module uses a step-by-step approximation method to adjust the simulated dosage of the precision dosing unit: The baseline dosage matching the current influent pH and iron ion concentration is retrieved from the built-in old kiln water dosing threshold library. Multiple sets of simulated pre-reaction experiments with different gradients are performed with the baseline dosage as the center to obtain the dosage range in which the steady-state pH and steady-state iron ion concentration both meet the preset compliance conditions. The lower limit and upper limit of this range are used as the lower limit and upper limit of the compliance critical range, respectively. Within the critical range of compliance, the dosage of the experimental point is calculated iteratively in round by round using an interval narrowing optimization algorithm. Based on the compliance judgment results of each round of experiments, the boundary of the critical range of compliance is narrowed round by round until the width of the critical range of compliance is less than the minimum dosing accuracy of the precision dosing unit, at which point the iteration terminates and the midpoint of the current range is taken as the critical dosage. The current model confidence level is obtained. The model confidence level is dynamically determined by the historical feedback correction amplitude and correction frequency of the effluent feedback correction module. A safety factor negatively correlated with the model confidence level is superimposed on the critical dosage to generate the final simulated dosage.

[0010] Furthermore, the control module adjusts the optimal dosage using the real-time influent parameters as weights, specifically including: The control module uses the instantaneous influent flow rate. As the sovereign entity, the feedforward dosing command value is calculated according to the following formula. : ,in, The optimal dosage per unit flow rate. This is the comprehensive water quality correction factor. Determined by the following formula:

[0011] in, and These are reference values ​​for pH and iron ion concentration, respectively. and These are the real-time influent pH value and the real-time influent iron ion concentration, respectively. and These are the correction factors for pH deviation and iron ion concentration deviation, respectively.

[0012] Furthermore, the control module corrects and compensates the feedforward dosing control command based on the deviation between the effluent pH value and the effluent iron ion concentration and the target value, specifically including: The control module calculates the pH deviation of the water. Deviation in effluent iron ion concentration ; The control module uses a PID control algorithm to calculate the feedback compensation amount. :

[0013] in, This is the overall deviation vector. , , These are the proportional coefficient, integral coefficient, and differential coefficient, respectively. The control module calculates the final dosing control command according to the following formula. : .

[0014] Furthermore, the control module is also used to adaptively update the PID control parameters of the effluent feedback correction module according to the fluctuation range of the influent water quality. The control module constructs a three-dimensional wave feature vector. ,in These represent the rates of change of pH value, iron ion concentration, and flow rate within adjacent sampling periods of the multi-parameter monitoring module. The control module counts the N consecutive sampling periods preceding the current sampling time. Standard deviation of each dimension , , and with As a comprehensive volatility indicator; The control module has multiple pre-set PID parameter combinations corresponding to different fluctuation levels. Each PID parameter combination includes a first set of parameters under high fluctuation conditions and a second set of parameters under low fluctuation conditions. The proportional coefficient and derivative coefficient in the first set of parameters are both greater than the corresponding coefficients in the second set of parameters, and the integral coefficient in the first set of parameters is less than the integral coefficient in the second set of parameters. The control module uses a comprehensive fluctuation index To index and query the PID parameter combinations, the proportional coefficient under the current operating condition is determined. Integral coefficient and differential coefficients and determined by matching , , Update the PID control law of the effluent feedback correction module.

[0015] Secondly, this invention proposes a pre-treatment intelligent control method for old kiln water based on an online micro-reaction simulator. This method is executed using the aforementioned pre-treatment intelligent control system for old kiln water, and includes the following steps: S1. The multi-parameter monitoring module collects the instantaneous flow rate, pH value, and iron ion concentration of the influent in real time. S2. The micro-reaction simulator diverts real-time water samples from the inlet pipe through a bypass pipeline to perform a simulated pre-reaction experiment and monitors the dynamic response curve of the pH value and iron ion concentration of the water sample in real time. S3. The control module receives the real-time influent parameters and the dynamic response curve, determines the optimal dosage corresponding to the real-time water sample based on the dynamic response curve, and corrects the optimal dosage with the real-time influent parameters as weights, generating a feedforward dosing control command. S4. The automatic dosing execution module responds to the feedforward dosing control command and executes the main process reagent dosing; S5. The effluent feedback correction module collects the effluent pH value and effluent iron ion concentration in real time and feeds them back to the control module. S6. The control module corrects and compensates the feedforward dosing control command based on the deviation between the pH value and the iron ion concentration of the effluent and the target value.

[0016] Furthermore, in step S2, the micro-reaction simulator performing the simulated pre-reaction experiment includes the following steps: S21. The precision dosing unit quantitatively adds reagents to the real-time water sample in the micro-reaction vessel; S22. The stirrer is set to a preset speed to ensure that the water sample and the reagent are fully mixed and reacted. S23. The online detection unit continuously collects the pH value and iron ion concentration of the water sample during the reaction process and generates the dynamic response curve. S24. The automatic cleaning unit performs cleaning and decontamination of the micro-reaction vessel after the experiment.

[0017] The beneficial effects of this invention are as follows: 1. This invention utilizes an online micro-reaction simulator to conduct advanced pre-reaction experiments on real-time influent, enabling the dosing system to determine the optimal dosing scheme for the current batch of water before the chemicals reach the main process dosing point. This eliminates the pure lag problem caused by hydraulic residence time in traditional feedback control modes. This invention uses feedforward prediction to handle the main quantity regulation task, reducing feedback correction to micro-level correction. Regardless of changes in influent water quality, the dosing scheme is generated based on the measured reaction characteristics of the batch of water samples, completely resolving the control failure and excessive discharge risks caused by lag. Simultaneously, the micro-reaction simulator employs a bypass water intake design, avoiding interference with the main process operation and adapting to the site and power supply conditions of existing old water treatment plants.

[0018] 2. This invention utilizes a step-by-step approximation method to rapidly converge to the optimal dosage within a limited number of experimental rounds. Combined with feedforward dosing and effluent feedback correction, it ensures consistently stable effluent compliance while avoiding redundant chemical dosing. The control module senses the amplitude of influent water quality fluctuations in real time based on a three-dimensional fluctuation feature vector, adaptively adjusting the PID control parameters of the effluent feedback correction module. It prioritizes response speed under high-fluctuation conditions and maintains steady-state accuracy under stable conditions, ensuring the system maintains excellent control quality even in complex water quality fluctuation scenarios. Attached Figure Description

[0019] Figure 1 This is a block diagram of the overall architecture of an intelligent control system for predicting chemical addition to old kiln water based on an online micro-reaction simulator, according to one embodiment of the present invention. Figure 2 This is a block diagram of the internal structure of a micro-reaction simulator in one embodiment of the present invention; Figure 3 This is a block diagram of the functional units of the control module in one embodiment of the present invention; Figure 4 This is a flowchart of a method for intelligent control of old kiln water pre-dosing based on an online micro-reaction simulator, according to one embodiment of the present invention. Detailed Implementation

[0020] The present application will now be described in further detail with reference to the accompanying drawings. It should be noted that the following specific embodiments are only used to further illustrate the present application and should not be construed as limiting the scope of protection of the present application. Those skilled in the art can make some non-essential improvements and adjustments to the present application based on the above application content.

[0021] First Embodiment Please combine Figure 1As shown, an intelligent control system for predicting pre-dosing of wastewater from old kilns, based on an online micro-reaction simulator, is applied to wastewater treatment scenarios in old kilns. Wastewater from old kilns is introduced into the treatment station via an inlet pipe, and the treated effluent is discharged via an outlet pipe. The system includes a multi-parameter monitoring module, a micro-reaction simulator, a control module, an automatic dosing execution module, and an effluent feedback correction module.

[0022] The multi-parameter monitoring module is deployed in the inlet water pipeline to collect the instantaneous flow rate, pH value, and iron ion concentration of the inlet water in real time.

[0023] As an example, the influent flow sensor uses an electromagnetic flow meter, whose measurement accuracy is unaffected by suspended solids and iron ion precipitation in the water. The influent pH sensor employs a composite electrode structure, with its reference electrode designed to prevent clogging. The influent iron ion concentration sensor uses an online spectrometer to determine the total iron content by measuring absorbance at a specific wavelength.

[0024] The miniature reaction simulator is connected to the inlet pipe via a bypass pipeline. It is used to divert real-time water samples and conduct simulated pre-reaction experiments to monitor the dynamic response curves of the water sample's pH value and iron ion concentration in real time. The bypass pipeline is equipped with a sampling valve and a flow regulating valve. The sampling valve controls the opening and closing of the water sample introduction. The flow regulating valve adjusts the diversion flow rate of the bypass pipeline, ensuring that the water sample flow rate entering the miniature reaction simulator remains stable within a preset range.

[0025] Furthermore, in combination Figure 2 As shown, the micro-reaction simulator includes a micro-reaction vessel, a precision dosing unit, an online detection unit, and an automatic cleaning unit.

[0026] The miniature reaction vessel is connected to the inlet water pipe via a bypass line and has a built-in stirrer to hold the real-time water sample obtained from the diversion. Optionally, the miniature reaction vessel is made of transparent, acid and alkali resistant material with a conical bottom to facilitate the discharge of precipitates and cleaning after the reaction.

[0027] The precision dosing unit is used to quantitatively add chemicals into a miniature reaction vessel to simulate the main process dosing procedure. Specifically, the precision dosing unit includes a high-precision miniature metering pump and multiple dosing lines. Each dosing line corresponds to a different type of water treatment chemical, and each line is independently controlled, allowing for the addition of different types and proportions of chemical combinations into the miniature reaction vessel.

[0028] The online detection unit is used to monitor the dynamic response curves of pH and iron ion concentration in water samples within a microreactor in real time. The online detection unit includes a micro pH electrode and a micro iron ion selective electrode. These electrodes are integrated at the sampling point in the lower middle part of the microreactor. The electrode output signals are amplified and filtered by a signal conditioning circuit, and then converted into digital signals by an analog-to-digital converter before being transmitted to the control module. As an example, the online detection unit is configured to detect at a frequency of more than once per second, enabling it to capture the complete evolution of pH and iron ion concentration in the water sample over time after reagent addition.

[0029] The automatic cleaning unit performs cleaning and wastewater discharge operations on the miniature reaction vessel after each simulated pre-reaction experiment. The automatic cleaning unit includes a water rinsing line, a compressed air purging line, and a wastewater discharge line. The water rinsing line injects clean water into the miniature reaction vessel, which, combined with high-speed stirring by the agitator, rinses the inner wall of the vessel and the surface of the sensor probe. The compressed air purging line introduces compressed air into the cleaned miniature reaction vessel to remove residual water droplets. The wastewater discharge line discharges the cleaning waste liquid and purging residue to a designated recycling system. Through the intervention of the automatic cleaning unit, the miniature reaction vessel is restored to a clean state after each simulated pre-reaction experiment, avoiding cross-interference from residual water samples and sediments in the water quality parameter measurements of the next simulated pre-reaction experiment.

[0030] Combination Figure 3 As shown, the control module is communicatively connected to the multi-parameter monitoring module and the micro-reaction simulator. The control module determines the optimal dosage for the real-time water sample based on the dynamic response curve, and corrects the optimal dosage using real-time influent parameters as weights, generating feedforward dosing control commands. Optionally, the control module uses an industrial-grade programmable logic controller or an embedded industrial computer, integrating a data acquisition card and communication interface. The data acquisition card receives sensor signals from the multi-parameter monitoring module and the micro-reaction simulator, and the communication interface interacts with the automatic dosing execution module and the effluent feedback correction module via fieldbus or industrial Ethernet.

[0031] An automatic dosing execution module is deployed in the dosing pipeline to respond to feedforward dosing control commands and execute pesticide dosing. Specifically, the automatic dosing execution module includes a variable frequency dosing pump, a pesticide storage tank, a precision metering valve, and a pipeline solenoid valve. The motor drive frequency of the variable frequency dosing pump is determined by the feedforward dosing control command output from the control module. By adjusting the motor speed, the plunger movement frequency of the dosing pump is changed, thereby controlling the pesticide output flow rate. The precision metering valve is located in the dosing pump outlet pipeline and performs secondary fine-tuning of the pesticide flow rate by controlling the valve core opening. The pipeline solenoid valve is located downstream of the metering valve and is used to quickly cut off the dosing path in the event of system shutdown or abnormal operating conditions.

[0032] The effluent feedback correction module is deployed in the effluent pipeline to collect effluent pH and iron ion concentration data in real time and feed them back to the control module. The effluent feedback correction module includes an effluent pH sensor and an effluent iron ion concentration sensor, and its selection and configuration principles are consistent with those of the multi-parameter monitoring module. The sensors of the effluent feedback correction module are installed at the monitoring point on the effluent pipeline, which is located downstream of the main process chemical dosing point. The pipeline length between the monitoring point and the chemical dosing point corresponds to the hydraulic residence time of the old kiln water flowing from the chemical dosing point to the monitoring point. Due to the hydraulic residence time, the water quality parameters collected by the effluent feedback correction module reflect the actual treatment effect of the chemical dosing operation several minutes to tens of minutes ago, rather than the effect of the current chemical dosing operation; therefore, the feedback signal has a natural lag.

[0033] The control module also corrects and compensates for deviations in the effluent pH and iron ion concentration from target values. The control module stores target effluent water quality values, including the target pH range and the upper limit for iron ion concentration. It compares the actual effluent pH value collected by the effluent feedback correction module with the target range and the actual effluent iron ion concentration with the upper limit, calculating the deviation. This deviation serves as the basis for feedback compensation, making minor corrections to the feedforward dosing control command at the next moment to compensate for system deviations introduced by factors such as model prediction errors, sensor drift, and actuator wear.

[0034] Furthermore, the sampling frequency of the multi-parameter monitoring module is configured to match the fluctuation rate of the old kiln water quality to capture dynamic changes in the influent water quality. The completion time of a single simulated pre-reaction experiment in the micro-reaction simulator is less than the hydraulic residence time of the old kiln water flowing from the dosing point to the effluent monitoring point, so that the feedforward dosing control command is output before the corresponding water sample arrives at the dosing point. Specifically, the control module statistically analyzes the fluctuation spectrum of water quality parameters based on historical data from the multi-parameter monitoring module. When water quality fluctuations are severe, the sampling frequency is automatically increased to more than once per second. The completion time of a single simulated pre-reaction experiment includes the sum of the water sample introduction time, the reagent mixing reaction time, the data acquisition and judgment time, and the cleaning and sewage discharge time, and this total time is controlled to be less than the hydraulic residence time. Through the above parameter matching, the system completes at least one complete simulated pre-reaction experiment within each hydraulic residence time window and outputs the feedforward command to the automatic dosing execution module.

[0035] Furthermore, the control module determines the optimal dosage for the real-time water sample based on the dynamic response curve, specifically including the following process.

[0036] The control module extracts features from the dynamic response curve to obtain the steady-state pH value, steady-state iron ion concentration, and steady-state arrival time. The steady-state pH value and steady-state iron ion concentration are the values ​​corresponding to the changes in pH value and iron ion concentration in the dynamic response curve being less than preset thresholds over multiple consecutive sampling periods, respectively.

[0037] Specifically, the control module records the pH value and iron ion concentration changes over time, with the time of reagent addition as the zero point. The control module then uses a sliding window method to differentiate the two curves and calculate the instantaneous rate of change at each sampling moment.

[0038] In this embodiment, when the rate of change of pH value is less than 0.02 pH units per second for five consecutive sampling cycles, and the rate of change of iron ion concentration is less than 0.5 mg / L for five consecutive sampling cycles, the control module determines that the reaction has reached a steady state, records the current time as the steady-state arrival time, records the current pH value as the steady-state pH value, and records the current iron ion concentration as the steady-state iron ion concentration. Through the above determination rules, the system can adaptively determine the reaction endpoint for different water quality conditions, avoiding excessive waiting or premature termination caused by a fixed reaction time.

[0039] The control module compares the steady-state pH value and steady-state iron ion concentration with preset target values ​​to determine the treatment effect under the current simulated dosage. The preset target values ​​include the lower and upper limits of the effluent pH target and the upper limit of the effluent iron ion concentration. If the steady-state pH value falls within the target range and the steady-state iron ion concentration is less than the upper limit, the treatment effect is considered satisfactory. If the steady-state pH value is lower than the lower target limit or higher than the upper target limit, or the steady-state iron ion concentration is greater than the upper limit, the treatment effect is considered unsatisfactory.

[0040] If the treatment effect does not meet the standard, the control module adjusts the simulated dosage of the precision dosing unit and starts the next round of simulated pre-reaction experiment until the treatment effect meets the standard. If the treatment effect meets the standard, the control module uses the current simulated dosage as the candidate optimal dosage.

[0041] Furthermore, the control module uses a step-by-step approximation method to adjust the simulated dosage of the precision dosing unit.

[0042] The system retrieves a baseline dosage from a built-in database of chemical dosing thresholds for old kiln water that matches the current influent pH and iron ion concentration. Multiple sets of simulated pre-reaction experiments with different gradients are then performed using this baseline dosage as the center. A dosage range where both steady-state pH and steady-state iron ion concentration meet the preset compliance conditions is obtained. The lower and upper limits of this range are used as the lower and upper limits of the critical compliance range, respectively.

[0043] Within the critical range for achieving the target, an interval narrowing optimization algorithm is used to iteratively calculate the dosage at each experimental point. Based on the target achievement results of each round of experiments, the boundary of the critical range for achieving the target is narrowed round by round until the width of the critical range for achieving the target is less than the minimum dosing accuracy of the precision dosing unit. The iteration then terminates, and the midpoint of the current range is taken as the critical dosage.

[0044] In this embodiment, the first stage uses a baseline dosage as the center and executes multiple sets of simulated pre-reaction experiments with dosages less than, equal to, and greater than the baseline dosage to quickly determine the approximate location of the critical range for achieving the target. This stage employs a large-step, coarse-grained experimental strategy, prioritizing search speed. The second stage uses the golden section method for iterative iterations within the critical range for achieving the target. In each iteration, the golden section point within the range is selected for simulated pre-reaction experiments. Based on the experimental results, the range boundary is narrowed, gradually approaching the critical dosage that just achieves the target treatment effect. This stage employs a small-step, fine-grained search strategy, prioritizing optimization accuracy. Through these two-stage, hierarchical progressions, the system simultaneously considers search speed and optimization accuracy within a limited number of experimental rounds.

[0045] Obtain the current model confidence level. The model confidence level is dynamically determined by the historical feedback correction magnitude and frequency of the effluent feedback correction module. Specifically, the control module records the feedback compensation amount and its direction output by the effluent feedback correction module for each instance. If the feedback compensation amount is consistently positive and the magnitude gradually increases, it indicates that the model prediction value is consistently lower than the actual demand value, and the model confidence level decreases. If the feedback compensation amount fluctuates slightly around zero, it indicates that the deviation between the model prediction value and the actual value is small, and the model confidence level remains high. A safety factor negatively correlated with the model confidence level is superimposed on the critical dosage to generate the final simulated dosage.

[0046] Furthermore, the control module adjusts the optimal dosage using real-time influent parameters as weights, specifically including the following process: The control module uses the instantaneous influent flow rate As the sovereign entity, the feedforward dosing command value is calculated according to the following formula. : ,in, The optimal dosage per unit flow rate. This is the comprehensive water quality correction factor. Determined by the following formula: ; in, and These are reference values ​​for pH and iron ion concentration, respectively. and These are the real-time influent pH value and the real-time influent iron ion concentration, respectively. and These are the correction factors for pH deviation and iron ion concentration deviation, respectively.

[0047] The physical meaning of the above correction relationship is: when the real-time influent pH value is lower than the reference value and acidity increases, A value greater than 1 indicates an increased feedforward dosing command, requiring the addition of more alkaline reagents to neutralize the extra hydrogen ions. When the real-time influent iron ion concentration exceeds the reference value, A value greater than 1 indicates an increase in the feedforward dosing command, requiring the addition of more oxidant to treat the extra iron ions. Conversely, a value less than 1 indicates a decrease in the feedforward dosing command. Through a dual regulation of flow weighting and comprehensive water quality correction, the feedforward dosing command achieves real-time response to changes in both influent flow rate and water quality.

[0048] Furthermore, the control module corrects and compensates the feedforward dosing control command based on the deviation between the effluent pH value and the effluent iron ion concentration and the target value, specifically including the following process.

[0049] The control module calculates the pH deviation of the effluent. Deviation in effluent iron ion concentration pH deviation of effluent This equals the difference between the real-time effluent pH value and the target effluent pH value. Effluent iron ion concentration deviation. It equals the difference between the real-time effluent iron ion concentration and the target value of the effluent iron ion concentration.

[0050] The control module uses a PID control algorithm to calculate the feedback compensation amount. : ; in, For the comprehensive deviation vector, , , These are the proportional coefficient, integral coefficient, and differential coefficient; proportional coefficient Determines the response speed of the feedback compensation amount to the current deviation. Integral coefficient. The differential coefficient determines the strength of the feedback compensation for the accumulation of historical deviations. The amount of feedback compensation determines the predictive adjustment of the deviation change trend.

[0051] The control module calculates the final dosing control command using the following formula. : .

[0052] With the above design, the feedforward section undertakes the vast majority of the dosage adjustment tasks, while the feedback section only makes minor corrections when the feedforward prediction error exceeds the allowable range, thus avoiding over-adjustment and system oscillation caused by the lag characteristics of feedback correction.

[0053] Furthermore, the control module is also used to adaptively update the PID control parameters of the effluent feedback correction module according to the fluctuation range of the influent water quality.

[0054] The control module constructs a three-dimensional wave feature vector. ,in These represent the rates of change of pH value, iron ion concentration, and flow rate within adjacent sampling periods of the multi-parameter monitoring module. It is equal to the absolute value of the difference between the influent pH value at the current sampling time and the influent pH value at the previous sampling time. It is equal to the absolute value of the difference between the influent iron ion concentration at the current sampling time and the influent iron ion concentration at the previous sampling time. It is equal to the absolute value of the difference between the instantaneous inflow rate at the current sampling time and the instantaneous inflow rate at the previous sampling time.

[0055] The control module counts the data from the N consecutive sampling periods preceding the current sampling time. Standard deviation of each dimension , , and with As a comprehensive fluctuation index, the standard deviation is calculated using an unbiased estimation formula, and the value of N is preset based on the on-site water quality fluctuation characteristics. The comprehensive fluctuation index integrates the fluctuation amplitudes of the three dimensions into a scalar value, which is used to quantitatively characterize the overall fluctuation intensity of the current influent water quality.

[0056] The control module has multiple pre-set PID parameter combinations corresponding to different fluctuation levels. Each PID parameter combination includes a first set of parameters for high fluctuation conditions and a second set of parameters for low fluctuation conditions. The proportional coefficient and derivative coefficient in the first set of parameters are both greater than the corresponding coefficients in the second set of parameters, and the integral coefficient in the first set of parameters is less than the integral coefficient in the second set of parameters. The control module uses a comprehensive volatility index To index and query PID parameter combinations, match and determine the proportional coefficient under the current operating condition. Integral coefficient and differential coefficients and determined by matching , , Update the PID control law of the effluent feedback correction module.

[0057] The logic behind the adaptive parameters described above is as follows: Under high-fluctuation conditions, rapid changes in influent water quality lead to frequent and rapid deviations in effluent. Increasing the proportional and derivative coefficients allows the feedback correction module to respond quickly to deviations and their trends, promptly suppressing deviation expansion. Simultaneously, decreasing the integral coefficient prevents excessive accumulation of the integral term due to frequent sign changes in deviations, thus avoiding integral saturation. Under low-fluctuation conditions, decreasing the proportional and derivative coefficients ensures stable system operation, preventing excessive sensitivity to measurement noise and unnecessary adjustment actions.

[0058] According to the above embodiments, this application proposes to conduct advanced pre-reaction experiments on real-time diverted water samples using a bypass-accessed micro-reaction simulator, replacing the traditional hysteresis correction mode that relies on effluent feedback. The system uses a multi-parameter monitoring module to acquire influent flow rate, pH, and iron ion concentration in real time. The micro-reaction simulator obtains the dynamic response curve of the water sample through precise dosing, mixing, and online detection. The control module determines the optimal dosing dosage for the current water sample under simulated conditions based on this curve, and then converts it into a main process dosing command using the actual influent flow rate as a weight. The automatic dosing execution module then executes the dosing. Simultaneously, the effluent feedback correction module collects the actual effluent water quality, and the control module performs minor compensation on the feedforward command based on its deviation from the target value.

[0059] Second Embodiment Combination Figure 4 As shown, a method for predictive intelligent control of chemical addition to old kiln water based on an online micro-reaction simulator is implemented using the intelligent control system as described in the first embodiment, and includes the following steps: S1. The multi-parameter monitoring module collects the instantaneous flow rate, pH value, and iron ion concentration of the influent in real time.

[0060] Specifically, the inlet flow sensor measures the volumetric flow rate of conductive water flowing through the inlet pipe using the principle of electromagnetic induction, and outputs an electrical signal proportional to the flow rate. The inlet pH sensor determines the pH value by measuring the potential difference generated by the activity of hydrogen ions in the water. The inlet iron ion concentration sensor determines the total iron content in the water using spectroscopic or electrochemical methods.

[0061] S2. The micro-reaction simulator diverts real-time water samples from the inlet pipe through a bypass pipeline to perform a simulated pre-reaction experiment and monitors the dynamic response curve of the pH value and iron ion concentration of the water sample in real time. Furthermore, in step S2, the micro-reaction simulator performs a simulated pre-reaction experiment, which includes the following sub-steps.

[0062] S21. The precision dosing unit quantitatively adds chemicals to the real-time water sample inside the micro-reactor. Specifically, the control module retrieves the initial simulated dosing amount from the dosing threshold database based on the current influent water quality parameters and converts this amount into the number of stepper motor drive pulses for the precision dosing unit. The precision dosing unit injects the corresponding volume of chemicals into the micro-reactor according to the number of drive pulses.

[0063] S22. The stirrer operates at a preset speed to ensure thorough mixing and reaction of the water sample and reagent. The stirrer speed is preset by the control module based on the water sample volume and reagent characteristics. The stirrer blades generate axial and radial flow within the micro-reaction vessel, rapidly dispersing and uniformly mixing the reagent in the water sample.

[0064] S23. The online detection unit continuously collects the pH value and iron ion concentration of the water sample during the reaction process, generating dynamic response curves. Starting from the moment the reagent is added, the online detection unit records real-time values ​​of pH value and iron ion concentration at a preset sampling frequency until the reaction reaches steady-state conditions. The recorded discrete data points are arranged in chronological order, forming response curves for pH value versus time and iron ion concentration versus time.

[0065] S24. The automatic cleaning unit performs cleaning and drainage of the miniature reaction vessel after the experiment. Specifically, the automatic cleaning unit first opens the water flushing pipeline to inject water into the miniature reaction vessel. After cleaning the inner wall of the vessel and the surface of the sensor probe with high-speed stirring by the stirrer, the drain valve is opened to discharge the cleaning waste liquid. Finally, the compressed air purging pipeline is opened to introduce compressed air into the miniature reaction vessel to blow away any remaining water droplets.

[0066] S3. The control module receives real-time influent parameters and dynamic response curves, determines the optimal dosage corresponding to the real-time water sample based on the dynamic response curves, and corrects the optimal dosage with real-time influent parameters as weights, generating feedforward dosing control commands.

[0067] Specifically, the control module first extracts the steady-state pH value and steady-state iron ion concentration from the dynamic response curve, and compares them with the preset target values ​​to determine the treatment effect of the current simulated dosage. If the target is not met, the simulated dosage is adjusted using a stepwise approximation method, and steps S21 to S23 are repeated until the target is met. If the target is met, the current simulated dosage is used as the candidate optimal dosage. Subsequently, the control module calculates the feedforward dosing command value using the aforementioned formula.

[0068] S4. The automatic dosing execution module responds to the feedforward dosing control command and executes the main process reagent dosing.

[0069] Specifically, the control module converts the feedforward dosing command value into the drive frequency value of the variable frequency dosing pump and the opening value of the precision metering valve. The variable frequency dosing pump operates according to the drive frequency value, pumping the reagent from the reagent storage tank, which is then injected into the dosing pipeline after secondary fine adjustment by the precision metering valve. The reagent is then transported along the dosing pipeline to the dosing point of the main process pipeline, where it mixes with the old kiln water flowing through that point.

[0070] S5, the effluent feedback correction module collects the effluent pH value and effluent iron ion concentration in real time and feeds them back to the control module.

[0071] The sensors in the effluent feedback correction module continuously monitor the effluent water quality parameters after the dosing reaction is completed, and transmit the actual effluent pH value and effluent iron ion concentration to the control module via the data bus.

[0072] S6. The control module corrects and compensates the feedforward dosing control command based on the deviation between the effluent pH value and the effluent iron ion concentration and the target value.

[0073] Specifically, the control module calculates the deviation between the actual effluent pH value and the target value. The deviation between the actual effluent iron ion concentration and the target value The control module calculates the feedback compensation amount using the previous method. Then, the feedback compensation is added to the feedforward dosing command value to generate the final dosing control command for the next moment. .

[0074] Third Embodiment In this embodiment, based on the second embodiment, the control module is also used to determine the early termination conditions of the simulated pre-reaction experiment.

[0075] The control module calculates the rate of change of pH and the rate of change of iron ion concentration in the water sample inside the micro-reaction vessel in real time.

[0076] Specifically, the control module performs differential processing on the pH value sequence and iron ion concentration sequence collected by the online detection unit, and calculates the instantaneous rate of change at each sampling moment. The pH change rate is obtained by dividing the difference between the current pH value and the pH value of the previous moment by the sampling time interval. The iron ion concentration change rate is obtained by dividing the difference between the current iron ion concentration and the iron ion concentration of the previous moment by the sampling time interval.

[0077] When the rates of change in pH and iron ion concentration for N consecutive sampling cycles both decrease to below the preset proportion of their corresponding peak values, the control module determines that the simulated pre-reaction experiment has essentially reached a steady state, terminates the current simulated pre-reaction experiment early, and records the current water quality parameters. The peak reaction rate is the instantaneous rate of change at the moment of most intense reaction after reagent addition. The peak reaction rate is determined as follows: the control module continuously monitors the rate of change after reagent addition and records the maximum monitored rate of change as the peak reaction rate.

[0078] Optionally, in this embodiment, the preset ratio is set to 5%. That is, when the rate of pH change for five consecutive sampling cycles decreases to less than 5% of the peak rate of pH, and the rate of iron ion concentration change for five consecutive sampling cycles decreases to less than 5% of the peak rate of iron ion concentration, the early termination condition is determined to be met.

[0079] Once the early termination condition is met, the control module no longer waits for the preset fixed reaction time, directly stopping data acquisition for the current experiment and recording the current pH value and iron ion concentration as steady-state values. By introducing the early termination condition, the duration of the simulated pre-reaction experiment is no longer fixed, but adaptively adjusted according to the actual reaction kinetics of the real-time water sample. For rapid reaction conditions, the experiment duration is automatically shortened, improving the system's response speed. For slow reaction conditions, the system waits until the early termination condition is met before outputting results, ensuring the reliability of steady-state data.

[0080] Fourth embodiment In this embodiment, based on the second embodiment, the number of micro-reaction simulators is two or more. Multiple micro-reaction simulators are connected in parallel to the water inlet pipe via bypass pipelines.

[0081] Each micro-reaction simulator has the same structure and function as the micro-reaction simulator described in the first embodiment, and all include a micro-reaction vessel, a precision dosing unit, an online detection unit, and an automatic cleaning unit. Each micro-reaction simulator is connected to the inlet water pipe via an independent bypass branch pipe. Each bypass branch pipe is equipped with an electromagnetic switch valve, which is independently controlled by the control module to open and close, thereby enabling and disabling each micro-reaction simulator independently.

[0082] The control module employs a pipelined parallel scheduling algorithm to allocate tasks to each microreactor. The core idea of ​​this algorithm is to decompose the process of a single microreactor completing a full simulation of a pre-reaction experiment into four sequentially executed stages: sampling, chemical dosing, reaction monitoring, and cleaning / discharging. Each stage occupies one time slice. Multiple microreactors execute different stages interleaved on the same timeline, forming a pipeline operation mode.

[0083] As an example, the pipeline scheduling process is illustrated using three micro-reaction simulators A, B, and C. In the first time slice, A performs the sampling phase, while B and C are in standby mode. In the second time slice, A enters the dosing phase, and B begins the sampling phase. In the third time slice, A enters the reaction monitoring phase, B enters the dosing phase, and C begins the sampling phase. In the fourth time slice, A enters the cleaning and decontamination phase, B enters the reaction monitoring phase, and C enters the dosing phase. In the fifth time slice, after completing cleaning and decontamination, A enters standby mode to prepare for the next round of sampling, B enters the cleaning and decontamination phase, and C enters the reaction monitoring phase. This cycle continues, with each micro-reaction simulator operating in a different phase, and these phases are staggered on the timeline, ensuring that the control module obtains a set of simulated pre-reaction experimental results within each time slice period.

[0084] By employing a pipelined parallel scheduling algorithm, the overall experimental output frequency of multiple micro-reaction simulators is increased to a multiple of the experimental frequency of a single simulator. While maintaining the total duration of a single experiment in the micro-reaction simulator, the time interval for the control module to obtain the simulated pre-reaction experiment results is shortened from the experimental cycle time of a single simulator to a fraction of the experimental cycle time.

[0085] Fifth embodiment This embodiment uses actual operating data from an old coal mine water treatment station as an example to illustrate the specific application process of the present invention.

[0086] The old kiln water treatment plant is designed to treat 50 cubic meters per hour. The influent water quality characteristics are: pH fluctuating between 2.5 and 5.5, and iron ion concentration fluctuating between 80 mg / L and 400 mg / L. The hydraulic retention time is 12 minutes.

[0087] During system initialization, operators set the target pH value of the effluent to 7.0 and the upper limit of the effluent iron ion concentration to 10 mg / L. The control module loads the old cellar water dosing threshold library and PID parameter mapping table from the storage medium. The dosing threshold library contains the baseline dosing dosage corresponding to different combinations of influent pH and iron ion concentrations, generated from previous laboratory beaker experimental data. The PID parameter mapping table contains three sets of PID parameter combinations for high fluctuation, medium fluctuation, and low fluctuation conditions.

[0088] After the system was put into operation, the multi-parameter monitoring module collected the instantaneous influent flow rate, influent pH value, and influent iron ion concentration at a sampling frequency of once per second. At a certain moment, the multi-parameter monitoring module collected the instantaneous influent flow rate as 45 cubic meters per hour, the influent pH value as 3.2, and the influent iron ion concentration as 220 milligrams per liter. After receiving the above parameters, the control module started the micro-reaction simulator to execute a simulated pre-reaction experiment.

[0089] The miniature reaction simulator diverts 100 ml of real-time water sample from the inlet pipe to the miniature reaction vessel via a bypass pipeline. The control module retrieves the baseline dosage matching the pH value of 3.2 and the iron ion concentration of 220 mg / L from the dosing threshold library, which is 2.5 ml of alkali solution per liter of water sample.

[0090] The control module performed the first round of simulated pre-reaction experiments at a baseline dosage of 2.5 ml / L. The precision dosing unit added 0.25 ml of alkali solution to the micro-reaction vessel. A stirrer mixed the water sample with the alkali solution at 300 rpm, and the online monitoring unit continuously monitored changes in pH and iron ion concentration. At 180 seconds into the reaction, both the rate of change in pH and the rate of change in iron ion concentration decreased to below 5% of their peak values. The control module prematurely terminated the experiment and recorded a steady-state pH of 6.2 and a steady-state iron ion concentration of 35 mg / L. The steady-state pH of 6.2 was lower than the target value of 7.0, and the steady-state iron ion concentration of 35 mg / L was higher than the upper limit of 10 mg / L, indicating that the treatment effect did not meet the target.

[0091] The control module initiates a stepped approximation method to adjust the simulated dosage. The first step centers on a baseline dosage of 2.5 ml / L, with three dosage groups of 1.25 ml / L, 2.5 ml / L, and 3.75 ml / L respectively, to conduct simulated pre-reaction experiments. The steady-state pH and iron ion concentrations for the three groups are as follows: pH 5.5 and iron ion concentration 65 mg / L at 1.25 ml / L; pH 6.2 and iron ion concentration 35 mg / L at 2.5 ml / L; and pH 7.3 and iron ion concentration 8 mg / L at 3.75 ml / L. Only the 3.75 ml / L group meets the target conditions, and the critical range for achieving the target is determined to be 3.0 ml / L to 4.5 ml / L.

[0092] The second step employs the golden section method for iterative testing within the critical dosing range. In the first round, experiments were conducted at the golden section points of 3.57 ml / L and 3.93 ml / L within the range of 3.0 ml / L to 4.5 ml / L. The results showed that at 3.57 ml / L, the pH was 7.1 and the iron concentration was 12 mg / L, which did not meet the standards; at 3.93 ml / L, the pH was 7.2 and the iron concentration was 9 mg / L, meeting the standards. The range narrowed to 3.57 ml / L to 4.5 ml / L. The second round continued to test at the golden section points within this narrowed range, further narrowing it. After the third iteration, the range width was less than the dosage equivalent corresponding to the minimum dosing accuracy of the precision dosing unit, terminating the iteration. The current midpoint of the range, 3.8 ml / L, was taken as the critical dosage. .

[0093] The control module reads the current model confidence level. The average absolute value of the feedback correction from the effluent feedback correction module is calculated based on the data from the last 20 effluent feedback corrections. The current feedforward dosing command value is 0.12. This is calculated as 3.8 ml per liter multiplied by 45 cubic meters per hour, which equals 171 ml per hour. The confidence level is then calculated using the formula... / get 1.0 - 0.12 / 171 = 0.999, the model confidence level is close to 1.0. The safety factor is taken as 1.02, and the final simulated dosage... Multiplying 3.8 ml per liter by 1.02 gives 3.876 ml per liter.

[0094] The control module uses the instantaneous influent flow rate of 45 cubic meters per hour as the primary weight to calculate the feedforward dosing command value. The optimal dosing rate per unit flow rate is then determined. 3.8 liters of alkali solution were added per cubic meter of water. The comprehensive water quality correction coefficient ρ was calculated according to the corresponding formula in the first embodiment. The pH reference value was set to 4.0, the iron ion concentration reference value to 150 mg / L, α to 0.1, and β to 0.005. Substituting the real-time pH value of 3.2 and the real-time iron ion concentration of 220 mg / L, the calculation yielded ρ = 1 + 0.1 × (4.0 - 3.2) + 0.005 × (150 - 220) = 1 + 0.08 - 0.35 = 0.73. (Feedforward dosing command value) =3.8×45×0.73=124.8 liters per hour.

[0095] The control module converts the feedforward dosing command value of 124.8 liters per hour into the drive frequency value of the variable frequency dosing pump and sends it to the automatic dosing execution module to execute the main process reagent dosing. The reagent is injected into the dosing point of the main process pipeline along the dosing pipeline and mixes and reacts with the old kiln water flowing through it.

[0096] Twelve minutes later, the corresponding batch of water samples flowed to the effluent monitoring point. The effluent feedback correction module detected an effluent pH value of 6.9 and an effluent iron ion concentration of 8 mg / L. The control module calculated the effluent pH deviation. The deviation of iron ion concentration in the effluent is -0.1. The value is -2 mg / L. Calculate the feedback compensation, assuming the current PID parameter combination is... It is 2.0. It is 0.5. The bias vector e(t) is calculated as a weighted sum of pH bias and iron ion concentration bias, with a value of 1.0. -2.3 liters per hour. Final dosing control command. =124.8-2.3=122.5 liters per hour.

[0097] The control module sends the final dosing control command of 122.5 liters per hour as the dosing command for the next moment to the automatic dosing execution module, completing the entire control loop of feedforward dosing and feedback correction.

[0098] Operational data shows that during a month of continuous operation, the effluent pH value remained stable between 6.8 and 7.2, and the effluent iron ion concentration remained stable between 5 mg / L and 10 mg / L. Chemical consumption was reduced by approximately 40% compared to manual, experience-based dosing. When the influent pH value fluctuated within the range of 2.5 to 5.5, the micro-reaction simulator was able to complete the simulated pre-reaction experiment and output feedforward commands within 8 minutes, earlier than the 12-minute hydraulic retention time.

[0099] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0100] In addition, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0101] The above 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.

Claims

1. A smart control system for predicting the pre-dosing of wastewater from old kilns based on an online micro-reaction simulator, applied to old kiln wastewater treatment scenarios, wherein the wastewater is introduced into the treatment station through an inlet pipe, and the treated effluent is discharged through an outlet pipe, characterized in that... include: A multi-parameter monitoring module is deployed in the inlet pipe to collect the instantaneous flow rate, pH value, and iron ion concentration of the inlet water in real time. A miniature reaction simulator is connected to the inlet pipe via a bypass pipeline to divert real-time water samples and conduct simulated pre-reaction experiments, so as to monitor the dynamic response curve of the pH value and iron ion concentration of the water sample in real time. The control module is used to determine the optimal dosage corresponding to the real-time water sample based on the dynamic response curve, and to correct the optimal dosage with the real-time influent parameters as weights, thereby generating a feedforward dosing control command. An automatic dosing execution module, deployed in the dosing pipeline, is used to respond to the feedforward dosing control command and execute the dosing of the agent; The water outlet feedback correction module is deployed in the water outlet pipe to collect the water outlet pH value and the water outlet iron ion concentration in real time and feed them back to the control module. The control module is also used to correct and compensate the feedforward dosing control command based on the deviation between the effluent pH value and the effluent iron ion concentration and the target value.

2. The intelligent control system for predicting the pre-dosing of chemicals in old kiln water based on an online micro-reaction simulator according to claim 1, characterized in that, The micro-reaction simulator includes a micro-reaction vessel, a precision dosing unit, and an online detection unit; The micro-reaction vessel is connected to the inlet pipe via a bypass pipeline and has a built-in stirrer for containing the real-time water sample obtained from the diversion. The precision dosing unit is used to quantitatively add reagents into the micro-reaction vessel to simulate the main process dosing procedure; The online detection unit is used to monitor the dynamic response curves of pH value and iron ion concentration of the water sample in the micro-reaction vessel in real time.

3. The intelligent control system for predicting the addition of chemicals to old kiln water based on an online micro-reaction simulator according to claim 1, characterized in that, The sampling frequency of the multi-parameter monitoring module is configured to match the fluctuation rate of the old kiln water quality in order to capture the dynamic changes in the influent water quality; the completion time of a single simulated pre-reaction experiment of the micro-reaction simulator is less than the hydraulic residence time of the old kiln water flowing from the dosing point to the effluent monitoring point, so that the feedforward dosing control command is output before the corresponding water sample arrives at the dosing point.

4. The intelligent control system for predicting the addition of chemicals to old kiln water based on an online micro-reaction simulator according to claim 1, characterized in that, The control module determines the optimal dosage for the real-time water sample based on the dynamic response curve, specifically including: The control module extracts features from the dynamic response curve to obtain the steady-state pH value, steady-state iron ion concentration, and steady-state arrival time; the steady-state pH value and steady-state iron ion concentration are the values ​​corresponding to the changes in pH value and iron ion concentration in the dynamic response curve being less than preset thresholds in multiple consecutive sampling periods, respectively. The control module compares the steady-state pH value and steady-state iron ion concentration with preset target values ​​to determine the treatment effect under the current simulated dosage. If the treatment effect does not meet the standard, the control module adjusts the simulated dosage of the precision dosing unit and starts the next round of simulated pre-reaction experiment until the treatment effect meets the standard. If the processing effect meets the standard, the control module will use the current simulated dosage as the candidate optimal dosage.

5. The intelligent control system for predicting the pre-dosing of chemicals in old kiln water based on an online micro-reaction simulator according to claim 4, characterized in that, The control module uses a step-by-step approximation method to adjust the simulated dosage of the precision dosing unit. The baseline dosage matching the current influent pH and iron ion concentration is retrieved from the built-in old kiln water dosing threshold library. Multiple sets of simulated pre-reaction experiments with different gradients are performed with the baseline dosage as the center to obtain the dosage range in which the steady-state pH and steady-state iron ion concentration both meet the preset compliance conditions. The lower limit and upper limit of this range are used as the lower limit and upper limit of the compliance critical range, respectively. Within the critical range of compliance, the dosage of the experimental point is calculated iteratively in round by round using an interval narrowing optimization algorithm. Based on the compliance judgment results of each round of experiments, the boundary of the critical range of compliance is narrowed round by round until the width of the critical range of compliance is less than the minimum dosing accuracy of the precision dosing unit, at which point the iteration terminates and the midpoint of the current range is taken as the critical dosage. The current model confidence level is obtained. The model confidence level is dynamically determined by the historical feedback correction amplitude and correction frequency of the effluent feedback correction module. A safety factor negatively correlated with the model confidence level is superimposed on the critical dosage to generate the final simulated dosage.

6. The intelligent control system for predicting the pre-dosing of chemicals in old kiln water based on an online micro-reaction simulator according to claim 1, characterized in that, The control module adjusts the optimal dosage using the real-time influent parameters as weights, specifically including: The control module uses the instantaneous influent flow rate. As the sovereign entity, the feedforward dosing command value is calculated according to the following formula. : ,in, The optimal dosage per unit flow rate. This is the comprehensive water quality correction factor. Determined by the following formula: in, and These are reference values ​​for pH and iron ion concentration, respectively. and These are the real-time influent pH value and the real-time influent iron ion concentration, respectively. and These are the correction factors for pH deviation and iron ion concentration deviation, respectively.

7. The intelligent control system for predicting the pre-dosing of chemicals in old kiln water based on an online micro-reaction simulator according to claim 6, characterized in that, The control module corrects and compensates for the feedforward dosing control command based on the deviations between the effluent pH value and the effluent iron ion concentration and the target values, specifically including: The control module calculates the pH deviation of the water. Deviation in effluent iron ion concentration ; The control module uses a PID control algorithm to calculate the feedback compensation amount. : in, For the comprehensive deviation vector, , , These are the proportional coefficient, integral coefficient, and differential coefficient, respectively. The control module calculates the final dosing control command according to the following formula. : 。 8. The intelligent control system for predicting the pre-dosing of chemicals in old kiln water based on an online micro-reaction simulator according to claim 7, characterized in that, The control module is also used to adaptively update the PID control parameters of the effluent feedback correction module according to the fluctuation range of the influent water quality. The control module constructs a three-dimensional wave feature vector. ,in These represent the rates of change of pH value, iron ion concentration, and flow rate within adjacent sampling periods of the multi-parameter monitoring module. The control module counts the N consecutive sampling periods preceding the current sampling time. Standard deviation of each dimension , , and with As a comprehensive volatility indicator; The control module has multiple pre-set PID parameter combinations corresponding to different fluctuation levels. Each PID parameter combination includes a first set of parameters under high fluctuation conditions and a second set of parameters under low fluctuation conditions. The proportional coefficient and derivative coefficient in the first set of parameters are both greater than the corresponding coefficients in the second set of parameters, and the integral coefficient in the first set of parameters is less than the integral coefficient in the second set of parameters. The control module uses a comprehensive fluctuation index To index and query the PID parameter combinations, the proportional coefficient under the current operating condition is determined. Integral coefficient and differential coefficients and determined by matching , , Update the PID control law of the effluent feedback correction module.

9. A pre-treatment intelligent control method for old kiln water based on an online micro-reaction simulator, executed by the pre-treatment intelligent control system for old kiln water according to any one of claims 1-8, characterized in that, Includes the following steps: S1. The multi-parameter monitoring module collects the instantaneous flow rate, pH value, and iron ion concentration of the influent in real time. S2. The micro-reaction simulator diverts real-time water samples from the inlet pipe through a bypass pipeline to perform a simulated pre-reaction experiment and monitors the dynamic response curve of the pH value and iron ion concentration of the water sample in real time. S3. The control module receives the real-time influent parameters and the dynamic response curve, determines the optimal dosage corresponding to the real-time water sample based on the dynamic response curve, and corrects the optimal dosage with the real-time influent parameters as weights, generating a feedforward dosing control command. S4. The automatic dosing execution module responds to the feedforward dosing control command and executes the main process reagent dosing; S5. The effluent feedback correction module collects the effluent pH value and effluent iron ion concentration in real time and feeds them back to the control module. S6. The control module corrects and compensates the feedforward dosing control command based on the deviation between the pH value and the iron ion concentration of the effluent and the target value.

10. The intelligent control method for pre-dosing chemicals in old kiln water based on an online micro-reaction simulator according to claim 9, characterized in that, In step S2, the micro-reaction simulator performs a simulated pre-reaction experiment, which includes the following steps: S21. The precision dosing unit quantitatively adds reagents to the real-time water sample in the micro-reaction vessel; S22. The stirrer is set to a preset speed to ensure that the water sample and the reagent are fully mixed and reacted. S23. The online detection unit continuously collects the pH value and iron ion concentration of the water sample during the reaction process and generates the dynamic response curve. S24. The automatic cleaning unit performs cleaning and decontamination of the micro-reaction vessel after the experiment.

Images