Intelligent control method and system for preparation process of pool water flocculant

Abstract

The application relates to the technical field of water treatment control, and discloses an intelligent control method and system for a swimming pool water flocculant preparation process, which comprises the following steps: synchronously acquiring real-time conductivity of a preparation tank, real-time flow of backwater, ultraviolet absorbance and water turbidity; performing time sequence difference operation on the real-time conductivity to obtain a dissolution completion degree index, terminating stirring when the index meets a threshold value, and calculating an efficiency attenuation coefficient; establishing a first-in-first-out data mapping queue, binding the efficiency attenuation coefficient with a liquid level interval of a dosing tank, and calling the dosing tank according to a consumed volume when extracting a medicament; obtaining an organic load factor through flow and absorbance, performing feedforward control operation in combination with the called efficiency coefficient, performing feedback control operation in combination with turbidity deviation, adding the two to generate a final instruction, and controlling a variable frequency metering pump. The application realizes closed-loop determination of medicament preparation and spatial tracking of batch efficiency, overcomes hydraulic lag of traditional control, and improves anti-interference capability.

Description

Technical Field

[0001] This invention relates to the field of water treatment control technology, specifically to an intelligent control method and system for the preparation process of pool water flocculants. Background Technology

[0002] In large public swimming pools and related circulating water treatment projects, adding polymeric flocculants to the water is a necessary step to remove suspended particles and dissolved organic matter. The quality of the flocculant's chemical preparation and the accuracy of the dosing control directly affect the retention effect of the sand filter and the final effluent water quality indicators.

[0003] Existing flocculant preparation and dosing systems typically employ a fixed-sequence mechanical stirring mode for reagent preparation, relying on fixed feedback logic to control the variable frequency metering pump during the dosing phase. In the reagent preparation stage, the control program drives the variable frequency mixer according to factory-set time parameters, failing to capture the actual dissolution progress and ion release status of the flocculant in the aqueous phase. Due to fluctuations in ambient water temperature and changes in water quality, the fixed stirring time can easily lead to incomplete hydration or over-stirring of the reagent. Incompletely hydrated reagent contains agglomerates, which can easily cause blockages in subsequent metering pipelines; while the shear force generated by excessive mechanical stirring can damage the molecular chain structure of polymers, leading to irreversible degradation of the reagent's flocculation efficiency.

[0004] The prepared reagent is then transferred to the dosing tank for storage and consumption. During this process, influenced by chemical factors such as the ionic strength of the background water, the polymer molecular chains experience electrostatic shielding, causing their bridging and trapping efficiency to decrease differentially with different batches and residence times. Conventional control systems treat the reagent inside the dosing tank as a fluid with constant and completely homogeneous physicochemical properties, without establishing a data tracking mechanism for the spatial distribution of the reagent within the container. This results in the system being unable to determine the true effectiveness of the extracted reagent during long-term continuous operation and consumption of different batches, thus failing to accurately compensate for the gain of the dosing command and easily leading to insufficient dosing or reagent waste.

[0005] In practical dosing control loops, existing systems often rely on an effluent turbidity meter installed downstream of the filtration equipment as the sole input parameter for closed-loop feedback regulation. Water treatment process networks typically have large volumes and long hydraulic retention times. When the organic load in the pool fluctuates instantaneously due to external factors such as sudden changes in passenger flow, the lagging feedback signal from the effluent end cannot promptly transmit the water quality deterioration to the upstream dosing actuator. This single feedback mechanism results in a control signal that lags significantly behind actual process requirements, leading to overshooting of the effluent turbidity index when water quality is impacted. Furthermore, the operating frequency of the variable frequency metering pump is prone to prolonged and repeated oscillations during adjustment, failing to guarantee stable system operation under dynamic conditions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an intelligent control method and system for the preparation process of pool water flocculants. This solves the problems in existing water treatment systems, such as the damage to the efficacy of the flocculant due to fixed-time stirring, the lack of a spatial tracking mechanism for the differences in efficacy between different batches of the flocculant, and the serious hydraulic lag caused by the single effluent turbidity feedback control, which leads to water quality overshoot and system oscillation when dealing with sudden changes in influent load.

[0007] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of this invention provides an intelligent control method for the preparation process of pool water flocculants, comprising the following steps: The real-time conductivity of the solution in the preparation tank, the real-time flow rate of the return water in the pipeline network, the ultraviolet absorbance, and the turbidity of the effluent were obtained. The real-time conductivity is subjected to time series difference operation to obtain the dissolution completion index. When the dissolution completion index falls within the steady-state tolerance threshold, the variable frequency mixer is shut down, and the current real-time conductivity is set as the steady-state conductivity. Based on the steady-state conductivity, the efficiency decay coefficient is obtained through a physical model. A data mapping queue is established to bind the efficiency attenuation coefficient to the liquid level of the reagent transferred to the dosing tank. When the variable frequency metering pump extracts the reagent, the corresponding efficiency attenuation coefficient is retrieved from the data mapping queue. The organic load factor is obtained based on the real-time flow rate and the ultraviolet absorbance. The dynamic feedforward gain coefficient is obtained by multiplying the efficiency attenuation coefficient and the basic feedforward gain. The feedforward output frequency is obtained by multiplying the dynamic feedforward gain coefficient and the organic load factor. The feedback fine-tuning frequency is obtained based on the set turbidity and the effluent turbidity. The feedforward output frequency and the feedback fine-tuning frequency are added together to obtain the final command. The variable frequency metering pump is controlled according to the final command.

[0008] Preferably, the process of performing time-series difference calculations on the real-time conductivity to obtain the dissolution completion index specifically includes: Calculate the difference between the real-time conductivity at the current discrete time step and the real-time conductivity at the previous discrete time step; The dissolution completion index is obtained by dividing the difference by the value of the discrete sampling period.

[0009] Preferably, shutting down the variable frequency mixer when the dissolution completion index falls within the steady-state tolerance threshold specifically includes: Determine whether the dissolution completion index is greater than zero and less than the steady-state tolerance threshold for a consecutive preset number of sampling periods; When the dissolution completion index is greater than zero and less than the steady-state tolerance threshold for a preset number of consecutive sampling periods, a maturation endpoint interruption signal is generated. The maturation endpoint interruption signal is sent to the drive end of the variable frequency mixer to stop the variable frequency mixer from running.

[0010] Preferably, the efficiency degradation coefficient derived from the steady-state conductivity using a physical model specifically includes: Obtain the steady-state conductivity constant and penalty weighting coefficient of the reference solvent; The difference is obtained by subtracting the steady-state conductivity constant from the steady-state conductivity. The penalty amount is obtained by multiplying the difference value by the penalty weight coefficient; The efficiency decay coefficient is obtained by adding the constant to the penalty amount.

[0011] Preferably, establishing the data mapping queue specifically includes: Obtain the depth parameters inside the dosing box; The depth parameter is divided into multiple liquid level intervals; Establish a first-in-first-out data structure queue containing multiple nodes within the storage space of the main control unit; Assign the multiple liquid level ranges to the multiple nodes, and use the first-in-first-out data structure queue as the data mapping queue.

[0012] Preferably, the efficiency attenuation coefficient is linked to the liquid level of the reagent transferred to the dosing tank. When the variable frequency metering pump extracts the reagent, retrieving the corresponding efficiency attenuation coefficient from the data mapping queue specifically includes: Obtain the volume of the transferred reagent after a single preparation is completed and transferred into the dosing tank; The corresponding liquid level range is determined based on the increase in liquid level inside the dosing tank caused by the volume of the transferred agent. Obtain the cumulative volume of consumed reagent pumped by the variable frequency metering pump; The liquid level range corresponding to the currently consumed agent inside the dosing tank is determined based on the volume of the consumed agent, and the efficiency decay coefficient stored in the corresponding node is read from the data mapping queue.

[0013] Preferably, deriving the organic loading factor based on the real-time flow rate and the ultraviolet absorbance specifically includes: Simultaneously extract the values ​​of the real-time flow rate and the ultraviolet absorbance at the current discrete time step; Multiplying the real-time flow rate by the ultraviolet absorbance value yields the organic load factor, which represents the total amount of dissolved organic matter entering the water treatment system per unit time.

[0014] Preferably, the step of deriving the feedback fine-tuning frequency based on the set turbidity and the effluent turbidity specifically includes: The turbidity deviation is obtained by subtracting the effluent turbidity from the set turbidity. Obtain the pre-set proportional coefficient, integral coefficient, and differential coefficient; The proportionality coefficient and the turbidity deviation are multiplied to obtain the proportional control term value; The integral control term value is obtained by multiplying the integral coefficient and the historical cumulative value of the turbidity deviation. Obtain the rate of change of the turbidity deviation between the current discrete time step and the previous discrete time step; The differential control term value is obtained by multiplying the differential coefficient and the deviation change rate. The feedback fine-tuning frequency is obtained by adding the values ​​of the proportional control term, the integral control term, and the derivative control term.

[0015] Preferably, controlling the variable frequency metering pump according to the final instruction specifically includes: Convert the final instruction into the target operating frequency value; The target operating frequency value is converted into a corresponding pulse frequency electrical signal; The pulse frequency electrical signal is transmitted to the frequency converter that drives the variable frequency metering pump through the communication bus, instructing the frequency converter to drive the variable frequency metering pump to operate according to the target operating frequency value.

[0016] A second aspect of the present invention provides an intelligent control system for the preparation process of pool water flocculants, comprising: The parameter acquisition module is used to obtain the real-time conductivity of the solution in the preparation tank, the real-time flow rate of the return water in the pipeline network, the ultraviolet absorbance, and the turbidity of the effluent. A preparation control module is used to perform time series difference calculation on the real-time conductivity to obtain a dissolution completion index. When the dissolution completion index falls within the steady-state tolerance threshold, the variable frequency mixer is shut down, the current real-time conductivity is set as the steady-state conductivity, and the efficiency decay coefficient is obtained based on the steady-state conductivity through a physical model. The data mapping module is used to establish a data mapping queue, bind the efficiency attenuation coefficient to the liquid level of the reagent transferred to the dosing tank, and call the corresponding efficiency attenuation coefficient from the data mapping queue when the variable frequency metering pump extracts the reagent. The dosing calculation execution module is used to derive the organic load factor based on the real-time flow rate and the ultraviolet absorbance, multiply the called efficiency attenuation coefficient by the basic feedforward gain to obtain the dynamic feedforward gain coefficient, multiply the dynamic feedforward gain coefficient by the organic load factor to obtain the feedforward output frequency, derive the feedback fine-tuning frequency based on the set turbidity and the effluent turbidity, add the feedforward output frequency and the feedback fine-tuning frequency to obtain the final command, and control the variable frequency metering pump according to the final command.

[0017] This invention provides an intelligent control method and system for the preparation process of pool water flocculants. It has the following beneficial effects: 1. This invention derives a dissolution completion index by performing time-series differential calculations on the real-time conductivity of the solution in the preparation tank. The index falling below a steady-state tolerance threshold is used as the trigger condition for shutting down the variable frequency stirrer. This transforms the traditional open-loop timed stirring process into a closed-loop control based on chemical state parameters. This avoids the phenomenon of incomplete dissolution or excessive mechanical shear degradation of polymers in the environment due to fixed time settings, ensuring the physicochemical stability of the flocculant in a single preparation.

[0018] 2. This invention calculates the efficacy decay coefficient reflecting the state of the drug by extracting the steady-state conductivity at the end of the ripening process, and constructs a data mapping queue to bind this coefficient to the drug level range in the dosing tank, establishing a data transmission link across the preparation and dosing stages. This parameter addressing and calling mechanism based on a first-in-first-out structure enables the dosing execution unit to acquire and match the efficacy variation characteristics of the currently consumed batch of drug due to differences in solvent background during continuous operation.

[0019] 3. In the dosing operation phase, this invention utilizes the organic load factor derived from real-time flow rate and UV absorbance as the feedforward input. It dynamically corrects the basic feedforward gain using the batch efficiency attenuation coefficient, and superimposes a feedback fine-tuning frequency generated from effluent turbidity calculations onto the feedforward output frequency. This compensates for external disturbances caused by sudden increases in water pollutants and internal disturbances caused by fluctuations in the physicochemical properties of the reagent itself, reducing the time lag and actuator frequency oscillations present in conventional single turbidity feedback control mechanisms. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the preparation method of the present invention; Figure 2 This is a schematic diagram of the system architecture of the present invention; Figure 3 This is a comparative data curve of the effluent turbidity under disturbed conditions according to the present invention; Figure 4This is a comparative data curve of the target operating frequency of the variable frequency metering pump of the present invention under disturbance conditions. Detailed Implementation

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Reference Figure 1 and Figure 2 In this embodiment, an intelligent control system for the preparation process of pool water flocculant includes a parameter acquisition module, a preparation control module, a data mapping module, and a dosing calculation execution module interconnected via an industrial baseboard bus or communication interface. These modules are integrated within the hardware environment of a microprocessor, programmable logic controller, or industrial computer host, and operate collaboratively according to a set program clock cycle.

[0023] The hardware input ports of the parameter acquisition module establish electrical communication connections with multi-source sensors deployed at the water treatment site. Specifically, the module's interface connects to an online conductivity sensor installed inside the preparation tank, an electromagnetic flow meter and an ultraviolet spectral absorption sensor installed on the main return water pipe of the pipeline network, and an online turbidity meter installed on the effluent side of the pipeline network. The module internally includes an analog-to-digital converter and a digital message parsing protocol stack to convert the raw electrical signals transmitted by each sensor into engineering-quantified values, thereby simultaneously acquiring the real-time conductivity of the solution inside the preparation tank, the real-time flow rate of the return water from the pipeline network, the ultraviolet absorbance, and the effluent turbidity. The acquired parameter data is uniformly stored in the module's shared buffer for other modules to read.

[0024] The data input terminal of the preparation control module is connected to the shared buffer of the parameter acquisition module via an internal bus. The preparation control module is internally equipped with a clock sequence generator and an arithmetic logic unit, used to perform time-series difference operations on the extracted real-time conductivity within consecutive discrete time steps to obtain the dissolution completion index. The output terminal of the preparation control module is equipped with a digital switch contact. When the dissolution completion index falls within the steady-state tolerance threshold solidified in the read-only memory, a low-level cutoff signal is output to the variable frequency stirrer drive circuit of the preparation chamber through this switch contact, shutting down the variable frequency stirrer. Simultaneously, the preparation control module sets the real-time conductivity at the moment of triggering the interrupt action as the steady-state conductivity, and based on this steady-state conductivity, derives the performance attenuation coefficient through its internal preset physical model logic path.

[0025] The data mapping module is connected to the data output bus of the preparation control module. The data mapping module dynamically allocates memory space in the main control device's random access memory (RAM) to establish a first-in-first-out (FIFO) data mapping queue. By writing instructions, the data mapping module stores the efficiency attenuation coefficient calculated by the preparation control module into the memory address segment, thereby transferring this parameter to the reagent level range within the dosing tank and binding it in digital space. Simultaneously, the data mapping module establishes feedback communication with the execution monitoring port controlling the variable frequency metering pump. While the variable frequency metering pump continuously extracts reagent, the data mapping module generates a memory offset address pointer based on the flow integral signal generated by the pump's operation. By changing the pointer's direction, it retrieves and calls the corresponding efficiency attenuation coefficient from the data mapping queue.

[0026] The data call ports of the dosing operation execution module point to the data register addresses of the parameter acquisition module and the data mapping module, respectively. The dosing operation execution module is configured with cascaded arithmetic units. Its preamplifier multiplier derives the organic load factor based on the synchronously extracted real-time flow rate and UV absorbance. Its correction arithmetic unit multiplies the called efficiency attenuation coefficient with the basic feedforward gain in memory to obtain the dynamic feedforward gain coefficient. Subsequently, the main arithmetic unit multiplies the dynamic feedforward gain coefficient and the organic load factor to obtain the feedforward output frequency. In the closed-loop channel, the difference comparator and proportional-integral-differential arithmetic unit of the dosing operation execution module derive the feedback fine-tuning frequency based on the set turbidity and the effluent turbidity. Finally, the adder of the dosing operation execution module adds the feedforward output frequency and the feedback fine-tuning frequency to obtain the final instruction. The dosing operation execution module is configured with a high-speed pulse output channel, which generates corresponding control pulse waveforms according to the final instruction and sends them to the controller of the variable frequency metering pump to directly control the mechanical operating frequency of the variable frequency metering pump.

[0027] Reference Figure 1 This invention provides an intelligent control method for the preparation process of pool water flocculant, the method comprising: Under a set discrete sampling period, the real-time conductivity of the solution in the preparation tank is acquired simultaneously, along with the real-time flow rate and UV absorbance of the return water from the pipeline network, and the turbidity of the effluent from the water treatment pipeline network. Real-time conductivity, real-time flow rate, UV absorbance, and effluent turbidity are used as input parameters for the control program.

[0028] A time-series difference operation is performed on the acquired real-time conductivity to obtain the dissolution completion index under the current state. During the operation, the numerical change of the dissolution completion index is monitored to determine whether the dissolution completion index falls within the pre-set steady-state tolerance threshold. When the dissolution completion index meets this boundary condition, a maturation endpoint interruption signal is generated and the variable frequency mixer is shut down, terminating the mechanical stirring program of the flocculant.

[0029] Within the same discrete time step of shutting down the variable frequency mixer, the current real-time conductivity is extracted and set as the steady-state conductivity of this batch of reagents. Based on the extracted steady-state conductivity, the corresponding efficiency attenuation coefficient is obtained through calculation using an internally preset physical model.

[0030] Establish a first-in, first-out (FIFO) data mapping queue. Obtain the physical liquid level range formed inside the dosing tank after a single preparation volume is transferred, and associate the calculated efficiency decay coefficient with the liquid level in that dosing tank. When the dosing procedure is executed and the variable frequency metering pump is extracting the reagent, retrieve the efficiency decay coefficient corresponding to the currently consumed reagent from the data mapping queue based on the cumulative volume extracted by the variable frequency metering pump.

[0031] The organic load factor is obtained by multiplying the real-time flow rate and UV absorbance. The efficiency attenuation coefficient retrieved from the data mapping queue is multiplied by the internally set base feedforward gain to obtain the dynamic feedforward gain coefficient. Subsequently, the dynamic feedforward gain coefficient is multiplied by the organic load factor to obtain the feedforward output frequency used in the feedforward control loop.

[0032] The deviation between the preset turbidity and the acquired effluent turbidity is calculated to obtain the feedback fine-tuning frequency for closed-loop control. The feedforward output frequency and the feedback fine-tuning frequency are added to obtain the final command. The corresponding variable frequency drive signal is generated according to the final command to control the variable frequency metering pump to perform the variable frequency dosing of flocculant.

[0033] Reference Figure 1 A global discrete time series is established within the main control unit, and a unified discrete sampling period is set. In the defined execution program, all parameter acquisition actions are performed at each discrete time step. Synchronous triggering within a fixed time window to align timestamps.

[0034] The main control unit acquires the simulated conductivity signal of the solution inside the preparation chamber. Due to the mechanical shearing action generated by the variable frequency stirrer during the preparation process, tiny bubbles are created within the fluid, leading to uneven local ion concentration distribution. This results in high-frequency electrical signal fluctuations in the directly acquired raw sensor data. After acquiring the raw conductivity data, the main control unit executes a moving average filtering algorithm to obtain the real-time conductivity used for state monitoring. The calculation formula is as follows: ; In the formula, Indicates real-time conductivity; Indicates the first The raw conductivity values ​​are obtained from the online conductivity sensor at discrete time steps; This represents the set sliding window length constant. The above algorithm eliminates the interference fluctuations caused by stirring at the conductivity measurement end; This represents the time step offset index variable in the summation operation.

[0035] In the same discrete time step of performing conductivity data acquisition Internally, the main control unit sends a data latching command to the measuring equipment installed on the return water side of the pipeline network, and simultaneously acquires the real-time flow rate of the return water in the pipeline network. and ultraviolet absorbance The ultraviolet absorbance value reflects the concentration of dissolved organic matter containing conjugated double bonds or aromatic ring structures in the water body. The main control unit synchronizes the real-time flow rate and ultraviolet absorbance data acquisition time through the same bus clock, so that the acquired values ​​correspond to the fluid state at the same cross-section of the physical pipe network.

[0036] At the same time, the main control unit acquires the effluent turbidity transmitted by the online turbidity meter located on the effluent side of the pipeline network. The effluent turbidity characterizes the final amount of suspended solids after treatment processes such as flocculation and filtration, and serves as a measured feedback parameter for the execution result of the system evaluation dosing command.

[0037] Real-time conductivity Real-time traffic UV absorbance and effluent turbidity In each discrete sampling period Before the process ends, the data register address of the main control unit is written uniformly so that it can be directly called by the subsequent preparation termination judgment and dosing control algorithm.

[0038] Reference Figure 1 After receiving the feeding completion signal and starting the variable frequency mixer, the main control unit performs time series difference calculation based on the acquired real-time conductivity. Specifically, the execution logic is as follows: calculate the difference between the real-time conductivity of the current discrete time step and the real-time conductivity of the previous discrete time step, and divide this difference by the value of the discrete sampling period to obtain the dissolution completion index. The mathematical formula for this calculation logic is defined as follows: ; In the formula, Indicates the current discrete time step The calculated dissolution completion index; Indicates the current discrete time step Real-time conductivity; Indicates the previous discrete time step Real-time conductivity; This indicates the discrete sampling period. The hydration process of polymeric flocculant dry powder in water is accompanied by the release of ions from the polymer molecular chains. The time derivative of the real-time conductivity, i.e., the degree of dissolution completion index, characterizes the current chemical evolution rate of the dissolution reaction.

[0039] During the calculation of the dissolution completion index, the main control unit executes adaptive shear force adjustment logic to cope with external environmental conditions such as low temperature in the preparation system that cause a decrease in hydration rate. The main control unit records the continuous running time of the variable frequency stirrer. When the system running time is less than the initial evaluation time threshold and the dissolution completion index is greater than the steady-state tolerance threshold, the main control unit obtains the first operating frequency of the variable frequency stirrer in the previous discrete time step. The main control unit adds a fixed step value to the first operating frequency to generate a second operating frequency, and outputs a frequency increase command to the variable frequency stirrer according to the second operating frequency. The calculation formula is as follows: ; In the formula, Indicates the current discrete time step The generated second operating frequency; Indicates the previous discrete time step The first operating frequency was obtained; This indicates a pre-set fixed step value.

[0040] As the flocculant hydration process continues, the ion release within the system tends towards saturation, and the value of the dissolution completion index gradually decreases. The main control unit determines whether this dissolution completion index is continuously greater than zero and less than the steady-state tolerance threshold for a preset number of sampling periods. When the dissolution completion index meets this continuous boundary condition, it indicates that the polymer ion release and molecular chain extension state have reached dynamic chemical equilibrium. The main control unit immediately generates a maturation endpoint interruption signal and sends it to the drive end of the variable frequency mixer, cutting off the motor drive circuit and stopping the variable frequency mixer.

[0041] Within the current discrete time step when the variable frequency mixer stops running, the main control unit extracts the current real-time conductivity and sets it as the steady-state conductivity of this batch of prepared reagent. The main control unit obtains the reference solvent steady-state conductivity constant and penalty weighting coefficient stored in the internal register. The steady-state conductivity constant is subtracted from the steady-state conductivity to obtain the difference value. This difference value is multiplied by the penalty weighting coefficient to obtain the penalty amount. Finally, the constant is added to the penalty amount to obtain the performance degradation coefficient. The physical model formula on which this calculation is based is: ; In the formula, Indicates the first The efficiency attenuation coefficient obtained from batch preparation of flocculants; This indicates the extraction of the set steady-state conductivity; Represents the steady-state conductivity constant of the reference solvent; This represents the penalty weighting coefficient.

[0042] The above physical model calculation logic establishes a numerical mapping between the physicochemical properties of the background solvent and the intrinsic efficacy of the reagent. The high background ion intensity in the water body induces an electrostatic shielding effect, which leads to the attenuation of the polymer's entrapment effect. The resulting physical morphological variation is quantified as a value greater than a constant one as an efficacy attenuation coefficient. The main control unit stores this efficacy attenuation coefficient in memory for subsequent process calls.

[0043] Reference Figure 1 The main control unit acquires the depth parameters inside the dosing tank and divides these parameters into multiple liquid level intervals. Specifically, it establishes a first-in-first-out (FIFO) data structure queue containing multiple nodes within the main control unit's storage space. The main control unit divides the vertical physical space of the dosing tank into equal-spaced sections based on the set total number of nodes. It then establishes a spatial correspondence between the resulting multiple liquid level intervals and the nodes in the FIFO data structure queue, using this correspondence as a data mapping queue. The formula for dividing the liquid level intervals is as follows: ; In the formula, This indicates the depth parameter obtained inside the dosing box; This represents the total number of nodes in a first-in-first-out (FIFO) data structure queue. Indicates the first The coordinates of the upper limit of the vertical height of each liquid level range; This represents the time step offset index variable in the summation operation; Indicates mathematical logic symbols as belonging to, and indicator variables. The value of is limited to the set to its right; It represents a sequence of positive integers increasing sequentially from 1 to positive integers. The discrete numerical set constituted.

[0044] During the reagent transfer process, the main control unit acquires the volume of reagent transferred to the dosing tank after a single preparation. Based on the fixed bottom area of ​​the dosing tank, the liquid level rise caused by this transfer operation is calculated, and the corresponding liquid level range is determined according to the increase in liquid level caused by the transferred reagent volume inside the dosing tank. The formula for this mapping conversion is: ; In the formula, This indicates the number of newly added nodes occupied by the transferred drug volume in the data mapping queue, with the symbol [symbol missing]. This indicates the floor function; Indicates the volume of the transferred drug obtained; This represents the constant of the bottom area inside the feeding box; The depth parameter inside the dosing box (i.e., the total height); This indicates the total number of nodes in the data mapping queue. The main control unit then writes the current batch performance degradation coefficient calculated during the preparation phase into the queue and binds it to the end of the queue. The nodes corresponding to each liquid level range.

[0045] During the dosing procedure, the main control unit integrates the operating frequency and time issued by the variable frequency metering pump to obtain the cumulative volume of consumed reagent pumped by the pump. The main control unit then determines the liquid level range corresponding to the current consumed reagent inside the dosing tank based on this volume. The specific addressing logic is as follows: The offset of the data node at the current physical suction position of the actuator is calculated based on the liquid level drop caused by the consumed reagent volume within the dosing tank. The formula for calculating the number of offset nodes is: ; In the formula, This indicates the cumulative number of nodes popped from the head of the data mapping queue corresponding to the liquid level range of the currently consumed medicine; Indicates the volume of consumed medicine obtained; This represents the constant of the bottom area inside the feeding box; The depth parameter inside the dosing box (i.e., the total height); This represents the total number of nodes in the data mapping queue.

[0046] The main control unit reads the performance degradation coefficient stored in the corresponding node from the data mapping queue based on the current head node position. During the continuous pumping process of the variable frequency metering pump, as... The increase, As the numerical values ​​increase, the head node of the data mapping queue dequeues and releases memory sequentially according to the first-in-first-out rule. The main control unit then sequentially calls the performance decay coefficient stored in the memory of the dequeued node for use by subsequent dosing operations. Through the above discrete space segmentation and data push-and-pop operations, the performance parameters called always precisely correspond to the physical liquid level when the dosing action consumes different batches of medicine at different time points.

[0047] Reference Figure 1 The main control unit synchronously extracts the real-time flow rate and UV absorbance values ​​at the current discrete time step. The main control unit performs a multiplication operation, multiplying the real-time flow rate value by the UV absorbance value to obtain the organic load factor. The organic load factor represents the total amount of dissolved organic matter entering the water treatment system per unit time and serves as the input parameter for the feedforward control loop. The formula corresponding to its calculation logic is as follows: ; In the formula, Indicates the current discrete time step The calculated organic loading factor; A numerical value representing real-time traffic; This represents the value of ultraviolet absorbance.

[0048] After obtaining the organic load factor, the main control unit reads the preset basic feedforward gain from the memory and obtains the performance decay coefficient corresponding to the current execution stage retrieved from the data mapping queue. The main control unit multiplies the retrieved performance decay coefficient by the basic feedforward gain to obtain the dynamic feedforward gain coefficient. Subsequently, the main control unit multiplies the dynamic feedforward gain coefficient by the organic load factor to obtain the feedforward output frequency used to drive the execution end action. This feedforward output frequency value constitutes the basic calculation result for compensating for the attenuation of water influent load and reagent physical state.

[0049] Synchronously, the main control unit performs feedback closed-loop calculations. The main control unit obtains the set turbidity from the memory, subtracts the obtained effluent turbidity from the set turbidity, and obtains the turbidity deviation at the current discrete time step.

[0050] The main control unit acquires the pre-set proportional coefficient, integral coefficient, and derivative coefficient. During the calculation process, the main control unit multiplies the proportional coefficient and turbidity deviation to obtain the proportional control term value; it also acquires the historical cumulative value of the turbidity deviation over all discrete time steps from startup to the current time, and multiplies the integral coefficient and this historical cumulative value to obtain the integral control term value.

[0051] Furthermore, the main control unit extracts the turbidity deviation of the current discrete time step and subtracts the turbidity deviation of the previous discrete time step to obtain the deviation change. The main control unit divides this deviation change by the value of the discrete sampling period to obtain the deviation change rate. Subsequently, the differential coefficient is multiplied by the deviation change rate to obtain the value of the differential control term.

[0052] The main control unit adds the calculated values ​​of the proportional control term, integral control term, and derivative control term to obtain the feedback fine-tuning frequency. The discrete output formula integrating the above feedforward and feedback calculation processes is as follows: ; In the formula, This represents the calculated feedback fine-tuning frequency; This represents the calculated turbidity deviation. This represents the turbidity deviation at the previous discrete time step; Indicates the proportionality coefficient; Indicates the integral coefficient; Represents the differential coefficient; Indicates the discrete sampling period; This represents the time step offset index variable in the summation operation; Indicates the current discrete time step; Indicates the first The turbidity deviation recorded at each historical discrete time step; the three terms on the right side of the formula correspond to the values ​​of the proportional control term, integral control term, and derivative control term, respectively.

[0053] After completing the calculations for the feedforward and feedback channels, the main control unit adds the generated feedforward output frequency to the feedback fine-tuning frequency to obtain the final command used to control the actuator's actions.

[0054] The main control unit converts the value corresponding to the final command into the target operating frequency value. Subsequently, the signal generation module inside the main control unit converts the target operating frequency value into a corresponding pulse frequency electrical signal. The main control unit transmits the pulse frequency electrical signal to the inverter input terminal driving the variable frequency metering pump via the communication bus. After receiving the pulse frequency electrical signal, the inverter generates a corresponding AC drive waveform according to the target operating frequency value, driving the motor inside the variable frequency metering pump to perform the physical dosing operation of the liquid reagent.

[0055] Specific application examples: Figure 3 This is a comparative data curve of effluent turbidity under disturbed conditions in a specific application embodiment of the present invention. The horizontal axis represents the system running time (minutes), and the vertical axis represents the measured effluent turbidity (NTU). The solid line represents the trajectory of effluent turbidity change using the intelligent control method of the present invention, and the dashed line represents the trajectory of effluent turbidity change using traditional fixed-time preparation and single feedback control. The dotted line is the set target baseline.

[0056] Figure 4 This is a comparative data curve of the target operating frequency of the variable frequency metering pump under disturbance conditions in a specific application embodiment of the present invention. The horizontal axis represents the system operating time (minutes), and the vertical axis represents the target operating frequency (Hz) sent to the frequency converter; the solid line represents the frequency command trajectory generated by the scheme of the present invention under the feedforward-feedback cascade architecture, and the dashed line represents the frequency command trajectory generated under the traditional single closed-loop architecture.

[0057] This embodiment is validated using the water treatment system of a large indoor public heated swimming pool. The pool has a total water volume of 2000 cubic meters, and the associated circulating water treatment system is designed to have a flow rate of 400 cubic meters per hour. The system uses a polymeric flocculant (such as polyacrylamide) for micro-flocculation filtration.

[0058] The discrete sampling period of the main control unit is set to 5 seconds. The target effluent turbidity is set to 0.5 NTU. The physical volume of the preparation tank is 1 cubic meter, the total height of the dosing tank is 1.2 meters, and the number of data mapping queue nodes is divided into 20 liquid level intervals.

[0059] In conventional systems that do not employ this invention, flocculant preparation uses a fixed timed stirring mode (set to 45 minutes per stirring cycle), without monitoring the physicochemical state of the dissolution process. Dosing control employs a single closed-loop feedback PID control based on effluent turbidity, adjusting the frequency of the variable frequency metering pump solely based on the deviation between the effluent turbidity measured by the online turbidity meter and the set value.

[0060] After starting the variable frequency mixer, the conductivity inside the preparation chamber was monitored in real time and differential calculations were performed. At the 32-minute mark, the dissolution completion index continuously fell below the steady-state tolerance threshold, indicating that the reagent had been fully hydrated. The mixer was immediately shut down (avoiding the ineffective stirring and molecular chain breakage caused by mechanical shearing during the remaining 13 minutes in the traditional method). At this point, the steady-state conductivity was extracted, and the efficiency attenuation coefficient affected by background ions in the water was calculated to be 1.08 for this batch.

[0061] Data mapping: When this batch of reagent is transferred to the dosing tank, the main control unit binds the liquid level range it occupies with the coefficient 1.08 in the data mapping queue.

[0062] During the dosing phase: The system synchronously collects the return water flow rate and UV absorbance of the pipeline network and calculates the organic load factor. When the metering pump extracts the reagent within this liquid level range, the system calls a coefficient of 1.08 to correct the basic feedforward gain, and performs feedforward-feedback cascade operations in conjunction with the measured turbidity deviation, ultimately outputting a pulse frequency to control the metering pump.

[0063] To verify the system's robustness under strong external disturbances, after the system had been running smoothly for 120 minutes, a peak in pool visitor flow was simulated, introducing a large amount of organic pollutants into the pool water, causing fluctuations in the return water flow and a sharp increase in ultraviolet absorbance. System operation data was recorded for 480 consecutive minutes.

[0064] Combination Figure 3 Analysis of the effluent turbidity control effect showed that during the stable period of 0-120 minutes, both schemes could maintain the effluent turbidity at around the set value of 0.5 NTU. However, after a sudden increase in passenger flow at the 120-minute mark, the traditional scheme, which relied solely on turbidity feedback from the effluent side, experienced a delay, leading to effluent turbidity overshoot, with a peak value reaching 1.35 NTU. Furthermore, during the subsequent recovery process, there was a decaying oscillation lasting for more than two hours.

[0065] The peak turbidity of the effluent from the present invention only slightly increased to 0.62 NTU, and then quickly converged and operated close to the set line of 0.5 NTU. This indicates that the system has withstood external load shocks.

[0066] Combination Figure 4Analyzing the response to the dosing command (inverter operating frequency), at the instant of the load change at the 120-minute mark, the present invention's solution rapidly senses the disturbance through a feedforward loop composed of flow rate and ultraviolet absorbance. The feedforward output frequency increases instantaneously, causing the overall target operating frequency to exhibit a step-like advance compensation response, thus injecting sufficient reagent into the pipeline network in advance. Simultaneously, by utilizing the efficiency attenuation coefficient (1.08) to amplify the gain, the efficiency loss of this batch of reagent is compensated for.

[0067] In the initial stage of a sudden load change, the frequency converter hardly operates. It is not until the high turbidity water reaches the outlet side (about 20 minutes later) that the feedback mechanism begins to passively increase the operating frequency. By this time, the pollutants have already penetrated the filter sand tank. Furthermore, due to the lack of compensation for the decline in the effectiveness of the chemicals, the pump frequency command continues to oscillate irregularly.

Claims

1. A smart control method for the preparation process of pool water flocculant, characterized in that, Includes the following steps: The real-time conductivity of the solution in the preparation tank, the real-time flow rate of the return water in the pipeline network, the ultraviolet absorbance, and the turbidity of the effluent were obtained. The real-time conductivity is subjected to time series difference operation to obtain the dissolution completion index. When the dissolution completion index falls within the steady-state tolerance threshold, the variable frequency mixer is shut down, and the current real-time conductivity is set as the steady-state conductivity. Based on the steady-state conductivity, the efficiency decay coefficient is obtained through a physical model. A data mapping queue is established to bind the efficiency attenuation coefficient to the liquid level of the reagent transferred to the dosing tank. When the variable frequency metering pump extracts the reagent, the corresponding efficiency attenuation coefficient is retrieved from the data mapping queue. The organic load factor is obtained based on the real-time flow rate and the ultraviolet absorbance. The dynamic feedforward gain coefficient is obtained by multiplying the efficiency attenuation coefficient by the basic feedforward gain. The feedforward output frequency is obtained by multiplying the dynamic feedforward gain coefficient by the organic load factor. The feedback fine-tuning frequency is obtained based on the set turbidity and the effluent turbidity. The feedforward output frequency and the feedback fine-tuning frequency are added together to obtain the final command. The variable frequency metering pump is controlled according to the final command.

2. The intelligent control method for the preparation process of pool water flocculant according to claim 1, characterized in that, The dissolution completion index is obtained by performing time series difference calculation on the real-time conductivity, specifically including: Calculate the difference between the real-time conductivity at the current discrete time step and the real-time conductivity at the previous discrete time step; The dissolution completion index is obtained by dividing the difference by the value of the discrete sampling period.

3. The intelligent control method for the preparation process of pool water flocculant according to claim 1, characterized in that, The specific steps for shutting down the variable frequency mixer when the solubility index falls within the steady-state tolerance threshold include: Determine whether the dissolution completion index is greater than zero and less than the steady-state tolerance threshold for a consecutive preset number of sampling periods; When the dissolution completion index is greater than zero and less than the steady-state tolerance threshold for a preset number of consecutive sampling periods, a maturation endpoint interruption signal is generated. The maturation endpoint interruption signal is sent to the drive end of the variable frequency mixer to stop the variable frequency mixer from running.

4. The intelligent control method for the preparation process of pool water flocculant according to claim 1, characterized in that, Based on the aforementioned steady-state conductivity, the efficiency attenuation coefficient is derived through a physical model, specifically including: Obtain the steady-state conductivity constant and penalty weighting coefficient of the reference solvent; The difference is obtained by subtracting the steady-state conductivity constant from the steady-state conductivity. The penalty amount is obtained by multiplying the difference value by the penalty weight coefficient; The efficiency decay coefficient is obtained by adding the constant to the penalty amount.

5. The intelligent control method for the preparation process of pool water flocculant according to claim 1, characterized in that, The establishment of the data mapping queue specifically includes: Obtain the depth parameters inside the dosing box; The depth parameter is divided into multiple liquid level intervals; Establish a first-in-first-out data structure queue containing multiple nodes within the storage space of the main control unit; Assign the multiple liquid level ranges to the multiple nodes, and use the first-in-first-out data structure queue as the data mapping queue.

6. The intelligent control method for the preparation process of pool water flocculant according to claim 1, characterized in that, Linking the performance attenuation coefficient to the liquid level of the reagent transferred to the dosing tank, and specifically retrieving the corresponding performance attenuation coefficient from the data mapping queue when the variable frequency metering pump extracts the reagent, includes: Obtain the volume of the transferred reagent after a single preparation is completed and transferred into the dosing tank; The corresponding liquid level range is determined based on the increase in liquid level inside the dosing tank caused by the volume of the transferred agent. Obtain the cumulative volume of consumed reagent pumped by the variable frequency metering pump; The liquid level range corresponding to the currently consumed agent inside the dosing tank is determined based on the volume of the consumed agent, and the efficiency decay coefficient stored in the corresponding node is read from the data mapping queue.

7. The intelligent control method for the preparation process of pool water flocculant according to claim 1, characterized in that, The organic loading factor is derived from the real-time flow rate and the ultraviolet absorbance, specifically including: Simultaneously extract the values ​​of the real-time flow rate and the ultraviolet absorbance at the current discrete time step; Multiplying the real-time flow rate by the ultraviolet absorbance value yields the organic load factor, which represents the total amount of dissolved organic matter entering the water treatment system per unit time.

8. The intelligent control method for the preparation process of pool water flocculant according to claim 1, characterized in that, The specific steps of deriving the feedback fine-tuning frequency based on the set turbidity and the effluent turbidity include: The turbidity deviation is obtained by subtracting the effluent turbidity from the set turbidity. Obtain the pre-set proportional coefficient, integral coefficient, and differential coefficient; The proportionality coefficient and the turbidity deviation are multiplied to obtain the proportional control term value; The integral control term value is obtained by multiplying the integral coefficient and the historical cumulative value of the turbidity deviation. Obtain the rate of change of the turbidity deviation between the current discrete time step and the previous discrete time step; The differential control term value is obtained by multiplying the differential coefficient and the deviation change rate. The feedback fine-tuning frequency is obtained by adding the values ​​of the proportional control term, the integral control term, and the derivative control term.

9. The intelligent control method for the preparation process of pool water flocculant according to claim 1, characterized in that, Controlling the variable frequency metering pump according to the final instruction specifically includes: Convert the final instruction into the target operating frequency value; The target operating frequency value is converted into a corresponding pulse frequency electrical signal; The pulse frequency electrical signal is transmitted to the frequency converter that drives the variable frequency metering pump through the communication bus, instructing the frequency converter to drive the variable frequency metering pump to operate according to the target operating frequency value.

10. An intelligent control system for the preparation process of pool water flocculant, characterized in that, A smart control method applied to the preparation process of a pool water flocculant as described in any one of claims 1-9, comprising: The parameter acquisition module is used to obtain the real-time conductivity of the solution in the preparation tank, the real-time flow rate of the return water in the pipeline network, the ultraviolet absorbance, and the turbidity of the effluent. A preparation control module is used to perform time series difference calculation on the real-time conductivity to obtain a dissolution completion index. When the dissolution completion index falls within the steady-state tolerance threshold, the variable frequency mixer is shut down, the current real-time conductivity is set as the steady-state conductivity, and the efficiency decay coefficient is obtained based on the steady-state conductivity through a physical model. The data mapping module is used to establish a data mapping queue, bind the efficiency attenuation coefficient to the liquid level of the reagent transferred to the dosing tank, and call the corresponding efficiency attenuation coefficient from the data mapping queue when the variable frequency metering pump extracts the reagent. The dosing calculation execution module is used to derive the organic load factor based on the real-time flow rate and the ultraviolet absorbance, multiply the called efficiency attenuation coefficient by the basic feedforward gain to obtain the dynamic feedforward gain coefficient, multiply the dynamic feedforward gain coefficient by the organic load factor to obtain the feedforward output frequency, derive the feedback fine-tuning frequency based on the set turbidity and the effluent turbidity, add the feedforward output frequency and the feedback fine-tuning frequency to obtain the final command, and control the variable frequency metering pump according to the final command.

Images