Adaptive control method and system for pure water preparation equipment

Abstract

The present application relates to the technical field of industrial process adaptive control, and particularly relates to an adaptive control method and system for a pure water preparation device, the method comprising: obtaining an online water quality monitoring value, a pure water flow value, a valve action time, a rechecked water quality monitoring value of a water storage path, and a target path effective volume of a water storage valve, a discharge valve and a backflow valve; updating an accumulated flow value according to the pure water flow value, generating a front boundary of an unqualified water section and a rear boundary of the unqualified water section based on the online water quality monitoring value, determining a boundary generation time and a corresponding accumulated flow value, and generating a water quality state boundary queue according to corresponding valve positions of the water storage valve, the discharge valve and the backflow valve. The present application can improve the matching between valve switching timing and actual water body transportation state, reduce waste caused by continuous discharge or backflow of qualified water sections, and improve water storage accuracy and adaptive control stability under water quality fluctuation.

Description

Technical Field

[0001] This invention relates to the field of adaptive control technology for industrial processes, and specifically to an adaptive control method and system for a pure water preparation equipment. Background Technology

[0002] Pure water production equipment typically obtains pure water that meets usage requirements through treatment units such as reverse osmosis, ion exchange, and precision filtration, and then completes subsequent treatment through paths such as water storage, discharge, or reflux. When the equipment starts or stops, the membrane module status fluctuates, the influent water quality is disturbed, or the flow rate changes, the effluent water quality may become unstable for a short period of time. If the control response is delayed, substandard water can easily enter the water storage side, affecting the safety of subsequent water use.

[0003] Existing control methods mostly rely on real-time water quality results from monitoring points to switch valves, or use fixed delays to compensate for the delivery lag between the monitoring point and the valve position. While these methods can meet basic control requirements when flow is stable and pipeline conditions are fixed, they can easily lead to discrepancies between the actual water body position corresponding to the monitoring results and the valve switching timing when pure water flow fluctuates, pipeline water volume changes, or valve responses differ.

[0004] Furthermore, existing equipment typically prioritizes alarms and discharge of substandard water, neglecting to adequately control the timing of water storage after the water quality has recovered from substandard to acceptable levels. If the recovery is too early, residual substandard water may still enter the storage side; if the recovery is too late, acceptable water may be continuously discharged or backflowed, reducing the utilization rate of pure water. Therefore, improving the matching between valve switching timing and the actual water delivery state during water quality fluctuations is a specific technical problem that needs to be solved in the adaptive control of pure water preparation equipment. Summary of the Invention

[0005] To address the aforementioned shortcomings of existing technologies, this invention provides an adaptive control method and system for a pure water preparation device, which can effectively solve the problems of easy deviation in valve switching timing and insufficient control of water storage timing in existing technologies.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides an adaptive control method for a pure water preparation device, comprising:

[0008] S1: Obtain online water quality monitoring values, pure water flow values, valve action time, verification water quality monitoring values ​​of the water storage path, and the effective volume of the target path of the water storage valve, discharge valve, and return valve;

[0009] S2: Update the cumulative flow value according to the pure water flow value, perform state transition detection based on the online water quality monitoring value to generate the front boundary and the back boundary of the unqualified water section, determine the boundary generation time and its corresponding cumulative flow value, and generate a water quality state boundary queue according to the corresponding valve positions of the water storage valve, discharge valve and return valve.

[0010] S3: Based on the cumulative flow rate value corresponding to the boundary generation time, perform flow rate integration on the pure water flow rate value to determine the cumulative transport volume;

[0011] S4: Based on the cumulative transport volume and the effective volume of the target path, propagation prediction is performed to generate a prediction result of the water quality state boundary reaching the corresponding valve position;

[0012] S5: Based on the prediction results and valve action time, perform advance trigger control to generate pre-action control commands, so that the unqualified water section enters the discharge path or return path, and the qualified water sections before and after the unqualified water section enter the water storage path.

[0013] S6: Based on the verified water quality monitoring values, perform feedback error correction on the prediction results corresponding to the water storage valve, determine the water quality state boundary prediction deviation, and update the effective volume of the target path and valve action time corresponding to the water storage valve based on the water quality state boundary prediction deviation. Use the updated parameters for propagation prediction and early trigger control in subsequent detection cycles.

[0014] In a second aspect, the present invention also provides an adaptive control system for a pure water preparation device, applied to the adaptive control method for the pure water preparation device as described in the first aspect, the system comprising:

[0015] The parameter acquisition module acquires online water quality monitoring values, pure water flow values, valve action time, verification water quality monitoring values ​​of the water storage path, and the effective volume of the target path of the water storage valve, discharge valve, and return valve.

[0016] The water quality boundary queue generation module updates the cumulative flow value based on the pure water flow value, performs state transition detection based on the online water quality monitoring value to generate the front boundary and the back boundary of the unqualified water segment, determines the boundary generation time and its corresponding cumulative flow value, and generates a water quality state boundary queue according to the corresponding valve positions of the water storage valve, discharge valve and return valve.

[0017] The cumulative delivery volume determination module uses the cumulative flow rate value corresponding to the boundary generation time as a benchmark to perform flow rate integration on the pure water flow rate value to determine the cumulative delivery volume.

[0018] The boundary propagation prediction module performs propagation prediction based on the cumulative transport volume and the effective volume of the target path, and generates a prediction result of the water quality state boundary reaching the corresponding valve position.

[0019] The valve pre-action control module performs advance trigger control based on the prediction results and valve action time, generates pre-action control commands, causes the unqualified water section to enter the discharge path or return path, and causes the qualified water sections before and after the unqualified water section to enter the water storage path.

[0020] The feedback error correction module corrects the prediction results corresponding to the water storage valve based on the verified water quality monitoring values, determines the water quality state boundary prediction deviation, and updates the effective volume of the target path and the valve action time corresponding to the water storage valve based on the water quality state boundary prediction deviation. The updated parameters are then used for propagation prediction and early trigger control in subsequent detection cycles.

[0021] The technical solution provided by this invention has the following beneficial effects:

[0022] This invention transforms the water quality state switching process corresponding to online water quality monitoring values ​​into the front and rear boundaries of the unqualified water segment. It then predicts the propagation state of the water quality state boundary at the corresponding valve positions of the storage valve, discharge valve, and return valve by combining pure water flow rate, cumulative flow rate, and the effective volume of the target path. This allows valve control to no longer rely solely on instantaneous water quality results or fixed delay estimations at the detection point, improving the matching between valve switching timing and the actual water transport state. Furthermore, this invention generates pre-action control commands based on the prediction results and valve action time. This facilitates blocking the storage path and switching to the discharge or return path before the unqualified water segment arrives, and timely restoration of the storage path after the rear boundary of the unqualified water segment is crossed. This reduces the probability of unqualified water entering the storage path and minimizes waste caused by continuous discharge or return of qualified water segments. Meanwhile, this invention uses the verified water quality monitoring value of the water storage path to correct the error of the prediction result corresponding to the water storage valve, and uses the updated target path effective volume and valve action time for subsequent detection cycles, so that the pure water preparation equipment can adjust parameters according to flow fluctuations, path state changes and valve response differences, thereby improving the water storage accuracy and adaptive control stability under water quality fluctuation conditions. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0024] Figure 1 This is a schematic diagram illustrating the steps of an adaptive control method for a pure water preparation device provided in an embodiment of the present invention;

[0025] Figure 2 A schematic flowchart of an adaptive control method for a pure water preparation device provided in an embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram of the structure of an adaptive control system for a pure water preparation device provided in an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of 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, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0028] Optional, see below Figure 1 and Figure 2 , Figure 1 This is a schematic diagram illustrating the steps of the adaptive control method for the pure water preparation equipment provided by the present invention. Figure 2 This is a flowchart illustrating the adaptive control method for a pure water preparation device provided by the present invention. In this embodiment, the entity executing the adaptive control method for the pure water preparation device is the adaptive control system of the pure water preparation device. The adaptive control method for the pure water preparation device includes:

[0029] Step S1: Obtain online water quality monitoring values, pure water flow rate values, valve action time, verification water quality monitoring values ​​of the water storage path, and the effective volume of the target path of the water storage valve, discharge valve, and return valve.

[0030] S1 is used to establish the adaptive control data foundation for the pure water preparation equipment, enabling subsequent state transition detection, water quality state boundary queue generation, propagation prediction, early trigger control, and feedback error correction to be executed based on data within the same detection cycle. Online water quality monitoring values ​​originate from online water quality monitoring points and reflect the water quality state of pure water before it enters the storage path, discharge path, and return path. When acquiring online water quality monitoring values, the values ​​are associated with and saved along with the corresponding detection time, allowing subsequent steps to generate the front and rear boundaries of the unqualified water segment based on changes in water quality state at adjacent detection times.

[0031] The pure water flow rate value is obtained from the pure water flow rate monitoring unit and is used to update the cumulative flow rate value. It also provides data for the cumulative flow rate value at the boundary generation time, the current cumulative flow rate value, and the cumulative transport volume. When acquiring the pure water flow rate value, it is associated with and saved with the corresponding detection time, so that the advancement status of the water quality state boundary on the target treatment path can be updated as the actual transport volume changes.

[0032] Valve action time includes the response time of the water storage valve, discharge valve, and return valve in the corresponding action direction. It is used to characterize the time required for the valve to reach the target open / closed state from receiving the control command. When acquiring the valve action time, the valve action time is associated with and saved with the corresponding valve position and corresponding action direction. This allows subsequent steps to determine the advance trigger window based on the valve action time and generate pre-action control commands that match the valve opening and closing actions.

[0033] The verification water quality monitoring values ​​for the water storage path are derived from the verification water quality monitoring points along the path and are used to reflect the actual changes in water quality after the water enters the storage path. When obtaining the verification water quality monitoring values ​​for the water storage path, the values ​​are associated with and saved along with the corresponding detection times. This allows subsequent steps to extract the occurrence times of the actual front and rear boundaries and compare them with the prediction results corresponding to the water storage valve.

[0034] The target path effective volumes of the water storage valve, discharge valve, and return valve correspond to the water transport paths from the online water quality monitoring point to the water storage valve, discharge valve, and return valve, respectively. These volumes characterize the effective transport volume required for the water quality boundary to propagate to the corresponding valve location. When obtaining the target path effective volume, it is associated with and saved to the corresponding valve location, enabling subsequent steps to retrieve the corresponding target path effective volume and determine the remaining transport volume.

[0035] Step S2: Update the cumulative flow value according to the pure water flow value, perform state transition detection based on the online water quality monitoring value to generate the front boundary and the back boundary of the unqualified water segment, determine the boundary generation time and its corresponding cumulative flow value, and generate a water quality state boundary queue according to the corresponding valve positions of the water storage valve, discharge valve and return valve.

[0036] S2 is used to convert changes in water quality status based on online water quality monitoring values ​​into controllable objects for sustainable maintenance. In practice, the cumulative flow rate is continuously updated based on the pure water flow rate, and the water quality status at the time of detection is determined based on the online water quality monitoring values. When the water quality status changes from acceptable to unacceptable, a front boundary of the unacceptable water segment is generated; when the water quality status changes from unacceptable to acceptable, a rear boundary of the unacceptable water segment is generated. The boundary generation time is used to identify the time location where the water quality status change occurs, and the cumulative flow rate value corresponding to the boundary generation time is used to identify the volume location when the water quality status boundary is generated.

[0037] Furthermore, the front and rear boundaries of the unqualified water section are used as water quality state boundaries, and water quality state boundary records are formed according to the corresponding valve positions of the storage valve, discharge valve, and return valve. Multiple water quality state boundary records form a water quality state boundary queue. Through S2, the same water quality state boundary can be propagated and maintained for the storage valve, discharge valve, and return valve respectively, enabling subsequent steps to generate separate prediction results for different valve positions. See steps S21-S23 for details.

[0038] Step S3: Using the cumulative flow rate value corresponding to the boundary generation time as a benchmark, perform flow rate integration on the pure water flow rate value to determine the cumulative transport volume.

[0039] S3 is used to determine the volume of water quality state boundary that advances to the corresponding valve position after its generation during the pure water transport process. In practice, within each detection cycle, the current cumulative flow rate is updated based on the pure water flow rate, and the cumulative flow rate corresponding to the boundary generation time of the water quality state boundary is read from the water quality state boundary queue. The difference between the current cumulative flow rate and the cumulative flow rate corresponding to the boundary generation time is determined as the cumulative transport volume.

[0040] Once the cumulative transport volume is written into the water quality state boundary queue, it can serve as a direct input for predicting the propagation of the water quality state boundary. Through S3, the advancement state of the water quality state boundary is not estimated based on a fixed time delay, but rather updated according to the actual transport volume formed by the pure water flow rate, which is beneficial for adapting to changes in the pure water flow rate within different detection cycles. See steps S31-S32 for details.

[0041] Step S4: Based on the cumulative transport volume and the effective volume of the target path, propagation prediction is performed to generate a prediction result of the water quality state boundary reaching the corresponding valve position.

[0042] S4 is used to determine the arrival status of the water quality state boundary relative to the storage valve, discharge valve, and return valve. In practice, the effective volume of the target path is retrieved based on the corresponding valve position in the water quality state boundary record, and the difference between the effective volume of the target path and the cumulative transport volume is determined as the remaining transport volume. The remaining transport volume reflects the amount of volume still needed to be transported from the water quality state boundary to the corresponding valve position.

[0043] Furthermore, based on the remaining transport volume of the previous detection cycle and the remaining transport volume of the current detection cycle, it is determined whether the water quality state boundary crosses the corresponding valve position. When it is determined that the water quality state boundary crosses the corresponding valve position, the arrival time is determined based on the pure water flow rate value from the previous detection cycle to the current detection cycle. The resulting prediction result includes the remaining transport volume, arrival determination result, arrival time, and corresponding valve position. The prediction result is used for subsequent early trigger control and water storage path feedback error correction. See steps S41-S44 for details.

[0044] Step S5: Based on the prediction results and valve action time, perform advance trigger control to generate a pre-action control command, so that the unqualified water section enters the discharge path or return path, and the qualified water sections before and after the unqualified water section enter the water storage path.

[0045] S5 is used to convert the propagation prediction results of the water quality state boundary into valve control actions. In specific implementation, the action delivery volume of the corresponding valve is determined based on the pure water flow rate value during the valve action time, and the action delivery volume is used as the advance trigger window of the corresponding valve. The advance trigger window is used to determine whether the pre-action control command needs to be issued before the water quality state boundary reaches the corresponding valve position, so as to compensate for the response process required for the valve to reach the target opening and closing state from receiving the control command.

[0046] For the upstream boundary of the substandard water section, when the remaining transport volume of the corresponding water storage valve and the valve to be opened enters the advance trigger window of the corresponding valve, a pre-action control command is generated to close the water storage valve and open the valve to be opened, causing the substandard water section to enter the discharge path or return path. For the downstream boundary of the substandard water section, when the remaining transport volume of the corresponding water storage valve and the valve to be opened enters the advance trigger window of the corresponding valve, a pre-action control command is generated to close the valve to be opened and restore the opening of the water storage valve, causing the qualified water sections before and after the substandard water section to enter the water storage path. The pre-action control command is associated with the prediction result of the corresponding water quality state boundary to support subsequent feedback error correction. See steps S51-S54 for details.

[0047] Step S6: Based on the verified water quality monitoring value, the prediction result corresponding to the water storage valve is corrected by feedback error, the water quality state boundary prediction deviation is determined, and the effective volume of the target path and the valve action time corresponding to the water storage valve are updated based on the water quality state boundary prediction deviation. The updated parameters are used for propagation prediction and early triggering control in subsequent detection cycles.

[0048] S6 is used to correct control parameters in subsequent monitoring cycles using actual water quality feedback from the water storage path. In practice, the actual water quality state switching process in the water storage path is determined based on the verified water quality monitoring values, and the occurrence times of the actual preceding and following boundaries are extracted from this process. The occurrence time of the actual preceding boundary is compared with the arrival time in the predicted result of the preceding boundary of the unqualified water section corresponding to the water storage valve to obtain the preceding boundary prediction deviation; similarly, the occurrence time of the actual following boundary is compared with the arrival time in the predicted result of the following boundary of the unqualified water section corresponding to the water storage valve to obtain the following boundary prediction deviation.

[0049] Furthermore, the parameter correction amount corresponding to the closure of the water storage valve is determined based on the preceding boundary prediction deviation, and the parameter correction amount corresponding to the reopening of the water storage valve is determined based on the following boundary prediction deviation. The effective volume of the target path and the valve action time corresponding to the water storage valve are then updated based on the parameter correction amounts. The updated parameters are incorporated into the propagation prediction and early trigger control of subsequent detection cycles, enabling the pure water preparation equipment to correct the valve pre-action trigger conditions based on the actual water quality state switching results in the water storage path, thereby improving the matching between the valve switching timing and the actual water delivery state under fluctuating water quality conditions. See steps S61-S67 for details.

[0050] Compared with related technologies, the present invention has at least the following beneficial effects:

[0051] Firstly, this invention transforms the changes in water quality status corresponding to online water quality monitoring values ​​into the front boundary and the back boundary of unqualified water sections, thereby expanding the control object from the immediate water quality results at the detection point to the water quality status boundary that can be continuously tracked along the target treatment path, which is beneficial to improving the targeting of control during water quality fluctuations.

[0052] Secondly, the present invention updates the cumulative flow value based on the pure water flow value and determines the cumulative transport volume based on the cumulative flow value corresponding to the boundary generation time, so that the advancement state of the water quality state boundary can be updated with the actual transport volume change, which is beneficial to reduce the timing offset caused by fixed delay control under flow fluctuation state.

[0053] Third, this invention combines the cumulative transport volume and the effective volume of the target path to generate a prediction result of the water quality state boundary reaching the corresponding valve position, so that the water storage valve, discharge valve and return valve can form a propagation prediction basis for different valve positions, which is conducive to improving the matching between the valve switching timing and the actual water transport state.

[0054] Fourth, the present invention generates pre-action control commands based on the prediction results and valve action time, so that the water storage path can be blocked and the discharge path or return path can be switched before the front boundary of the unqualified water section is reached, and the water storage path can be restored after the rear boundary of the unqualified water section is crossed. This helps to reduce the probability of unqualified water entering the water storage path and reduce the amount of water continuously discharged or returned from the qualified water section.

[0055] Fifth, the present invention corrects the error of the prediction result corresponding to the water storage valve based on the verified water quality monitoring value of the water storage path, and updates the effective volume of the target path and the valve action time corresponding to the water storage valve based on the prediction deviation of the water quality state boundary. This allows the updated parameters to be used for propagation prediction and early triggering control in subsequent detection cycles, which is beneficial to improving the adaptive control stability of pure water preparation equipment under water quality fluctuation conditions.

[0056] In one embodiment, steps S21-S23 are described as follows:

[0057] S21. Based on the water quality qualification threshold, the online water quality monitoring values ​​at each detection time are determined to be either qualified or unqualified.

[0058] In practice, online water quality monitoring values ​​and pure water flow rates are acquired within each monitoring cycle, and the cumulative flow rate value corresponding to the current monitoring time is updated based on the pure water flow rate value. A water quality compliance threshold is used to classify the online water quality monitoring values ​​into different states. When the online water quality monitoring value meets the compliance conditions corresponding to the water quality compliance threshold, the water quality status at the corresponding monitoring time is recorded as compliant; when the online water quality monitoring value does not meet the compliance conditions corresponding to the water quality compliance threshold, the water quality status at the corresponding monitoring time is recorded as non-compliant.

[0059] Furthermore, the online water quality monitoring values, water quality status, monitoring time, and corresponding cumulative flow values ​​at each monitoring moment are associated and saved. This allows subsequent steps to read the water quality status at adjacent monitoring moments and obtain the cumulative flow value corresponding to the boundary generation moment when a state transition occurs. For multiple consecutive monitoring moments, the water quality status is arranged in chronological order to form a water quality status sequence for state transition detection. The first monitoring moment is used to establish the initial water quality status, without generating the front and rear boundaries of the unqualified water segment. Starting from the second monitoring moment, the water quality status at the current monitoring moment is compared with the water quality status at the previous monitoring moment.

[0060] S22, compare the water quality status at adjacent detection times. When the water quality status changes from qualified to unqualified, generate the front boundary of the unqualified water segment; when the water quality status changes from unqualified to qualified, generate the rear boundary of the unqualified water segment, and determine the next detection time after the status change as the corresponding boundary generation time.

[0061] In practice, the water quality status at the previous detection time and the water quality status at the current detection time are read, and a state transition detection is performed between the two water quality statuses. If the water quality status at the previous detection time is qualified and the water quality status at the current detection time is unqualified, a state transition from qualified to unqualified is determined, the front boundary of the unqualified water segment is generated, and the current detection time is determined as the boundary generation time of the front boundary of the unqualified water segment. The cumulative flow value corresponding to the current detection time is determined as the cumulative flow value corresponding to the boundary generation time of the front boundary of the unqualified water segment.

[0062] If the water quality at the current testing moment is unqualified, and then becomes qualified, a state transition from unqualified to qualified is determined, generating the boundary of the unqualified water segment. The current testing moment is then designated as the boundary generation moment of this boundary. The cumulative flow value corresponding to the current testing moment is determined as the cumulative flow value corresponding to the boundary generation moment of the unqualified water segment.

[0063] When the water quality status at the previous detection moment is consistent with the current detection moment, no new front and rear boundaries for the unqualified water segment are generated; only the water quality status record and cumulative flow value record at the corresponding detection moment are maintained. Through the above processing, the front boundary of the unqualified water segment is used to identify the starting position of the unqualified water segment entering the subsequent target treatment path, and the rear boundary of the unqualified water segment is used to identify the position where the unqualified water segment ends and is restored to a qualified water segment.

[0064] S23, take the front boundary and the rear boundary of the unqualified water section as the water quality state boundary, generate water quality state boundary records according to the water storage valve, discharge valve and return valve respectively, and write the water quality state boundary, the cumulative flow value corresponding to the boundary generation time, the corresponding valve position and the effective volume of the target path into each water quality state boundary record, and generate a water quality state boundary queue.

[0065] In practice, after generating the front or rear boundary of the unqualified water segment, the generated front or rear boundary of the unqualified water segment is determined as the water quality state boundary. For a single water quality state boundary, three water quality state boundary records are established according to the storage valve, discharge valve, and return valve, respectively, so that the same water quality state boundary can be used for propagation state maintenance corresponding to different valve positions.

[0066] Each water quality status boundary record must include at least the water quality status boundary, the boundary generation time, the cumulative flow rate at the boundary generation time, the corresponding valve location, the effective volume of the target path at the corresponding valve location, and the boundary arrival status. The corresponding valve location is one of the following: storage valve, discharge valve, or return valve. The effective volume of the target path at the corresponding valve location is the effective volume of the target path from the online water quality monitoring point to the corresponding valve location. The boundary arrival status is initially set to "not arrived" when the water quality status boundary record is generated, and is used in subsequent steps to update it to "arrived" based on the cumulative transport volume and the effective volume of the target path.

[0067] Furthermore, the water quality state boundary records generated according to the storage valve, discharge valve, and return valve are written into the water quality state boundary queue, and the water quality state boundary queue is maintained in the order of boundary generation time. The water quality state boundary records in the water quality state boundary queue are used in subsequent steps to read the cumulative flow value, corresponding valve position, and effective volume of the target path at the boundary generation time, thereby determining the cumulative transport volume, remaining transport volume, arrival determination result, and arrival time.

[0068] In one embodiment, steps S31-S32 are described as follows:

[0069] S31, within each detection cycle, update the current cumulative flow value based on the pure water flow value, and read the current cumulative flow value and the cumulative flow value corresponding to the boundary generation time.

[0070] In practice, the pure water preparation equipment continuously acquires pure water flow rate values ​​according to the detection cycle and saves these values ​​in association with the corresponding detection time. For the current detection cycle, the cumulative flow rate value from the previous detection cycle is first read. Then, the change in transport volume within the detection cycle is determined based on the pure water flow rate values ​​between the previous and current detection cycles. This change in transport volume is then added to the cumulative flow rate value from the previous detection cycle to obtain the current cumulative flow rate value. The current cumulative flow rate value represents the basic cumulative transport volume of pure water after passing through the online water quality monitoring point up to the current detection time.

[0071] Furthermore, the system reads water quality state boundary records in the unreached state from the water quality state boundary queue. Each water quality state boundary record already contains the water quality state boundary, the cumulative flow value corresponding to the boundary generation time, the corresponding valve position, and the effective volume of the target path. For water quality state boundary records that need to be updated in the current detection cycle, the system reads the current cumulative flow value and the cumulative flow value corresponding to the boundary generation time from the same water quality state boundary record, making the current cumulative flow value and the cumulative flow value corresponding to the boundary generation time comparable data under the same volume benchmark.

[0072] For water quality state boundary records that have already been written as arriving in subsequent propagation predictions, the already written cumulative transport volume, remaining transport volume, arrival determination result, and arrival time can be retained, and they will no longer participate in the cumulative transport volume update of the current detection cycle. Through the above processing, water quality state boundary records that are still in the process of propagation in the water quality state boundary queue can obtain an update basis in each detection cycle, avoiding the substitution of actual transport volume changes with a fixed time delay.

[0073] S32, determine the cumulative transport volume of the corresponding water quality state boundary in the current detection cycle, and write the cumulative transport volume into the water quality state boundary queue.

[0074] In practice, the water quality state boundary records in the unreached state are traversed in the water quality state boundary queue. For any water quality state boundary record, the current cumulative flow value is subtracted from the cumulative flow value corresponding to the boundary generation time written in the water quality state boundary record to obtain the cumulative transport volume of the corresponding water quality state boundary in the current detection cycle. The cumulative transport volume is used to represent the volume of water quality state boundary that has advanced towards the corresponding valve position since the boundary generation time during the pure water transport process.

[0075] Furthermore, the obtained cumulative transport volume is written into the same water quality state boundary record, and the correlation between the cumulative transport volume and the water quality state boundary, the cumulative flow value corresponding to the boundary generation time, the corresponding valve position, and the effective volume of the target path is maintained. When the same water quality state boundary generates water quality state boundary records for storage valves, discharge valves, and return valves respectively, different water quality state boundary records can read the same cumulative flow value corresponding to the boundary generation time, and save the corresponding valve position and the effective volume of the target path respectively, so that subsequent propagation prediction can be performed for storage valves, discharge valves, and return valves respectively.

[0076] In cases where the current cumulative flow value is less than the cumulative flow value at the boundary generation time, an abnormal data state occurs. The corresponding water quality state boundary record does not generate a valid cumulative transport volume, and the propagation state already written in the previous detection cycle is maintained. This prevents errors in subsequent propagation predictions caused by abnormal cumulative flow value collection order. The cumulative transport volume written into the water quality state boundary queue in the current detection cycle is used in step S4 to compare with the effective volume of the target path, thereby determining the remaining transport volume, arrival determination result, and arrival time.

[0077] In one embodiment, steps S41-S44 are described as follows:

[0078] S41, based on the corresponding valve position recorded in the water quality state boundary queue, call the effective volume of the target path at the corresponding valve position, and determine the difference between the effective volume of the target path and the cumulative transport volume as the remaining transport volume of the current detection cycle, and write the remaining transport volume of the current detection cycle into the water quality state boundary queue.

[0079] In practice, the water quality state boundary queue is read within the current monitoring cycle, and propagation prediction is performed on water quality state boundary records where the boundary has not yet been reached. For any water quality state boundary record, the corresponding valve position is first read, and then the effective volume of the target path for the corresponding valve position among the storage valve, discharge valve, and return valve is retrieved based on the corresponding valve position. The effective volume of the target path represents the effective transport volume required for the water quality state boundary to propagate from the online water quality monitoring point to the corresponding valve position.

[0080] Furthermore, the cumulative transport volume already written in the same water quality state boundary record is read, and the effective volume of the target path is subtracted from the cumulative transport volume to obtain the remaining transport volume for the current detection cycle. A remaining transport volume greater than zero indicates that the water quality state boundary has not yet propagated to the corresponding valve position; a remaining transport volume not greater than zero indicates that the water quality state boundary has reached or passed the corresponding valve position. The remaining transport volume for the current detection cycle is written into the same water quality state boundary record and is associated with the water quality state boundary, the cumulative flow value at the boundary generation time, the cumulative transport volume, the corresponding valve position, and the effective volume of the target path.

[0081] For the same water quality state boundary, multiple water quality state boundary records are generated separately for the storage valve, discharge valve, and return valve. The effective volume of the target path at the corresponding valve position is called up for each, and the remaining transport volume for the current detection cycle is determined for each. Through this process, the front boundary or the rear boundary of the same non-compliant water section can form independent propagation prediction basis at different valve positions.

[0082] S42, read the remaining transport volume written in the previous detection cycle in the same water quality state boundary record. When the remaining transport volume in the previous detection cycle is greater than zero and the remaining transport volume in the current detection cycle is not greater than zero, generate the arrival judgment result of the water quality state boundary crossing the corresponding valve position.

[0083] In practice, after the remaining transport volume of the current detection cycle is written into the water quality state boundary record, the remaining transport volume written in the previous detection cycle is read from the same water quality state boundary record. A continuity judgment is then made between the remaining transport volume of the previous detection cycle and the remaining transport volume of the current detection cycle. If the remaining transport volume of the previous detection cycle is greater than zero, it indicates that the water quality state boundary had not yet reached the corresponding valve position in the previous detection cycle; if the remaining transport volume of the current detection cycle is not greater than zero, it indicates that the water quality state boundary has reached or crossed the corresponding valve position in the current detection cycle. When both of the aforementioned conditions are met, an arrival determination result for whether the water quality state boundary has crossed the corresponding valve position is generated.

[0084] Furthermore, the arrival determination result is written into the same water quality state boundary record, and the boundary arrival status is updated from "not arrived" to "arrived". For water quality state boundary records where the remaining transport volume in the previous detection cycle is greater than zero and the remaining transport volume in the current detection cycle is also greater than zero, the boundary arrival status is kept as "not arrived", and the cumulative transport volume and remaining transport volume are updated in subsequent detection cycles. For water quality state boundary records that have already been written with an arrival status in the previous detection cycle, arrival determination results for water quality state boundaries crossing corresponding valve positions are not generated again to avoid the same water quality state boundary being repeatedly determined relative to the same corresponding valve position.

[0085] For the detection cycle in which the remaining transport volume is first written into the water quality state boundary record, if the remaining transport volume written in the previous detection cycle does not yet exist in the same water quality state boundary record, then only the remaining transport volume of the current detection cycle is saved, and cross-cycle arrival determination is performed at the beginning of the next detection cycle. Through the above processing, the arrival determination result can reflect the event of the water quality state boundary crossing the corresponding valve position between adjacent detection cycles, rather than only reflecting the static distance state at a certain detection moment.

[0086] S43, when the arrival determination result indicates that the water quality state boundary has crossed the corresponding valve position, the arrival time of the water quality state boundary reaching the corresponding valve position is determined based on the remaining transport volume of the previous detection cycle, the remaining transport volume of the current detection cycle, and the pure water flow rate value from the previous detection cycle to the current detection cycle. See steps S431-S433 for details.

[0087] In one embodiment, steps S431-S433 are described as follows:

[0088] S431, starting from the detection time of the previous detection cycle, the pure water flow rate value from the previous detection cycle to the current detection cycle is accumulated and integrated to obtain the change in transport volume within the cycle over time.

[0089] In practice, when the judgment result indicates that the water quality state boundary has crossed the corresponding valve position, the detection time of the previous detection cycle, the detection time of the current detection cycle, the remaining transport volume of the previous detection cycle, and the pure water flow rate value from the previous detection cycle to the current detection cycle are read. Starting from the detection time of the previous detection cycle, the pure water flow rate value is accumulated and integrated in chronological order to obtain the change in transport volume over time. The change in transport volume over time is used to represent the transport volume formed from the detection time of the previous detection cycle to any time within the cycle.

[0090] When pure water flow rate values ​​are obtained using discrete sampling, the pure water flow rate values ​​between adjacent sampling times can be converted into the volume change of the corresponding sampling period according to the order of sampling times. The volume changes of the transport volume within each sampling period are then accumulated sequentially to form a time-increasing sequence of volume changes within a period. This time-increasing sequence of volume changes within a period is used to determine the time position at which the water quality boundary reaches the corresponding valve position between the previous and current detection periods.

[0091] S432, the time when the change in conveying volume within the cycle first reaches the remaining conveying volume of the previous detection cycle is determined as the arrival offset time from the previous detection cycle to the current detection cycle.

[0092] In practice, the change in transport volume within a cycle is compared with the remaining transport volume of the previous testing cycle. The remaining transport volume of the previous testing cycle represents the volume that still needs to be transported from the water quality boundary to the corresponding valve position at the testing time of the previous testing cycle; the change in transport volume within a cycle represents the transport volume already completed since the testing time of the previous testing cycle. When the change in transport volume within a cycle first reaches the remaining transport volume of the previous testing cycle, it indicates that the water quality boundary has reached the corresponding valve position.

[0093] Furthermore, the time corresponding to the first attainment of the remaining transport volume of the previous detection cycle is determined as the arrival offset time. If the pure water flow rate is obtained using discrete sampling, and the change in transport volume within the cycle crosses the remaining transport volume of the previous detection cycle between adjacent sampling times, the arrival offset time can be determined based on the pure water flow rate between adjacent sampling times and the remaining unreached transport volume, ensuring that the arrival offset time falls between the detection time of the previous detection cycle and the detection time of the current detection cycle. Through the above processing, the arrival offset time can reflect the actual arrival position of the water quality state boundary within the detection cycle, avoiding the direct use of the detection time of the current detection cycle as the arrival time.

[0094] S433, based on the detection time of the previous detection cycle and the arrival offset time, determine the arrival time of the water quality state boundary to the corresponding valve position.

[0095] In practice, the detection time of the previous detection cycle is combined with the arrival offset time to obtain the arrival time of the water quality state boundary at the corresponding valve position. The arrival time is written into the same water quality state boundary record and is associated with and saved along with the water quality state boundary, the corresponding valve position, the remaining transport volume, and the arrival determination result. The arrival time is used to indicate the specific time position at which the water quality state boundary crosses the corresponding valve position between the previous detection cycle and the current detection cycle.

[0096] For water quality state boundary records where the arrival determination result does not indicate that the water quality state boundary crosses the corresponding valve position, the arrival offset duration determination process is not performed, no new arrival time is written, and the propagation status continues to be updated in subsequent detection cycles. For water quality state boundary records where the arrival time has already been written, subsequent steps can read the arrival time for associating the pre-action control command with the prediction result of the corresponding water quality state boundary, and for comparing the prediction result corresponding to the water storage valve with the actual water quality state switching process of the water storage path.

[0097] S44 uses the remaining conveying volume, arrival determination result, arrival time, and corresponding valve position as the prediction result.

[0098] In practice, after determining the remaining transport volume, generating the arrival judgment result, and determining the arrival time for the current testing cycle, the remaining transport volume, arrival judgment result, arrival time, and corresponding valve position are combined into a prediction result, which is then written into the corresponding water quality state boundary record. For water quality state boundary records where the arrival judgment result indicates that the water quality state boundary has not yet crossed the corresponding valve position, the prediction result includes at least the remaining transport volume for the current testing cycle, the non-arrival status, and the corresponding valve position; for water quality state boundary records where the arrival judgment result indicates that the water quality state boundary has crossed the corresponding valve position, the prediction result includes the remaining transport volume for the current testing cycle, the arrival judgment result, the arrival time, and the corresponding valve position.

[0099] Furthermore, the prediction results are saved separately according to the water quality state boundary records. The prediction results corresponding to the storage valve are used to generate subsequent pre-action control commands for closing and reopening the storage valve, and are also used for feedback error correction after the formation of the water quality monitoring values ​​for the storage path verification. The prediction results corresponding to the discharge valve are used to generate subsequent pre-action control commands for opening and closing the discharge valve. The prediction results corresponding to the return valve are used to generate subsequent pre-action control commands for opening and closing the return valve.

[0100] Through steps S41-S44, each water quality state boundary record in the water quality state boundary queue can form the remaining transport volume, arrival judgment result, and arrival time according to the corresponding valve position. This allows the propagation prediction results of the water quality state boundary reaching the storage valve, discharge valve, and return valve to be maintained separately, and provides a data basis for subsequent early trigger control and feedback error correction.

[0101] In one embodiment, steps S51-S54 are described as follows:

[0102] S51, determine the corresponding valve's action delivery volume based on the pure water flow rate value during the valve's action time, and use the action delivery volume as the corresponding valve's advance trigger window.

[0103] In practice, the valve action times of the storage valve, discharge valve, and return valve are read within the current detection cycle, and the valve action times to be calculated are determined based on the valve action direction corresponding to the pre-action control command. The storage valve closing pre-action control command corresponds to the valve action time when the storage valve closes; the storage valve reopening pre-action control command corresponds to the valve action time when the storage valve reopens; the discharge valve opening pre-action control command and the discharge valve closing pre-action control command correspond to the valve action times when the discharge valve opens and closes, respectively; and the return valve opening pre-action control command and the return valve closing pre-action control command correspond to the valve action times when the return valve opens and closes, respectively.

[0104] Furthermore, using the detection time of the current detection cycle as the trigger judgment time, the volume of water that the corresponding valve continues to transport before completing its action is determined based on the pure water flow rate value during the valve's action time, and this volume is defined as the action transport volume. The action transport volume serves as an advance trigger window for the corresponding valve, used to determine whether the pre-action control command needs to be issued before the water quality state boundary reaches the corresponding valve position.

[0105] When performing the early trigger judgment, the control system determines the action delivery volume within the valve action time based on the pure water flow rate value already acquired in the current detection cycle. Specifically, the control system uses the detection time of the current detection cycle as the trigger judgment time, reads the pure water flow rate value already acquired at the trigger judgment time, and combines it with the valve action time of the corresponding valve in the corresponding action direction to determine the volume of water that continues to be delivered before the valve completes its action, and determines this volume as the action delivery volume.

[0106] In one specific embodiment, the motion delivery volume can be determined using the following process:

[0107]

[0108]

[0109] in, This indicates the valve that participates in pre-action control. It is one of the water storage valve, drain valve, and return valve;

[0110] Indicates the direction of valve movement;

[0111] Indicates the current testing cycle;

[0112] Indicates the detection time in the current detection cycle;

[0113] Indicates valve In the direction of valve movement Valve operating time;

[0114] This indicates the pure water flow rate during the valve's operating time.

[0115] Indicates valve The corresponding valve action direction within the current detection cycle. The action conveying volume, and serves as the advance trigger window for the corresponding valve;

[0116] Indicates the boundary of water quality status;

[0117] Indicates the boundary of water quality status In the current inspection cycle relative to the valve The remaining conveying volume;

[0118] Indicates the boundary of water quality status Relative to valve and valve operating direction The result of determining whether to enter the prematurely triggered window;

[0119] like This indicates that the remaining transport volume corresponding to the water quality state boundary has entered the early trigger window of the corresponding valve; if This indicates that the current detection cycle does not meet the early triggering conditions for the corresponding valve.

[0120] When the pure water flow rate is obtained using a discrete sampling method, the pure water flow rate value acquired at the trigger judgment time is converted into the change in delivery volume within the corresponding sampling period, and the changes in delivery volume are accumulated to obtain the action delivery volume. Through the above processing, the valve action time is converted into an advance trigger window with the same dimension as the remaining delivery volume, enabling subsequent trigger judgments to be executed on a volume scale.

[0121] Furthermore, if the remaining transport volume of the water quality state boundary relative to the corresponding valve position is greater than zero, and the remaining transport volume is not greater than the transport volume of the corresponding valve in the corresponding action direction, then the water quality state boundary is determined to have entered the early trigger window of the corresponding valve. If the remaining transport volume of the water quality state boundary relative to the corresponding valve position is not greater than zero, then the water quality state boundary is determined to have reached or passed the corresponding valve position, and the corresponding detection cycle is no longer used as the early trigger judgment cycle for the corresponding valve position.

[0122] S52, when the pre-action control command is used to cause the unqualified water section to enter the discharge path, the discharge valve is determined as the valve to be opened; when the pre-action control command is used to cause the unqualified water section to enter the return path, the return valve is determined as the valve to be opened.

[0123] In practice, the control system determines the valve to be opened based on the non-conforming water segment handling path corresponding to the pre-action control command. When the pre-action control command is used to cause the non-conforming water segment to enter the discharge path, the discharge valve is identified as the valve to be opened; when the pre-action control command is used to cause the non-conforming water segment to enter the return path, the return valve is identified as the valve to be opened. After the valve to be opened is determined, the control system calls the valve action time corresponding to the valve to be opened and determines the advance trigger window corresponding to the valve to be opened based on the valve action time. The storage valve and the valve to be opened are jointly the valve control objects corresponding to the same non-conforming water segment. The storage valve is used to block the non-conforming water segment from entering the storage path, and the valve to be opened is used to cause the non-conforming water segment to enter the discharge path or the return path.

[0124] S53, when the remaining conveying volume of the water storage valve and the valve to be opened corresponding to the front boundary of the unqualified water section enters the advance trigger window of the corresponding valve, a pre-action control command for closing the water storage valve and opening the valve to be opened is generated.

[0125] In practice, the prediction results of the water storage valve corresponding to the upstream boundary of the substandard water section are read, and the remaining transport volume of the upstream boundary of the substandard water section relative to the water storage valve is obtained from it. Simultaneously, the prediction results of the valve to be opened corresponding to the upstream boundary of the substandard water section are read, and the remaining transport volume of the upstream boundary of the substandard water section relative to the valve to be opened is obtained from it. The remaining transport volume corresponding to the water storage valve is compared with the advance trigger window corresponding to the closing of the water storage valve, and the remaining transport volume corresponding to the valve to be opened is compared with the advance trigger window corresponding to the opening of the valve to be opened.

[0126] When the remaining transport volume of the non-compliant water segment relative to the front boundary of the storage valve enters the pre-trigger window corresponding to the closing of the storage valve, and the remaining transport volume of the non-compliant water segment relative to the valve to be opened enters the pre-trigger window corresponding to the opening of the valve to be opened, a pre-action control command is generated to close the storage valve and open the valve to be opened. This pre-action control command includes at least the water quality state boundary, the non-compliant water segment front boundary marker, the storage valve closing action, the valve to be opened opening action, the trigger judgment time, and the predicted position of the corresponding valve. The control system sends control signals to the storage valve and the valve to be opened according to the pre-action control command, causing the storage valve to enter the closing process before the non-compliant water segment front boundary reaches its position, and causing the valve to enter the opening process before the non-compliant water segment front boundary reaches its position.

[0127] Furthermore, if the remaining transport volume of the non-compliant water segment's leading boundary relative to the storage valve does not enter the advance trigger window corresponding to the storage valve closing, or if the remaining transport volume of the non-compliant water segment's leading boundary relative to the valve to be opened does not enter the advance trigger window corresponding to the valve to be opened, then no pre-action control command for closing the storage valve and opening the valve to be opened will be generated in the current detection cycle. Instead, advance trigger judgment will continue in subsequent detection cycles based on the predicted results of the corresponding water quality state boundary. Through this process, the control system generates corresponding pre-action control commands after the advance trigger conditions for closing the storage valve and opening the valve to be opened are met at the leading boundary of the non-compliant water segment, ensuring that the timing of the actions of the storage valve and the valve to be opened matches the propagation state of the non-compliant water segment's leading boundary reaching the corresponding valve position.

[0128] S54, when the remaining transport volume of the water storage valve and the valve to be opened corresponding to the downstream boundary of the unqualified water section enters the advance trigger window of the corresponding valve, a pre-action control command is generated to close the valve to be opened and restore the opening of the water storage valve, and the pre-action control command is associated with the prediction result of the corresponding water quality state boundary.

[0129] In practice, the prediction results of the water storage valve corresponding to the downstream boundary of the substandard water section are read, and the remaining transport volume of the downstream boundary of the substandard water section relative to the water storage valve is obtained from it. Simultaneously, the prediction results of the valve to be opened corresponding to the downstream boundary of the substandard water section are read, and the remaining transport volume of the downstream boundary of the substandard water section relative to the valve to be opened is obtained from it. The remaining transport volume corresponding to the water storage valve is compared with the advance trigger window corresponding to the reopening of the water storage valve, and the remaining transport volume corresponding to the valve to be opened is compared with the advance trigger window corresponding to the closing of the valve to be opened.

[0130] When the remaining transport volume of the non-compliant water segment relative to the downstream boundary of the storage valve enters the advance trigger window corresponding to the reopening of the storage valve, and the remaining transport volume of the non-compliant water segment relative to the valve to be opened enters the advance trigger window corresponding to the closing of the valve to be opened, a pre-action control command is generated to close the valve to be opened and reopen the storage valve. This pre-action control command includes at least the water quality state boundary, the downstream boundary marker of the non-compliant water segment, the closing action of the valve to be opened, the reopening action of the storage valve, the trigger judgment time, and the predicted result of the corresponding valve position. The control system sends control signals to the valve to be opened and the storage valve according to the pre-action control command, causing the discharge path or return path to gradually exit the treatment state after the downstream boundary of the non-compliant water segment is crossed, and allowing the compliant water segment to enter the storage path.

[0131] Furthermore, the generated pre-action control commands are associated and saved with the prediction results of the corresponding water quality state boundaries. For the front boundary of the unqualified water section, the pre-action control commands are associated with the prediction results of the corresponding water storage valve and the valve to be opened at the front boundary of the unqualified water section; for the rear boundary of the unqualified water section, the pre-action control commands are associated with the prediction results of the corresponding water storage valve and the valve to be opened at the rear boundary of the unqualified water section. The associated and saved content is used for subsequent feedback error correction based on the verified water quality monitoring values ​​of the water storage path, enabling the control system to trace the correspondence between the closing of the water storage valve, the reopening of the water storage valve, and the actual water quality state switching of the water storage path.

[0132] If the remaining transport volume of the substandard water segment relative to the storage valve does not enter the advance trigger window corresponding to the reopening of the storage valve, or if the remaining transport volume of the substandard water segment relative to the valve to be opened does not enter the advance trigger window corresponding to the closing of the valve to be opened, then no pre-action control command for closing the valve to be opened and reopening the storage valve will be generated in the current detection cycle. Instead, advance trigger judgment will continue in subsequent detection cycles based on the predicted results of the corresponding water quality state boundary. Through this process, the control system generates corresponding pre-action control commands after the advance trigger conditions for reopening the storage valve and closing the valve to be opened are met at the substandard water segment's downstream boundary, ensuring that the timing of the actions of the storage valve and the valve to be opened matches the propagation state of the substandard water segment's downstream boundary reaching the corresponding valve position.

[0133] In one embodiment, steps S61-S67 are described as follows:

[0134] S61, determine the actual water quality state switching process in the water storage path based on the verified water quality monitoring value, and extract the occurrence time of the actual front boundary and the occurrence time of the actual back boundary from the actual water quality state switching process.

[0135] In practice, the verification water quality monitoring values ​​of the water storage path are read in chronological order of the detection times. Based on the water quality qualification threshold, the verification water quality monitoring value of the water storage path at each detection time is determined to be either qualified or unqualified. The water quality status of the water storage path at adjacent detection times is compared. When the water quality status of the water storage path changes from qualified to unqualified, the subsequent detection time of the change is determined as the actual occurrence time of the front boundary; when the water quality status of the water storage path changes from unqualified to qualified, the subsequent detection time of the change is determined as the actual occurrence time of the back boundary.

[0136] Furthermore, the occurrence times of the actual front boundary and the actual back boundary are associated and saved with the verification water quality monitoring values ​​of the water storage path. The actual front boundary indicates the actual time position when the unqualified water section enters the water storage path, and the actual back boundary indicates the actual time position when the water quality in the water storage path recovers to a qualified state. These time positions are then compared with the arrival times in the prediction results corresponding to the water storage valves.

[0137] In this embodiment, the verified water quality monitoring values ​​of the water storage path are used to reflect the water quality state switching results after the water body enters the water storage path. The occurrence time of the actual front boundary is used to indicate the actual time position in the water storage path where the water quality changes from a qualified state to a non-qualified state, and the occurrence time of the actual rear boundary is used to indicate the actual time position in the water storage path where the water quality changes from a non-qualified state to a qualified state. The occurrence time of the actual front boundary is compared with the arrival time in the predicted result of the front boundary of the non-qualified water section corresponding to the water storage valve, and the occurrence time of the actual rear boundary is compared with the arrival time in the predicted result of the rear boundary of the non-qualified water section corresponding to the water storage valve.

[0138] In this embodiment, the effective volume of the target path corresponding to the water storage valve is used to represent the effective transport volume between the online water quality monitoring point and the valve position corresponding to the water storage valve. The arrival time in the prediction result corresponding to the water storage valve is used to represent the time and position when the water quality state boundary reaches the valve position corresponding to the water storage valve. The verification water quality monitoring value of the water storage path is used to correct the feedback error of the prediction result corresponding to the water storage valve without changing the meaning of the effective volume of the target path corresponding to the water storage valve and the arrival time in the prediction result corresponding to the water storage valve.

[0139] When the water quality monitoring value of the water storage path does not change from qualified to unqualified during the verification process associated with the front boundary of the corresponding unqualified water section, the actual occurrence time of the front boundary is not extracted, the front boundary prediction deviation is not generated, and the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the closing of the water storage valve are not updated in this feedback error correction.

[0140] When the water quality monitoring value of the water storage path does not change from unqualified to qualified during the verification process associated with the downstream boundary of the corresponding unqualified water section, the actual occurrence time of the downstream boundary is not extracted, the downstream boundary prediction deviation is not generated, and the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the resumption of opening of the water storage valve are not updated in this feedback error correction.

[0141] When the verified water quality monitoring value of the water storage path consistently meets the qualification conditions corresponding to the water quality qualification threshold during the verification process associated with the corresponding water quality state boundary, it indicates that no actual front boundary and actual back boundary have been formed in the water storage path that can be compared with the prediction results corresponding to the water storage valve. In this case, the occurrence time of the actual front boundary and the occurrence time of the actual back boundary are not extracted, the front boundary prediction deviation and the back boundary prediction deviation are not generated, the corresponding water quality state boundary prediction deviation is not generated, and the effective volume of the target path corresponding to the water storage valve and the valve action time are not updated in this feedback error correction.

[0142] The reduction in the rate of continuous discharge or backflow of qualified water sections is achieved through the predicted results corresponding to the downstream boundary of unqualified water sections, the advance trigger window corresponding to the reopening of the water storage valve, and the pre-action control commands used to close the valve to be opened and reopen the water storage valve. When no actual upstream and downstream boundaries are formed in the water storage path, the control system does not generate parameter corrections based on the verified water quality monitoring values ​​that lack actual comparison objects.

[0143] S62, compare the occurrence time of the actual front boundary with the arrival time in the prediction result of the front boundary of the unqualified water section corresponding to the water storage valve to obtain the front boundary prediction deviation.

[0144] In practice, the prediction result of the water storage valve corresponding to the front boundary of the unqualified water section is read, and the arrival time of the front boundary of the unqualified water section to the valve position corresponding to the water storage valve is obtained from the prediction result. When the occurrence time of the actual front boundary has been extracted in step S61, the occurrence time of the actual front boundary is compared with the arrival time in the prediction result of the front boundary of the unqualified water section corresponding to the water storage valve to obtain the front boundary prediction deviation.

[0145] If the current boundary prediction deviation indicates that the actual occurrence time of the preceding boundary is earlier than the predicted arrival time of the preceding boundary of the non-compliant water section corresponding to the water storage valve, it means that the preceding boundary of the non-compliant water section appears earlier in the water storage path than predicted, and the corresponding water storage valve closing trigger timing needs to be corrected in advance. If the current boundary prediction deviation indicates that the actual occurrence time of the preceding boundary is later than the predicted arrival time of the preceding boundary of the non-compliant water section corresponding to the water storage valve, it means that the water storage valve closing trigger timing needs to be corrected for being too early.

[0146] When the actual occurrence time of the front boundary is not extracted in step S61, the comparison between the occurrence time of the actual front boundary and the arrival time in the prediction result of the front boundary of the unqualified water section corresponding to the water storage valve is not performed, and no front boundary prediction deviation is generated.

[0147] S63, compare the occurrence time of the actual back boundary with the arrival time in the prediction result of the back boundary of the unqualified water section corresponding to the water storage valve to obtain the back boundary prediction deviation, and generate the water quality state boundary prediction deviation based on the front boundary prediction deviation and the back boundary prediction deviation.

[0148] In practice, the prediction result of the water storage valve corresponding to the downstream boundary of the unqualified water section is read, and the arrival time of the downstream boundary of the unqualified water section to the valve position corresponding to the water storage valve is obtained from the prediction result. When the occurrence time of the actual downstream boundary has been extracted in step S61, the occurrence time of the actual downstream boundary is compared with the arrival time in the prediction result of the downstream boundary of the unqualified water section corresponding to the water storage valve to obtain the downstream boundary prediction deviation.

[0149] Furthermore, the generated front boundary prediction bias and back boundary prediction bias are written into the water quality state boundary prediction bias. The water quality state boundary prediction bias includes at least the water quality state boundary corresponding to the bias, the direction of the water storage valve's action corresponding to the bias, the arrival time in the prediction result corresponding to the water storage valve, the actual occurrence time in the water storage path, and the bias direction. The front boundary prediction bias corresponds to the water storage valve closing control correction, and the back boundary prediction bias corresponds to the water storage valve reopening control correction.

[0150] When the actual occurrence time of the rear boundary is not extracted in step S61, the comparison between the actual occurrence time of the rear boundary and the arrival time in the prediction result of the rear boundary of the unqualified water section corresponding to the water storage valve is not performed, and no rear boundary prediction deviation is generated. When only the front boundary prediction deviation is generated, the water quality state boundary prediction deviation only includes the front boundary prediction deviation and is only used for parameter correction corresponding to the closing of the water storage valve; when only the rear boundary prediction deviation is generated, the water quality state boundary prediction deviation only includes the rear boundary prediction deviation and is only used for parameter correction corresponding to the reopening of the water storage valve.

[0151] S64, determine the parameter correction amount corresponding to closing the water storage valve based on the prediction deviation of the front boundary, and determine the parameter correction amount corresponding to reopening the water storage valve based on the prediction deviation of the rear boundary.

[0152] In this embodiment, both the front boundary prediction bias and the back boundary prediction bias are used to determine the parameter correction amounts for the effective volume of the target path and the valve action time corresponding to the water storage valve. When both the front boundary prediction bias and the back boundary prediction bias are generated in the same feedback error correction process, the effective volume of the target path corresponding to the water storage valve is updated sequentially according to steps S65 and S66, and the effective volume of the target path corresponding to the water storage valve obtained after sequential updates is used as the input parameter for the propagation prediction corresponding to the water storage valve in subsequent detection cycles.

[0153] The specifics are described in steps S641-S642.

[0154] In one embodiment, steps S641-S642 are described as follows:

[0155] S641, based on the pure water flow rate value within the time interval between the arrival time of the predicted front boundary of the unqualified water section corresponding to the water storage valve and the actual occurrence time of the front boundary, determine the effective volume correction amount of the target path corresponding to the closing of the water storage valve, and determine the time interval as the valve action time correction amount corresponding to the closing of the water storage valve.

[0156] In practice, the arrival time of the predicted boundary of the unqualified water section corresponding to the water storage valve, the occurrence time of the actual boundary, and the pure water flow rate between the two times are read. The change in transport volume within the time interval is determined based on the pure water flow rate between the two times, and this change in transport volume is used as the effective volume correction amount for the target path corresponding to the closing of the water storage valve. The time interval between the two times is used as the valve action time correction amount for the closing of the water storage valve, and the correction direction is determined based on the relationship between the occurrence time of the actual boundary and the arrival time in the predicted result.

[0157] S642: Based on the pure water flow rate value within the time interval between the arrival time of the predicted result of the unqualified water section downstream boundary corresponding to the water storage valve and the actual occurrence time of the downstream boundary, determine the effective volume correction amount of the target path corresponding to the resumption of opening of the water storage valve, and determine the time interval as the valve action time correction amount corresponding to the resumption of opening of the water storage valve.

[0158] In practice, the arrival time of the predicted boundary of the unqualified water section corresponding to the water storage valve, the occurrence time of the actual boundary, and the pure water flow rate between the two times are read. The change in transport volume within the time interval is determined based on the pure water flow rate between the two times, and this change in transport volume is used as the effective volume correction amount for the target path corresponding to the reopening of the water storage valve. The time interval between the two times is used as the valve action time correction amount for the reopening of the water storage valve, and the correction direction is determined based on the relationship between the occurrence time of the actual boundary and the arrival time in the predicted result.

[0159] In one specific implementation, the water quality state boundary prediction deviation, the target path effective volume correction, and the valve action time correction can be determined according to the following formula:

[0160]

[0161]

[0162]

[0163] in, This indicates the bias in the front boundary prediction; Indicates the moment when the actual front boundary occurs; This indicates the arrival time in the predicted result of the non-conforming water section front boundary corresponding to the water storage valve;

[0164] This indicates the effective volume correction amount for the target path corresponding to the closure of the water storage valve; This indicates the valve action time correction amount corresponding to the closing of the water storage valve;

[0165] Indicates the prediction bias at the back boundary; Indicates the actual time when the back boundary occurs; This indicates the arrival time in the predicted result of the downstream boundary of the unqualified water section corresponding to the water storage valve;

[0166] This indicates the effective volume correction amount for the target path corresponding to the reopening of the water storage valve; This indicates the valve action time correction amount corresponding to the water storage valve resuming its opening.

[0167] This represents the pure water flow rate value within the corresponding time interval;

[0168] The sign function is used to preserve the direction of the deviation between the actual occurrence time and the arrival time in the prediction result.

[0169] When the pure water flow rate is obtained using a discrete sampling method, the changes in the transport volume corresponding to each sampling period within the time interval are accumulated to obtain the corresponding target path effective volume correction amount.

[0170] Using the above formula, the front boundary prediction bias and the back boundary prediction bias are converted into the target path effective volume correction at the volume scale and the valve action time correction at the time scale. The target path effective volume correction is used in steps S65 and S66 to update the target path effective volume corresponding to the water storage valve, and the valve action time correction is used in steps S65 and S66 to update the valve action time corresponding to the water storage valve closing and the valve action time corresponding to the water storage valve reopening.

[0171] Furthermore, the effective volume correction amount for the target path corresponding to the closing of the water storage valve is used to correct the propagation prediction of the non-conforming water segment's front boundary reaching the corresponding valve position in subsequent detection cycles. The valve action time correction amount for the closing of the water storage valve is used to correct the premature trigger window corresponding to the closing of the water storage valve in subsequent detection cycles. The effective volume correction amount for the target path corresponding to the reopening of the water storage valve is used to correct the propagation prediction of the non-conforming water segment's rear boundary reaching the corresponding valve position in subsequent detection cycles. The valve action time correction amount for the reopening of the water storage valve is used to correct the premature trigger window corresponding to the reopening of the water storage valve in subsequent detection cycles.

[0172] In subsequent detection cycles, the control system generates a prediction result of the water quality state boundary reaching the corresponding valve position of the water storage valve based on the updated target path effective volume, and generates an advance trigger window for the water storage valve based on the updated valve action time. The prediction deviation of the same water quality state boundary is no longer repeatedly added to the pre-action control command as an additional advance or additional delay.

[0173] S65, based on the parameter correction amount corresponding to the closing of the water storage valve, update the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the closing of the water storage valve.

[0174] In practice, when a leading boundary prediction deviation has been generated, the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the water storage valve closure are updated based on the parameter correction amount corresponding to the water storage valve closure. The correction amount for the effective volume of the target path corresponding to the water storage valve closure is a non-negative volume quantity, and the correction amount for the valve action time corresponding to the water storage valve closure is a non-negative time quantity. The correction direction is determined based on the relationship between the actual occurrence time of the leading boundary and the arrival time in the prediction result of the leading boundary of the non-conforming water section corresponding to the water storage valve.

[0175] The current boundary prediction deviation indicates that when the actual occurrence time of the front boundary is earlier than the arrival time in the prediction result of the front boundary of the unqualified water section corresponding to the water storage valve, the updated effective volume of the target path corresponding to the water storage valve is equal to the original effective volume of the target path corresponding to the water storage valve minus the correction amount of the effective volume of the target path corresponding to the closing of the water storage valve; the updated valve action time corresponding to the closing of the water storage valve is equal to the original valve action time corresponding to the closing of the water storage valve plus the correction amount of the valve action time corresponding to the closing of the water storage valve.

[0176] The current boundary prediction deviation indicates that when the actual occurrence time of the front boundary is later than the arrival time in the prediction result of the front boundary of the unqualified water section corresponding to the water storage valve, the updated effective volume of the target path corresponding to the water storage valve is equal to the original effective volume of the target path corresponding to the water storage valve plus the correction amount of the effective volume of the target path corresponding to the closing of the water storage valve; the updated valve action time corresponding to the closing of the water storage valve is equal to the original valve action time corresponding to the closing of the water storage valve minus the correction amount of the valve action time corresponding to the closing of the water storage valve.

[0177] When no front boundary prediction deviation is generated, the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the closing of the water storage valve are not updated, and the parameters that were used in the previous detection cycle for the front boundary propagation prediction of the non-conforming water section and the early triggering control of the closing of the water storage valve are retained.

[0178] S66, based on the parameter correction amount corresponding to the water storage valve resuming opening, update the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the water storage valve resuming opening.

[0179] In practice, when a back boundary prediction deviation has been generated, the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the water storage valve's reopening are updated based on the parameter correction amount corresponding to the water storage valve's reopening. The correction amount for the effective volume of the target path corresponding to the water storage valve's reopening is a non-negative volume quantity, and the correction amount for the valve action time corresponding to the water storage valve's reopening is a non-negative time quantity. The correction direction is determined based on the relationship between the actual occurrence time of the back boundary and the arrival time in the prediction result of the back boundary of the non-conforming water section corresponding to the water storage valve.

[0180] When the prediction deviation of the back boundary indicates that the actual occurrence time of the back boundary is earlier than the arrival time in the prediction result of the back boundary of the unqualified water section corresponding to the water storage valve, the updated effective volume of the target path corresponding to the water storage valve is equal to the effective volume of the target path corresponding to the water storage valve before the update minus the correction amount of the effective volume of the target path corresponding to the resumption of the water storage valve; the updated valve action time corresponding to the resumption of the water storage valve is equal to the valve action time corresponding to the resumption of the water storage valve before the update plus the correction amount of the valve action time corresponding to the resumption of the water storage valve.

[0181] When the back boundary prediction deviation indicates that the actual occurrence time of the back boundary is later than the arrival time in the prediction result of the back boundary of the unqualified water section corresponding to the water storage valve, the updated effective volume of the target path corresponding to the water storage valve is equal to the original effective volume of the target path corresponding to the water storage valve plus the correction amount of the effective volume of the target path corresponding to the resumption of the water storage valve; the updated valve action time corresponding to the resumption of the water storage valve is equal to the original valve action time corresponding to the resumption of the water storage valve minus the correction amount of the valve action time corresponding to the resumption of the water storage valve.

[0182] When no back boundary prediction deviation is generated, the update of the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the resumption of the water storage valve is not performed, and the parameters that were used in the previous detection cycle for the back boundary propagation prediction of the unqualified water section and the early trigger control for the resumption of the water storage valve are retained.

[0183] Furthermore, after updating the effective volume of the target path corresponding to the water storage valve based on the target path effective volume correction, if the updated target path effective volume is less than zero, then the updated target path effective volume is set to zero. Similarly, after updating the valve action time corresponding to the water storage valve closing or reopening based on the valve action time correction, if the updated valve action time is less than zero, then the updated valve action time is set to zero. Through these processes, the effective volume of the target path used for propagation prediction and the valve action time used for early trigger control remain non-negative parameters in subsequent detection cycles.

[0184] S67 uses the updated parameters for propagation prediction and early trigger control in subsequent detection cycles.

[0185] In specific implementation, the effective volume of the target path corresponding to the water storage valve, obtained after updates in steps S65 and S66, is used as the input parameter for the propagation prediction of the water storage valve in subsequent detection cycles. The valve action time corresponding to the closing of the water storage valve is used as the input parameter for the early trigger control of the closing of the water storage valve in subsequent detection cycles, and the valve action time corresponding to the reopening of the water storage valve is used as the input parameter for the early trigger control of the reopening of the water storage valve in subsequent detection cycles. If only the front boundary prediction deviation is generated, the parameters used in subsequent detection cycles are determined according to the update result of step S65; if only the rear boundary prediction deviation is generated, the parameters used in subsequent detection cycles are determined according to the update result of step S66; if neither the front boundary prediction deviation nor the rear boundary prediction deviation is generated, the corresponding parameters already saved in the previous detection cycle are retained.

[0186] For feedback error correction processes where no preceding boundary prediction deviation or no following boundary prediction deviation is generated, the control system does not update the corresponding parameters lacking actual boundaries and continues to use the corresponding parameters already saved in the previous detection cycle. Through steps S61-S67, the actual water quality state switching process formed by the verification water quality monitoring values ​​of the water storage path can be fed back to the prediction results corresponding to the water storage valve; when there is an actual water quality state switching, the preceding boundary prediction deviation and the following boundary prediction deviation are converted into corresponding parameter correction quantities, and further applied to the propagation prediction and early trigger control of subsequent detection cycles; when there is no actual water quality state switching, no parameter correction quantity without an actual comparison object is generated.

[0187] In this embodiment, feedback error correction is performed based on the verification water quality monitoring values ​​of the water storage path. These values ​​reflect the actual water quality state switching process after the water enters the storage path. Therefore, steps S61-S67 are used to update the effective volume of the target path and the valve actuation time corresponding to the water storage valve. The effective volume of the target path corresponding to the discharge valve and the return valve, as well as the valve actuation time, are used as input parameters obtained in step S1 and used for the advance trigger control in step S5, and are not updated based on the verification water quality monitoring values ​​of the water storage path. Through the above processing, the parameter update object of feedback error correction is consistent with the monitoring object of the verification water quality monitoring values ​​of the water storage path.

[0188] Reference Figure 3 , Figure 3 This is a schematic diagram of the adaptive control system of the pure water preparation equipment provided by the present invention. The system includes:

[0189] The parameter acquisition module 310 acquires online water quality monitoring values, pure water flow values, valve action time, verification water quality monitoring values ​​of the water storage path, and the effective volume of the target path of the water storage valve, discharge valve, and return valve.

[0190] The water quality boundary queue generation module 320 updates the cumulative flow value according to the pure water flow value, performs state transition detection based on the online water quality monitoring value to generate the front boundary and the back boundary of the unqualified water section, determines the boundary generation time and its corresponding cumulative flow value, and generates a water quality state boundary queue according to the corresponding valve positions of the water storage valve, discharge valve and return valve.

[0191] The cumulative transport volume determination module 330 determines the cumulative transport volume by integrating the pure water flow rate value based on the cumulative flow rate value corresponding to the boundary generation time.

[0192] The boundary propagation prediction module 340 performs propagation prediction based on the cumulative transport volume and the effective volume of the target path, and generates a prediction result of the water quality state boundary reaching the corresponding valve position.

[0193] The valve pre-action control module 350 performs advance trigger control based on the prediction results and valve action time, generates pre-action control commands, causes the unqualified water section to enter the discharge path or return path, and causes the qualified water sections before and after the unqualified water section to enter the water storage path.

[0194] The feedback error correction module 360 ​​corrects the prediction results corresponding to the water storage valve based on the verified water quality monitoring values, determines the water quality state boundary prediction deviation, and updates the effective volume of the target path and the valve action time corresponding to the water storage valve based on the water quality state boundary prediction deviation. The updated parameters are then used for propagation prediction and early trigger control in subsequent detection cycles.

[0195] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0196] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0197] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention 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 will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of adaptive control of a pure water production plant, characterized in that, Includes the following steps: S1: Obtain online water quality monitoring values, pure water flow values, valve action time, verification water quality monitoring values ​​of the water storage path, and the effective volume of the target path of the water storage valve, discharge valve, and return valve; S2: Update the cumulative flow value according to the pure water flow value, perform state transition detection based on the online water quality monitoring value to generate the front boundary and the back boundary of the unqualified water section, determine the boundary generation time and its corresponding cumulative flow value, and generate a water quality state boundary queue according to the corresponding valve positions of the water storage valve, discharge valve and return valve. S3: Based on the cumulative flow rate value corresponding to the boundary generation time, perform flow rate integration on the pure water flow rate value to determine the cumulative transport volume; S4: Based on the cumulative transport volume and the effective volume of the target path, propagation prediction is performed to generate a prediction result of the water quality state boundary reaching the corresponding valve position; S5: Based on the prediction results and valve action time, perform advance trigger control to generate pre-action control commands, so that the unqualified water section enters the discharge path or return path, and the qualified water sections before and after the unqualified water section enter the water storage path. S6: Based on the verified water quality monitoring values, perform feedback error correction on the prediction results corresponding to the water storage valve, determine the water quality state boundary prediction deviation, and update the effective volume of the target path and valve action time corresponding to the water storage valve based on the water quality state boundary prediction deviation. Use the updated parameters for propagation prediction and early trigger control in subsequent detection cycles.

2. The self-adaptive control method of a pure water preparation apparatus according to claim 1, characterized by, Specifically, S2 is: S21: Based on the water quality qualification threshold, the online water quality monitoring values ​​at each detection time are determined to be qualified or unqualified. S22: Compare the water quality status at adjacent detection times. When the water quality status changes from qualified to unqualified, generate the front boundary of the unqualified water segment; when the water quality status changes from unqualified to qualified, generate the rear boundary of the unqualified water segment, and determine the next detection time after the status change as the corresponding boundary generation time. S23: Take the front boundary and the rear boundary of the unqualified water section as the water quality state boundary, generate water quality state boundary records according to the water storage valve, discharge valve and return valve respectively, and write the water quality state boundary, the cumulative flow value corresponding to the boundary generation time, the corresponding valve position and the effective volume of the target path into each water quality state boundary record, and generate a water quality state boundary queue.

3. The self-adaptive control method of a pure water preparation apparatus according to claim 2, characterized by, Specifically, S3 is: S31: In each detection cycle, update the current cumulative flow value according to the pure water flow value, and read the current cumulative flow value and the cumulative flow value corresponding to the boundary generation time; S32: The difference between the current cumulative flow value and the cumulative flow value corresponding to the boundary generation time of the water quality state boundary is determined as the cumulative transport volume of the corresponding water quality state boundary in the current detection cycle, and the cumulative transport volume is written into the water quality state boundary queue.

4. The self-adaptive control method of a pure water preparation apparatus according to claim 3, characterized by, Specifically, S4 is: S41: Based on the corresponding valve position recorded in the water quality state boundary queue, call the effective volume of the target path at the corresponding valve position, and determine the difference between the effective volume of the target path and the cumulative transport volume as the remaining transport volume of the current detection cycle, and write the remaining transport volume of the current detection cycle into the water quality state boundary queue. S42: Read the remaining transport volume written in the previous detection cycle in the same water quality state boundary record. When the remaining transport volume in the previous detection cycle is greater than zero and the remaining transport volume in the current detection cycle is not greater than zero, generate the arrival judgment result of the water quality state boundary crossing the corresponding valve position. S43: When the arrival determination result indicates that the water quality state boundary has crossed the corresponding valve position, the arrival time of the water quality state boundary reaching the corresponding valve position is determined based on the remaining transport volume of the previous detection cycle, the remaining transport volume of the current detection cycle, and the pure water flow rate value from the previous detection cycle to the current detection cycle. S44: Use the remaining conveying volume, arrival determination result, arrival time and corresponding valve position as the prediction result.

5. The self-adaptive control method of a pure water preparation apparatus according to claim 4, characterized by, Specifically, S43 is: S431: Starting from the detection time of the previous detection cycle, the pure water flow rate value from the previous detection cycle to the current detection cycle is accumulated and integrated to obtain the change in transport volume over time. S432: The time when the change in conveying volume within the cycle first reaches the remaining conveying volume of the previous detection cycle is determined as the arrival offset time from the previous detection cycle to the current detection cycle. S433: Based on the detection time of the previous detection cycle and the arrival offset time, determine the arrival time of the water quality state boundary to the corresponding valve position.

6. The self-adaptive control method of a pure water preparation apparatus according to claim 4, wherein, S5 generates a pre-action control command by performing advance triggering control based on the prediction result and the valve action time, specifically as follows: S51: Determine the corresponding valve's action delivery volume based on the pure water flow rate during the valve's action time, and use the action delivery volume as the corresponding valve's advance trigger window; S52: When the pre-action control command is used to cause the unqualified water section to enter the discharge path, the discharge valve is determined as the valve to be opened; when the pre-action control command is used to cause the unqualified water section to enter the return path, the return valve is determined as the valve to be opened. S53: When the remaining conveying volume of the water storage valve and the valve to be opened corresponding to the front boundary of the unqualified water section enters the advance trigger window of the corresponding valve, a pre-action control command is generated to close the water storage valve and open the valve to be opened. S54: When the remaining transport volume of the water storage valve and the valve to be opened corresponding to the downstream boundary of the unqualified water section enters the advance trigger window of the corresponding valve, a pre-action control command is generated to close the valve to be opened and restore the opening of the water storage valve, and the pre-action control command is associated with the prediction result of the corresponding water quality state boundary.

7. The self-adaptive control method of a pure water preparation apparatus according to claim 5, wherein, The step S6 involves correcting the prediction results corresponding to the water storage valve based on the verified water quality monitoring values ​​to determine the water quality state boundary prediction deviation. Specifically: S61: Determine the actual water quality state switching process in the water storage path based on the verified water quality monitoring values, and extract the occurrence time of the actual front boundary and the occurrence time of the actual back boundary from the actual water quality state switching process; S62: Compare the actual occurrence time of the front boundary with the arrival time in the prediction result of the front boundary of the unqualified water section corresponding to the water storage valve to obtain the front boundary prediction deviation. S63: Compare the occurrence time of the actual back boundary with the arrival time in the prediction result of the back boundary of the unqualified water section corresponding to the water storage valve to obtain the back boundary prediction deviation, and generate the water quality state boundary prediction deviation based on the front boundary prediction deviation and the back boundary prediction deviation.

8. The adaptive control method for the pure water preparation equipment according to claim 7, characterized in that, The S6 update of the effective volume of the target path and the valve action time corresponding to the water storage valve based on the water quality state boundary prediction deviation is as follows: S64: Determine the parameter correction amount corresponding to closing the water storage valve based on the front boundary prediction deviation, and determine the parameter correction amount corresponding to reopening the water storage valve based on the rear boundary prediction deviation. S65: Based on the parameter correction amount corresponding to the closing of the water storage valve, update the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the closing of the water storage valve; S66: Based on the parameter correction amount corresponding to the water storage valve resuming opening, update the effective volume of the target path corresponding to the water storage valve and the valve action time corresponding to the water storage valve resuming opening.

9. The adaptive control method for the pure water preparation equipment according to claim 8, characterized in that, Specifically, S64 is: S641: Based on the pure water flow rate value within the time interval between the arrival time of the predicted front boundary of the unqualified water section corresponding to the water storage valve and the actual occurrence time of the front boundary, determine the effective volume correction amount of the target path corresponding to the closing of the water storage valve, and determine the time interval as the valve action time correction amount corresponding to the closing of the water storage valve. S642: Based on the pure water flow rate value within the time interval between the arrival time of the predicted result of the unqualified water section downstream boundary corresponding to the water storage valve and the actual occurrence time of the downstream boundary, determine the effective volume correction amount of the target path corresponding to the resumption of opening of the water storage valve, and determine the time interval as the valve action time correction amount corresponding to the resumption of opening of the water storage valve.

10. An adaptive control system for a pure water preparation device, characterized in that, An adaptive control method for implementing the pure water preparation equipment according to any one of claims 1 to 9, the system comprising: The parameter acquisition module acquires online water quality monitoring values, pure water flow values, valve action time, verification water quality monitoring values ​​of the water storage path, and the effective volume of the target path of the water storage valve, discharge valve, and return valve. The water quality boundary queue generation module updates the cumulative flow value based on the pure water flow value, performs state transition detection based on the online water quality monitoring value to generate the front boundary and the back boundary of the unqualified water segment, determines the boundary generation time and its corresponding cumulative flow value, and generates a water quality state boundary queue according to the corresponding valve positions of the water storage valve, discharge valve and return valve. The cumulative delivery volume determination module uses the cumulative flow rate value corresponding to the boundary generation time as a benchmark to perform flow rate integration on the pure water flow rate value to determine the cumulative delivery volume. The boundary propagation prediction module performs propagation prediction based on the cumulative transport volume and the effective volume of the target path, and generates a prediction result of the water quality state boundary reaching the corresponding valve position. The valve pre-action control module performs advance trigger control based on the prediction results and valve action time, generates pre-action control commands, causes the unqualified water section to enter the discharge path or return path, and causes the qualified water sections before and after the unqualified water section to enter the water storage path. The feedback error correction module corrects the prediction results corresponding to the water storage valve based on the verified water quality monitoring values, determines the water quality state boundary prediction deviation, and updates the effective volume of the target path and the valve action time corresponding to the water storage valve based on the water quality state boundary prediction deviation. The updated parameters are then used for propagation prediction and early trigger control in subsequent detection cycles.

Images