A decision method and system for optimal dosage of a medicament

Abstract

The application discloses a decision method and system for optimal medicament dosage, and the method comprises the following steps: detecting the turbidity of raw water to be treated and the turbidity change step, and selecting a running mode according to the detection result; under the selected running mode, the supernatant turbidity value corresponding to different medicament dosing concentrations is obtained through corresponding coagulation stirring experiments or corresponding historical experimental data; and the optimal medicament dosing amount is decided according to the supernatant turbidity value and the corresponding medicament dosing concentration. By using the scheme of the application, an intelligent decision framework is established, and the automation, precision and high efficiency of medicament dosing in the water treatment process are realized. The application can realize real-time sensing of the dynamic of raw water quality, and automatically switch to the most suitable running mode according to the preset logic. By distinguishing different running modes, the application can adapt to the actual water condition, avoid the blindness of the traditional dosing mode, and significantly improve the scientificity and pertinence of the decision.

Description

TECHNICAL FIELD

[0001] The present application generally relates to the technical field of water treatment. More particularly, the present application relates to a decision-making method and system for optimal dosage of reagent. BACKGROUND

[0002] Coagulation and sedimentation is the core process of water purification in waterworks. In order to determine the optimal dosage of coagulant (such as polyaluminum chloride PAC), waterworks generally use six-beaker stirring experiments for simulation.

[0003] Currently, the experiment is mainly carried out in two ways: first, traditional manual stirring experiment: this method completely relies on the experience of experimenters. From water taking, reagent preparation, setting stirring intensity and time, to the final completion of sedimentation, the optimal dosage is determined by observing the morphology of flocculation body with naked eyes. This method has many disadvantages: first, the setting of experimental parameters (such as GT value of stirring) is highly subjective, it is difficult to accurately reproduce the hydraulic conditions of actual production in waterworks, and it is easy to cause decision deviation; second, the whole process is time-consuming and inefficient; third, the result is determined by manual visual observation, which is easy to cause wrong dosage decision due to subjective error. Insufficient dosage will lead to insufficient sedimentation, while excessive dosage may cause the disintegration of flocculation body, both of which will cause serious water quality problems. Second, semi-automatic six-beaker coagulation stirring instrument: this is a prior art device, which can automatically execute gradient stirring process according to the preset program through a multi-axis stirring system driven synchronously. However, this device still relies heavily on manual operation. The water and reagent adding steps before the experiment need to be completed manually, and more importantly, after the stirring and sedimentation program is completed, the experimental results in the six beakers still need to be determined by manual visual observation, so as to select the optimal dosage. This makes the determination of experimental results still subjective and uncertain, and data recording and unmanned operation cannot be realized.

[0004] In summary, the existing coagulation stirring experiment method, whether it is manual operation or the use of semi-automatic stirring instrument, faces the key challenges of high dependence on manual experience, long experimental process, strong subjectivity of result determination, and inability to automatically record and utilize data, which is difficult to meet the needs of fine and automated operation of modern waterworks.

[0005] Therefore, it is urgent to provide a decision-making scheme for optimal dosage of reagent, which can automatically complete the experiment, objectively analyze the results and directly guide the production and reagent dosage, in order to solve the pain points of the prior art. SUMMARY

[0006] In order to at least solve one or more of the above-mentioned technical problems, the present application proposes a decision-making scheme for optimal dosage of reagent in multiple aspects.

[0007] In a first aspect, the application provides a method for determining an optimal coagulant dosage, comprising: detecting the turbidity of raw water to be treated and a turbidity change step, and selecting an operation mode according to the detection results; under the selected operation mode, obtaining supernatant turbidity values corresponding to different coagulant dosages through corresponding coagulation and stirring experiments or corresponding historical experimental data; and determining the optimal coagulant dosage according to the supernatant turbidity values and the corresponding coagulant dosage concentrations.

[0008] In some embodiments, the operation mode includes: a daily mode, an emergency mode, and a high turbidity mode; wherein the daily mode is selected when the turbidity of raw water to be treated is less than or equal to a turbidity threshold value and the turbidity change step of raw water to be treated at two adjacent time points is less than or equal to a turbidity change threshold value; the emergency mode is selected when the turbidity of raw water to be treated is less than or equal to the turbidity threshold value and the turbidity change step of raw water to be treated at two adjacent time points is greater than the turbidity change threshold value; and the high turbidity mode is selected when the turbidity of raw water to be treated is greater than the turbidity threshold value.

[0009] In some embodiments, under the daily mode, nine reactors are evenly divided into three groups of reactors and are started at intervals of a preset time, coagulant is added in each group of reactors according to a first coagulant concentration gradient to perform a triple coagulation and stirring experiment, and a plurality of supernatant turbidity values corresponding to each group of reactors are obtained, wherein the coagulant dosage concentration intervals of each group of reactors are different.

[0010] In some embodiments, under the daily mode, in the process of determining the optimal coagulant dosage according to the supernatant turbidity values and the corresponding coagulant dosage concentrations, the following steps are performed: forming a concentration-turbidity curve corresponding to each group of reactors based on the obtained supernatant turbidity values and the corresponding coagulant dosage concentrations; determining whether all concentration-turbidity curves are monotonous numerical curves; in response to not all concentration-turbidity curves being monotonous numerical curves, determining the optimal coagulant dosage by identifying the lowest point of the concentration-turbidity curve that is not a monotonous numerical curve; in response to all concentration-turbidity curves being monotonous numerical curves, obtaining a plurality of supernatant turbidity values and a corresponding concentration-turbidity curve by performing a nine-coagulation and stirring experiment, determining whether the concentration-turbidity curve is a monotonous numerical curve; in response to the concentration-turbidity curve not being a monotonous numerical curve, determining the optimal coagulant dosage by identifying the lowest point of the concentration-turbidity curve; and in response to the concentration-turbidity curve being a monotonous numerical curve, adjusting the coagulant dosage concentration interval of the nine-coagulation and stirring experiment and repeating the nine-coagulation and stirring experiment until the corresponding concentration-turbidity curve obtained is not a monotonous numerical curve.

[0011] In some embodiments, in the process of obtaining the plurality of supernatant turbidity values by performing the nine-coagulation stirring experiment, the first medicament concentration gradient is reduced to a second medicament concentration gradient, and the medicament dosage concentration corresponding to the second-lowest supernatant turbidity value obtained in the process of adding the medicament in each group of reactors to perform the three-coagulation stirring experiment is taken as the current initial medicament dosage concentration, the medicament is added in the nine reactors according to the second medicament concentration gradient, and a plurality of supernatant turbidity values are obtained.

[0012] In some embodiments, in the emergency mode, the medicament is added in the three reactors according to a third medicament concentration gradient to perform the three-coagulation stirring experiment, and supernatant turbidity values at different medicament concentrations are obtained, wherein the third medicament concentration gradient is greater than the first medicament concentration gradient.

[0013] In some embodiments, in the process of determining the optimal medicament dosage based on the supernatant turbidity values and the corresponding medicament dosage concentrations in the emergency mode, the following steps are performed: forming a current concentration-turbidity curve based on the obtained supernatant turbidity values and the corresponding medicament dosage concentrations; determining whether the current concentration-turbidity curve is a monotonic numerical curve; in response to the current concentration-turbidity curve not being a monotonic numerical curve, obtaining a preliminary optimal medicament dosage by identifying the lowest point of the current concentration-turbidity curve, and determining the optimal medicament dosage by performing the six-coagulation stirring experiment; and in response to the current concentration-turbidity curve being a monotonic numerical curve, taking the lowest point of the current concentration-turbidity curve as the initial medicament dosage concentration, and repeating the process of adding the medicament in the three reactors according to the third medicament concentration gradient to perform the three-coagulation stirring experiment until the corresponding concentration-turbidity curve obtained is not a monotonic numerical curve.

[0014] In some embodiments, in the process of determining the optimal medicament dosage by performing the six-coagulation stirring experiment, the following steps are performed: reducing the third medicament concentration gradient to a fourth medicament concentration gradient; taking the medicament dosage concentration corresponding to the lowest supernatant turbidity value obtained in the process of adding the medicament in the three reactors according to the third medicament concentration gradient to perform the three-coagulation stirring experiment as the current initial medicament dosage concentration, adding the medicament in the six reactors according to the fourth medicament concentration gradient, and obtaining a plurality of supernatant turbidity values; forming a corresponding concentration-turbidity curve based on the obtained supernatant turbidity values and the corresponding medicament dosage concentrations, and determining the optimal medicament dosage by identifying the lowest point of the concentration-turbidity curve.

[0015] In some embodiments, in the high turbidity mode, the neural network model is trained by historical experimental data, and the turbidity of the raw water to be treated is input into the trained neural network model to determine the optimal medicament dosage, wherein the historical experimental data includes different medicament dosage concentrations and corresponding supernatant turbidity values at different medicament dosage concentrations.

[0016] In a second aspect, the application provides a decision system for optimal dosage of a medicament, which adopts the decision method for optimal dosage of a medicament as described in any of the embodiments of the first aspect to decide the optimal dosage of the medicament, and the system comprises: a running mode selection module for detecting the turbidity of raw water to be treated and a turbidity variation step, and selecting a running mode according to the detection result; a supernatant turbidity value acquisition module for acquiring supernatant turbidity values corresponding to different medicament dosage concentrations under the selected running mode through corresponding coagulation and stirring experiments or corresponding historical experimental data; and an optimal medicament dosage decision module for deciding the optimal medicament dosage according to the supernatant turbidity values and the corresponding medicament dosage concentrations.

[0017] Through the decision scheme for optimal dosage of a medicament as provided above, the embodiments of the application realize the automation, precision and high efficiency of medicament dosage in the water treatment process by establishing an intelligent decision framework. It can perceive the dynamics of raw water quality in real time, and automatically switch to the most suitable running mode according to the preset logic. By distinguishing different running modes, it can adapt to the actual water conditions, avoiding the blindness of traditional dosage methods, and significantly improving the scientificity and pertinence of the decision.

[0018] Further, in some embodiments, in the daily mode, first, the nine reactors are divided into three groups, and experiments are carried out in parallel and at different time with different concentration intervals. The system can acquire possible optimal dosage points with high frequency and wide range, which is an efficient and economical daily monitoring method. At the same time, the possibility that the regular experiment cannot find the optimal value is foreseen, and instead of simply stopping or alarming, the system seamlessly switches to the nine-union coagulation and stirring experiment with a wider coverage or finer gradient, and adjusts the concentration interval through iteration until the optimal dosage point is successfully locked. This design greatly improves the success rate and reliability of the system in finding the optimal solution. And only when necessary, the nine-union fine experiment with more resource consumption is started. This hierarchical strategy realizes the optimal configuration of reactor resources and experimental time consumption on the premise of ensuring the accuracy of the results, achieving the best balance between efficiency and cost.

[0019] Further, in some embodiments, under the emergency mode, first, a three-union experiment is carried out with a large gradient concentration. This sacrifices part of the accuracy, but in return, it ensures a fast detection speed and a wider concentration coverage, ensuring that a roughly effective action interval can be quickly captured, which occupies valuable time to respond to sudden deterioration of water quality. When the obtained concentration-turbidity curve is monotonic, the large gradient experiment is continuously carried out until an effective preliminary optimal point is found. When the preliminary optimal point is found, the system immediately switches to a more precise six-union experiment mode, which will automatically narrow the concentration gradient and carry out a more precise search centered on the preliminary optimal point. This seamless connection from large-range coarse adjustment to small-range fine adjustment takes into account both the speed of emergency response and the accuracy of the final decision.

[0020] Further, in some embodiments, in the high turbidity mode, by training a neural network model in advance with a large amount of historical experimental data, the successful experience and complex rules in dealing with high turbidity water are internalized, so that when a real high turbidity event occurs, the time-consuming physical coagulation experiment can be completely bypassed, and the optimal dosage can be responded to within seconds. This zero-time delay decision-making capability wins the most valuable response time for the user, and is the ultimate barrier to ensure water supply safety. BRIEF DESCRIPTION OF DRAWINGS

[0021] The above and other objects, features and advantages of the example embodiments of the present application will be more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which several embodiments of the present application are shown by way of example, and wherein like or corresponding elements show like or corresponding parts, by referring to which; drawings:

[0022] Figure 1 An example flowchart of a decision-making method for the optimal dosage of a medicament according to an embodiment of the present application is shown;

[0023] Figure 2 An example flowchart of a decision-making method for the optimal dosage of a medicament in a daily mode according to an embodiment of the present application is shown;

[0024] Figure 3 A concentration-turbidity curve diagram of a triple coagulation stirring experiment in a daily mode according to an embodiment of the present application is shown;

[0025] Figure 4 A concentration-turbidity curve diagram of a triple coagulation stirring experiment in a daily mode according to another embodiment of the present application is shown;

[0026] Figure 5 A concentration-turbidity curve diagram of a nine-coagulation stirring experiment in a daily mode according to another embodiment of the present application is shown;

[0027] Figure 6 An example flowchart of a decision-making method for the optimal dosage of a medicament in an emergency mode according to an embodiment of the present application is shown;

[0028] Figure 7 A concentration-turbidity curve diagram of a triple coagulation stirring experiment in an emergency mode according to an embodiment of the present application is shown;

[0029] Figure 8 A concentration-turbidity curve diagram of a six-coagulation stirring experiment in an emergency mode according to an embodiment of the present application is shown;

[0030] Figure 9 A concentration-turbidity curve diagram obtained in a high turbidity mode according to an embodiment of the present application is shown.

[0031] Figure 10 An exemplary structural block diagram of a decision system for determining the optimal dosage of a medicament according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0033] It should be understood that the terms “include” and “contain” used in the specification and claims of the present application indicate the presence of the described features, whole, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, whole, steps, operations, elements, components and / or sets thereof.

[0034] It should also be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. As used in the specification and claims of the present application, the singular forms “a”, “an” and “the” are intended to include plural forms unless the context clearly indicates otherwise. It should be further understood that the term “and / or” used in the specification and claims of the present application refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0035] Figure 1 An exemplary flow chart of a method 100 for determining the optimal dosage of a medicament according to an embodiment of the present application is shown.

[0036] As shown in step S110, the turbidity of the raw water to be treated and the turbidity change step are detected, and a running mode is selected according to the detection results. Figure 1

[0037] In the embodiments of the present application, the turbidity of the raw water to be treated is periodically acquired, and the difference between the turbidity at the current time and the turbidity at the last time is taken as the turbidity change step.

[0038] ​In the embodiments of the present application, the operation modes include: a daily mode, an emergency mode and a high turbidity mode. Specifically, the daily mode is selected when the turbidity of the raw water to be treated is less than or equal to a turbidity threshold and the turbidity change step of the raw water to be treated at two adjacent time instants is less than or equal to a turbidity change threshold. The emergency mode is selected when the turbidity of the raw water to be treated is less than or equal to the turbidity threshold and the turbidity change step of the raw water to be treated at two adjacent time instants is greater than the turbidity change threshold. The high turbidity mode is selected when the turbidity of the raw water to be treated is greater than the turbidity threshold.

[0039] In some embodiments of the present application, the turbidity threshold is 100 NTU and the turbidity change threshold is 5 NTU. In other embodiments of the present application, the turbidity threshold and the turbidity change threshold can also be set according to actual needs and historical experience.

[0040] Through step S110, the water quality is no longer judged based on only a single, static turbidity value, but a key dynamic indicator, the turbidity change step, is innovatively introduced. By considering both the good or bad of the current water quality (turbidity value) and the speed of water quality change (change step), the true state of the current water situation is accurately obtained, and the best response mode matching the current situation is automatically switched to. By introducing the turbidity change step, two completely different situations can be accurately distinguished: one is that the water quality is stable but slightly turbid, and the daily fine management should be adopted; the other is that the water quality seems clear but is rapidly deteriorating, which is actually a precursor to sudden pollution. This greatly improves the pertinence of decision-making. The establishment of the emergency mode can start the emergency plan in advance when the water quality has not yet exceeded the safety threshold but has shown a rapid deterioration trend. This gains valuable golden processing time for responding to sudden water pollution events and kills the risk in the embryonic state.

[0041] After step S110 is performed, in step S120, under the selected operation mode, the supernatant turbidity values corresponding to different medicament dosing concentrations are obtained through corresponding coagulation and stirring experiments or corresponding historical experimental data.

[0042] In the embodiments of the present application, in the daily mode, the nine reactors are evenly divided into three groups of reactors and are started at intervals of a preset time. The medicament is dosed in the reactors in each group according to a first medicament concentration gradient to perform triple coagulation and stirring experiments, and a plurality of supernatant turbidity values corresponding to each group of reactors are obtained, wherein the medicament dosing concentration intervals of each group of reactors are different.

[0043] In some embodiments of the present application, each group of reactors in the daily mode takes 60 minutes, and the triple coagulation and stirring experiments of the next group of reactors are started every 20 minutes after the start of one group of reactors. The nine reactors can ensure that three groups of experiments are performed at the same time, and the instrument outputs the supernatant turbidity value once every 20 minutes.

[0044] In the embodiments of the present application, in the emergency mode, the coagulant is added in three reactors according to a third coagulant concentration gradient to carry out triple coagulation and stirring experiments, and supernatant turbidity values under different coagulant concentrations are obtained, wherein the third coagulant concentration gradient is greater than the first coagulant concentration gradient.

[0045] In some embodiments of the present application, the first coagulant concentration gradient is 0.5 mg / L, and the third coagulant concentration gradient is 2 mg / L. In other embodiments of the present application, the first coagulant concentration gradient and the third coagulant concentration gradient can also be set according to actual needs and historical experience, which are not limited in the present application.

[0046] By using the grouping, staggered, and small gradient method in the daily mode with stable water quality, the most economical and effective dosing point is locked with the highest accuracy through continuous and comprehensive detection, so as to realize cost reduction and efficiency increase in daily operation. In the emergency mode with sudden change of water quality, the method is switched to a large gradient and small scale, sacrificing part of the accuracy to exchange for the extreme response speed, and the only goal is to find a reliable dosing range in the shortest time to quickly control the deteriorating water quality and ensure the safety of water supply. This differentiated strategy design makes the whole system not only thrifty in ordinary times, but also decisive in emergencies, showing high flexibility and practicality.

[0047] In the embodiments of the present application, in the process of obtaining the supernatant turbidity value in the daily mode and the emergency mode, first, the turbidity probe is controlled by the mechanical arm to enter the preset liquid level in the reactor for multi-angle automatic measurement, and a plurality of turbidity measurement results are obtained. Then, it is judged whether the deviation rates between the plurality of turbidity measurement results two by two exceed the preset deviation rate. In response to the deviation rates between the plurality of turbidity measurement results two by two exceeding the preset deviation rate, the step of controlling the turbidity probe by the mechanical arm to enter the preset liquid level in the reactor for multi-angle automatic measurement is returned. In response to the existence of the deviation rates between the plurality of turbidity measurement results two by two not exceeding the preset deviation rate, it is judged whether the deviation rates between the plurality of turbidity measurement results two by two all do not exceed the preset deviation rate. In response to the deviation rates between the plurality of turbidity measurement results two by two all not exceeding the preset deviation rate, the average value of the plurality of turbidity measurement results is taken as the turbidity value of the supernatant in the corresponding reactor. In response to the existence of the deviation rates between the plurality of turbidity measurement results two by two exceeding the preset deviation rate, the turbidity measurement result causing the deviation rate between the turbidity measurement results exceeding the preset deviation rate is removed, and the average value of the remaining turbidity measurement results is taken as the turbidity value of the supernatant in the corresponding reactor.

[0048] In some embodiments of the present application, the turbidity probe is controlled by a three-coordinate gantry robot to extend into the liquid level of the reactor by one-third, and the turbidity of the supernatant is measured. Each reactor robot measures three turbidity values, and after measuring the first value, the probe is rotated by 90° to measure the second value, and after measuring the second value, the probe is rotated by 90° to measure the third value.

[0049] In some embodiments of the present application, the preset deviation rate is 5%. When three turbidity values are measured by each reactor robot, if the deviation rate between the three values is not more than 5%, the three turbidity values are averaged to obtain the turbidity value of the supernatant. If the deviation rate between one turbidity value and any one of the other two turbidity values exceeds 5%, the remaining two turbidity values are averaged to obtain the turbidity value of the supernatant. If the deviation rate between the three turbidity values exceeds 5%, the process returns to the step of measuring the turbidity of the supernatant at the preset liquid level in the reactor by the robot-controlled turbidity probe.

[0050] By using the robot to accurately place the probe at the preset liquid level and performing multi-angle measurement, the errors caused by inconsistent liquid levels and random sampling points in manual operation are eliminated, and the repeatability of the measurement is ensured. By comparing the deviation rate between multiple measurement values in real time, this method can intelligently identify and process abnormal data. Whether the measurement is generally unstable (which triggers a retest) or individual outliers occur (which are automatically excluded), the method can independently respond to ensure the effectiveness of the data. This closed-loop process of multi-point measurement, deviation testing, abnormality exclusion, and average calculation greatly enhances the robustness of the measurement results, effectively avoids false readings caused by single accidental factors (such as small bubbles and suspended particles), and thus provides a solid and reliable data foundation for subsequent determination of the optimal medicament dosage.

[0051] In the high-turbidity mode, the supernatant turbidity values corresponding to different medicament dosing concentrations are obtained from historical experimental data.

[0052] After step S120 is performed, in step S130, the optimal medicament dosage is determined based on the supernatant turbidity value and the corresponding medicament dosage concentration.

[0053] In the daily mode, the specific process involved in step S130 can be referred to in the description of the present application. Figure 2 .

[0054] Figure 2 An exemplary flowchart of the present application in the daily mode is shown.

[0055] As Figure 2As shown, in step S210, a concentration-turbidity curve corresponding to each group of reactors is formed based on the obtained supernatant turbidity values and the corresponding medicament dosage concentrations. In step S220, it is determined whether all the concentration-turbidity curves are monotonous numerical curves. In response to all the concentration-turbidity curves not being monotonous numerical curves, in step S230, the optimal medicament dosage is determined by identifying the lowest point of the concentration-turbidity curve that is not a monotonous numerical curve. In response to all the concentration-turbidity curves being monotonous numerical curves, in step S240, a plurality of supernatant turbidity values and a corresponding concentration-turbidity curve are obtained by performing the nine-coagulation stirring experiment. In step S250, it is determined whether the concentration-turbidity curve is a monotonous numerical curve. In response to the concentration-turbidity curve not being a monotonous numerical curve, in step S260, the optimal medicament dosage is determined by identifying the lowest point of the concentration-turbidity curve. In response to the concentration-turbidity curve being a monotonous numerical curve, the medicament dosage concentration interval of the nine-coagulation stirring experiment is adjusted and the process returns to step S240, and the nine-coagulation stirring experiment is repeated until the corresponding concentration-turbidity curve obtained is not a monotonous numerical curve.

[0056] In an embodiment of the present application, the monotonous numerical curve includes a monotonously increasing numerical curve and a monotonously decreasing numerical curve, at which time, the optimal concentration range cannot be determined, and a range expansion experiment is required.

[0057] In an embodiment of the present application, in the process of obtaining a plurality of supernatant turbidity values by performing the nine-coagulation stirring experiment, the first medicament concentration gradient is reduced to a second medicament concentration gradient, and the medicament dosage concentration corresponding to the second-lowest supernatant turbidity value obtained in the process of adding medicament in each group of reactors to perform the three-coagulation stirring experiment is taken as the current initial medicament dosage concentration. The medicament is added in the nine reactors according to the second medicament concentration gradient, and a plurality of supernatant turbidity values are obtained.

[0058] In some embodiments of the present application, the second medicament concentration gradient is 0.3 mg / L. In other embodiments of the present application, the second medicament concentration gradient can also be set according to actual needs and historical experience, which is not limited in the present application.

[0059] In an embodiment of the present application, in the daily mode, the nine reactors are evenly divided into three groups of reactors and are started at intervals of 20 minutes, the first medicament concentration gradient is set to 0.5 mg / L, so that the medicament dosage concentration intervals of the three groups of reactors are different. The medicament is added in the three groups of reactors according to the first medicament concentration gradient to perform the three-coagulation stirring experiment, and a plurality of supernatant turbidity values corresponding to each group of reactors are obtained. Then, the concentration-turbidity curves corresponding to each group of reactors are obtained according to the plurality of supernatant turbidity values corresponding to each group of reactors, and the concentration-turbidity curves that are not monotonous numerical curves obtained are as shown in FIG. 2. Figure 3As shown. According to Figure 3 The lowest point of the concentration-turbidity curve corresponds to a dosage of 15.5 mg / L, which is taken as the optimal dosage.

[0060] In another embodiment of this application, in normal operation, the nine reactors are divided into three groups and started at 20-minute intervals. A first reagent concentration gradient of 0.5 mg / L is set, resulting in different reagent dosage ranges for each group of reactors. Reagents are added to each group of reactors according to the first reagent concentration gradient to conduct a triple coagulation and stirring experiment, obtaining multiple supernatant turbidity values ​​for each group of reactors. Then, concentration-turbidity curves are obtained for each group of reactors based on these multiple supernatant turbidity values. Furthermore, the concentration-turbidity curves for each reactor are monotonic numerical curves, where the lowest turbidity concentration-turbidity curve is shown below. Figure 4 As shown.

[0061] Next, the reagent dosage concentration of 15.5 mg / L, corresponding to the second lowest supernatant turbidity value in the concentration-turbidity curve, was used as the initial reagent dosage concentration. A second reagent concentration gradient of 0.3 mg / L was set, and the reagent was added to nine reactors according to the second reagent concentration gradient. A nine-stage coagulation and stirring experiment was conducted to obtain multiple supernatant turbidity values, and the corresponding concentration-turbidity curves were generated as follows: Figure 5 As shown. According to Figure 5 At this point, the concentration-turbidity curve is not a monotonic numerical curve, and the lowest point corresponds to a dosage of 16.7 mg / L, which is taken as the optimal dosage.

[0062] In the embodiments of this application, the specific process involved in executing step S130 in emergency mode can be found in the following reference. Figure 6 .

[0063] Figure 6 An exemplary flowchart illustrating an embodiment of this application is shown for determining the optimal dosage of a drug in emergency mode.

[0064] like Figure 6As shown in FIG. 6, in step S610, a current concentration-turbidity curve is formed based on the obtained supernatant turbidity values and the corresponding medicament dosage concentration. In step S620, it is determined whether the current concentration-turbidity curve is a monotonic numerical curve. In response to the current concentration-turbidity curve not being a monotonic numerical curve, in step S630, the preliminary optimal medicament dosage is obtained by identifying the lowest point of the current concentration-turbidity curve, and the optimal medicament dosage is determined by performing the six-coagulation stirring experiment. In response to the current concentration-turbidity curve being a monotonic numerical curve, in step S640, the lowest point of the current concentration-turbidity curve is taken as the initial medicament dosage concentration, the three-coagulation stirring experiment is repeated by adding medicament in the three reactors according to the third medicament concentration gradient to obtain a corresponding concentration-turbidity curve, and the process returns to step S620 until the obtained corresponding concentration-turbidity curve is not a monotonic numerical curve.

[0065] In the embodiments of the present application, in the process of determining the optimal medicament dosage by performing the six-coagulation stirring experiment, first, the third medicament concentration gradient is reduced to the fourth medicament concentration gradient. Then, the medicament dosage concentration corresponding to the lowest supernatant turbidity value obtained in the three-coagulation stirring experiment of adding medicament in the three reactors according to the third medicament concentration gradient is taken as the current initial medicament dosage concentration, medicament is added in the six reactors according to the fourth medicament concentration gradient, and a plurality of supernatant turbidity values are obtained. Then, a corresponding concentration-turbidity curve is formed based on the obtained supernatant turbidity values and the corresponding medicament dosage concentration, and the optimal medicament dosage is determined by identifying the lowest point of the concentration-turbidity curve.

[0066] In some embodiments of the present application, the fourth medicament concentration gradient is 0.5 mg / L. In other embodiments of the present application, the fourth medicament concentration gradient can also be set according to actual needs and history, which is not limited in the present application.

[0067] In an embodiment of the present application, in the emergency mode, the third medicament concentration gradient is set to 2 mg / L, medicament is added in the three reactors according to the third medicament concentration gradient to perform the three-coagulation stirring experiment, and a plurality of supernatant turbidity values corresponding to each reactor are obtained. Then, a corresponding concentration-turbidity curve is obtained based on the plurality of supernatant turbidity values corresponding to each reactor, and the obtained concentration-turbidity curve is as shown in FIG. 6. According to the concentration-turbidity curve, the medicament dosage corresponding to the lowest point of the concentration-turbidity curve is 20 mg / L. Figure 7 Figure 7

[0068] ​​Then, after the start of the three reactors for 20 minutes, the six reactors are started for six coagulation mixing experiments. During the six coagulation mixing experiments, 20 mg / L is used as the current initial reagent dosage concentration, and the fourth reagent concentration gradient is set to 0.5 mg / L. According to the fourth reagent concentration gradient, the reagent is added in the six reactors, the supernatant turbidity values are obtained, and the corresponding concentration-turbidity curve is obtained according to the supernatant turbidity values, as shown in Figure 8 According to Figure 8 , the lowest point of the concentration-turbidity curve corresponds to the reagent dosage of 21.5 mg / L, which is used as the optimal reagent dosage.

[0069] In the embodiment of the present application, in the high turbidity mode, the neural network model is trained by historical experimental data, and the turbidity of the raw water to be treated is input into the trained neural network model to determine the optimal reagent dosage. Specifically, the historical experimental data includes different reagent dosage concentrations and the corresponding supernatant turbidity values of different reagent dosage concentrations.

[0070] In an embodiment of the present application, the corresponding relationship between the reagent dosage concentration and the water sample turbidity in the historical experimental data is provided to the neural network model for learning, with the water sample turbidity as the input feature to predict the target characteristics. In the case of high turbidity, the linear relationship between the optimal reagent dosage concentration of the water sample and the water sample turbidity can be referred to Figure 9 Therefore, the neural network model can fit the optimal reagent dosage concentration in the case of high turbidity and determine the dosage.

[0071] In summary, through the optimal reagent dosage determination scheme provided above, the embodiment of the present application establishes an intelligent decision-making framework, realizes the automation, precision and high efficiency of reagent dosage in water treatment process. It can real-time perceive the dynamics of raw water quality, and automatically switch to the most suitable operation mode according to the preset logic. By distinguishing different operation modes, it can adapt to the actual water conditions, avoiding the blindness of traditional dosage method, and significantly improving the scientificity and pertinence of decision-making.

[0072] Further, in some embodiments, in the daily mode, first, by dividing the nine reactors into three groups, experiments are carried out in parallel and staggered at different concentration intervals, and the system can obtain the possible optimal dosing point at a high frequency and a wide range, which is an efficient and economical daily monitoring method. At the same time, it is foreseen that the regular experiment may not be able to find the optimal value, and instead of simply stopping or alarming, it seamlessly switches to the nine-union coagulation stirring experiment with a wider coverage or finer gradient, and adjusts the concentration interval through iteration until the optimal dosing point is successfully locked. This design greatly improves the success rate and reliability of the system to find the optimal solution. And only when necessary, the more resource-consuming nine-union fine experiment is started. This hierarchical strategy optimizes the configuration of reactor resources and experimental time consumption under the premise of ensuring the accuracy of the results, achieving the best balance between efficiency and cost.

[0073] Further, in some embodiments, under the emergency mode, first, a three-union experiment is carried out with a large gradient concentration. This sacrifices part of the accuracy, but in return, it ensures a very fast detection speed and a wider concentration coverage range, ensuring that a roughly effective action interval can be quickly captured, and valuable time is seized to respond to sudden water quality deterioration. When the obtained concentration-turbidity curve is monotonic, the large gradient experiment is continuously carried out until an effective preliminary optimal point is found. When the preliminary optimal point is found, it is immediately switched to a more precise six-union experiment mode, which will automatically narrow the concentration gradient and conduct a more precise search centered on the preliminary optimal point. This seamless connection from large-range coarse adjustment to small-range fine adjustment takes into account both the speed of emergency response and the accuracy of the final decision.

[0074] Further, in some embodiments, in the high turbidity mode, by pre-training a neural network model using a large amount of historical experimental data, the successful experience and complex rules in the past in dealing with high turbidity water are internalized, so that when a real high turbidity event occurs, the time-consuming physical coagulation experiment can be completely bypassed, and a second-level response of the optimal dosing amount is realized. This zero-time-delay decision-making capability wins the most valuable response time for the user, and is the ultimate barrier to ensure water supply safety.

[0075] The embodiment of the present application also provides a decision system for the optimal dosing amount of the medicament, which can use the foregoing decision method 100 for the optimal dosing amount of the medicament to decide the optimal dosing amount of the medicament, or can use other methods to decide the optimal dosing amount of the medicament, which is not limited herein.

[0076] Figure 10 An exemplary structural block diagram of the decision system for the optimal dosing amount of the medicament is shown.

[0077] As Figure 10As shown, the system 1000 comprises a running mode selection module 1010, a supernatant turbidity value acquisition module 1020 and an optimal coagulant dosage decision module 1030. In the embodiments of the present application, the running mode selection module 1010, the supernatant turbidity value acquisition module 1020 and the optimal coagulant dosage decision module 1030 can be separate units or can be integrated in the same controller, which is not limited in the present application.

[0078] Specifically, the running mode selection module 1010 is configured to detect the turbidity of the raw water to be treated and the turbidity variation step, and select a running mode according to the detection result.

[0079] Specifically, the supernatant turbidity value acquisition module 1020 is configured to acquire the supernatant turbidity value corresponding to different coagulant dosage concentrations under the selected running mode through corresponding coagulation and stirring experiments or corresponding historical experimental data.

[0080] Specifically, the optimal coagulant dosage decision module 1030 is configured to determine the optimal coagulant dosage according to the supernatant turbidity value and the corresponding coagulant dosage concentration.

[0081] When the system 1000 adopts the aforementioned optimal coagulant dosage decision method 100 to determine the optimal coagulant dosage, the aforementioned step S110 is performed by the running mode selection module 1010, the aforementioned step S120 is performed by the supernatant turbidity value acquisition module 1020, and the aforementioned step S130 is performed by the optimal coagulant dosage decision module 1030. The specific execution process can be referred to the foregoing, which will not be described here.

[0082] Although the embodiments of the present application have been shown and described herein, it should be apparent to those skilled in the art that such embodiments are merely illustrative of the present application. Many changes, modifications and substitutions can be suggested to one skilled in the art without departing from the spirit and scope of the present application. It is therefore intended that the following claims be interpreted as encompassing all such changes, modifications and substitutions.

Claims

1. A method of determining an optimal dosage of a medicament, characterized by, The method comprises the following steps: detecting the turbidity of raw water to be treated and the turbidity change step, and selecting an operation mode according to the detection result, wherein the difference between the turbidity at the current time and the turbidity at the last time is taken as the turbidity change step; under the selected operation mode, the supernatant turbidity values corresponding to different medicament dosing concentrations are obtained through corresponding coagulation stirring experiments or corresponding historical experimental data; the optimal medicament dosing amount is determined according to the supernatant turbidity values and the corresponding medicament dosing concentrations; the operation mode includes: daily mode, emergency mode and high turbidity mode; when the turbidity of raw water to be treated is less than or equal to the turbidity threshold value and the turbidity change step of raw water to be treated at two adjacent times is less than or equal to the turbidity change threshold value, the daily mode is selected; when the turbidity of raw water to be treated is less than or equal to the turbidity threshold value and the turbidity change step of raw water to be treated at two adjacent times is greater than the turbidity change threshold value, the emergency mode is selected; when the turbidity of raw water to be treated is greater than the turbidity threshold value, the high turbidity mode is selected; under the daily mode, nine reactors are evenly divided into three groups of reactors and are started at intervals of a preset time, medicament is dosed in each group of reactors according to a first medicament concentration gradient to perform triple coagulation stirring experiments, and a plurality of supernatant turbidity values corresponding to each group of reactors are obtained, wherein the medicament dosing concentration intervals of each group of reactors are different; under the daily mode, the following steps are performed in the process of determining the optimal medicament dosing amount according to the supernatant turbidity values and the corresponding medicament dosing concentrations: based on the obtained supernatant turbidity values and the corresponding medicament dosing concentrations, a concentration-turbidity curve corresponding to each group of reactors is formed; it is judged whether all concentration-turbidity curves are monotonous numerical curves; in response to all concentration-turbidity curves not being monotonous numerical curves, the optimal medicament dosing amount is determined by identifying the lowest point of the concentration-turbidity curve which is not a monotonous numerical curve; in response to all concentration-turbidity curves being monotonous numerical curves, a plurality of supernatant turbidity values and a corresponding concentration-turbidity curve are obtained by performing nine coagulation stirring experiments, and it is judged whether the concentration-turbidity curve is a monotonous numerical curve; in response to the concentration-turbidity curve not being a monotonous numerical curve, the optimal medicament dosing amount is determined by identifying the lowest point of the concentration-turbidity curve; in response to the concentration-turbidity curve being a monotonous numerical curve, the medicament dosing concentration interval of the nine coagulation stirring experiments is adjusted and the nine coagulation stirring experiments are repeated until the corresponding concentration-turbidity curve obtained is not a monotonous numerical curve; under the emergency mode, medicament is dosed in three reactors according to a third medicament concentration gradient to perform triple coagulation stirring experiments, and the supernatant turbidity values under different medicament concentrations are obtained, wherein the third medicament concentration gradient is greater than the first medicament concentration gradient; under the high turbidity mode, a neural network model is trained through historical experimental data, the turbidity of raw water to be treated is input into the trained neural network model to determine the optimal medicament dosing amount, wherein the historical experimental data includes different medicament dosing concentrations and supernatant turbidity values corresponding to different medicament dosing concentrations.

2. The optimal amount of medicament dosing decision method according to claim 1, characterized by, In the process of obtaining multiple supernatant turbidity values by performing nine coagulation stirring experiments, the first medicament concentration gradient is reduced to the second medicament concentration gradient, and the medicament dosage concentration corresponding to the second-lowest supernatant turbidity value obtained in the three coagulation stirring experiments performed by adding medicament in each group of reactors is taken as the current initial medicament dosage concentration. The medicament is added in the nine reactors according to the second medicament concentration gradient, and multiple supernatant turbidity values are obtained.

3. The optimal amount of medicament dosing decision method according to claim 1, characterized in that, In the emergency mode, in the process of determining the optimal medicament dosage according to the supernatant turbidity values and the corresponding medicament dosage concentrations, the following steps are performed: forming a current concentration-turbidity curve based on the obtained supernatant turbidity values and the corresponding medicament dosage concentrations; determining whether the current concentration-turbidity curve is a monotonic numerical curve; in response to the current concentration-turbidity curve not being a monotonic numerical curve, obtaining a preliminary optimal medicament dosage by identifying the lowest point of the current concentration-turbidity curve, and determining the optimal medicament dosage by performing six coagulation stirring experiments; in response to the current concentration-turbidity curve being a monotonic numerical curve, taking the lowest point of the current concentration-turbidity curve as the initial medicament dosage concentration, and repeating the three coagulation stirring experiments performed by adding medicament in three reactors according to the third medicament concentration gradient until the corresponding concentration-turbidity curve obtained is not a monotonic numerical curve.

4. The optimal amount of medicament dosing decision method according to claim 3, characterized by, In the process of determining the optimal medicament dosage by performing six coagulation stirring experiments, the following steps are performed: reducing the third medicament concentration gradient to the fourth medicament concentration gradient; taking the medicament dosage concentration corresponding to the lowest supernatant turbidity value obtained in the three coagulation stirring experiments performed by adding medicament in three reactors according to the third medicament concentration gradient as the current initial medicament dosage concentration, adding medicament in six reactors according to the fourth medicament concentration gradient, and obtaining multiple supernatant turbidity values; forming a corresponding concentration-turbidity curve based on the obtained supernatant turbidity values and the corresponding medicament dosage concentrations, and determining the optimal medicament dosage by identifying the lowest point of the concentration-turbidity curve.

5. A decision system for optimal dosage of a medicament, characterized by, The system comprises: an operating mode selection module for detecting the turbidity of the raw water to be treated and the turbidity change step, and selecting an operating mode according to the detection results, the operating mode including a daily mode, an emergency mode, and a high turbidity mode; wherein when the turbidity of the raw water to be treated is less than or equal to a turbidity threshold value and the turbidity change step of the raw water to be treated at two adjacent time points is less than or equal to a turbidity change threshold value, the daily mode is selected; when the turbidity of the raw water to be treated is less than or equal to the turbidity threshold value and the turbidity change step of the raw water to be treated at two adjacent time points is greater than the turbidity change threshold value, the emergency mode is selected; and when the turbidity of the raw water to be treated is greater than the turbidity threshold value, the high turbidity mode is selected; a supernatant turbidity value acquisition module for obtaining supernatant turbidity values corresponding to different medicament dosage concentrations by corresponding coagulation stirring experiments or corresponding historical experimental data under the selected operating mode; and a supernatant turbidity value acquisition module for obtaining supernatant turbidity values corresponding to different medicament dosage concentrations by corresponding coagulation stirring experiments or corresponding historical experimental data under the selected operating mode. The optimal medicament dosage decision module is configured to determine an optimal medicament dosage based on the supernatant turbidity value and the corresponding medicament dosage concentration.

Images