Method for determining the amount of medicament to be added and device therefor

Abstract

The application provides a medicament adding amount determination method and device. The method comprises the following steps: obtaining a first state parameter of a concentration tank corresponding to a to-be-treated mixture at a current moment and historical medicament amounts added to the to-be-treated mixture at each moment in a previous medicament adding period; determining a medicament efficacy of the medicament in the concentration tank at the current moment according to the historical medicament amounts, and weighting the medicament efficacy and the first state parameter to determine a first medicament amount added to the concentration tank at the current moment; inputting the medicament efficacy and the first state parameter into a preset network model to obtain a second medicament amount output by the network model; inputting the medicament efficacy and a second state parameter of the concentration tank in the previous medicament adding period into the network model to obtain a second supernatant depth output by the network model; and determining a target medicament amount added to the concentration tank at the current moment according to a difference between the second supernatant depth and a first supernatant depth in the first state parameter, the first medicament amount and the second medicament amount. Therefore, the accuracy of the medicament adding amount is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal mines, and in particular to a method for determining the amount of reagent to be added and a device thereof. BACKGROUND

[0002] In order to improve the recovery rate of clean coal, a certain amount of reagent (such as a flocculating agent) needs to be added to the slime water to promote the settlement of the slime. However, according to experience, the amount of reagent added manually is inaccurate, which causes waste of reagent or a low recovery rate of clean coal, and thus a high production cost. SUMMARY

[0003] The present application provides a method for determining the amount of reagent to be added and a device thereof. The specific scheme is as follows:

[0004] In one aspect, the present application provides a method for determining the amount of reagent to be added, comprising:

[0005] obtaining, from a system, a first state parameter of a concentration tank corresponding to a mixture to be treated at a current time and a historical reagent amount added to the mixture to be treated at each time in a previous reagent adding period, wherein the first state parameter comprises a first supernatant depth, an inlet flow of the concentration tank, and a particle size of the mixture to be treated;

[0006] determining the efficacy of the reagent in the concentration tank at the current time according to the historical reagent amounts, and weighting the efficacy and the first state parameter to determine a first reagent amount to be added to the concentration tank at the current time;

[0007] inputting the efficacy and the first state parameter into a preset network model to obtain a second reagent amount output by the network model;

[0008] inputting the efficacy and a second state parameter of the concentration tank in the previous reagent adding period into the network model to obtain a second supernatant depth output by the network model; wherein the second state parameter is obtained from the system when determining the amount of reagent to be added, and the second state parameter is consistent with the type of the first state parameter;

[0009] determining a target reagent amount to be added to the concentration tank at the current time according to a difference between the second supernatant depth and the first supernatant depth, the first reagent amount, and the second reagent amount; wherein when the difference is greater than a threshold value, the first reagent amount is determined as the target reagent amount, and when the difference is less than the threshold value, the second reagent amount is determined as the target reagent amount.

[0010] In another aspect, the present application provides a device for determining the amount of reagent to be added, comprising:

[0011] The acquisition module is configured to acquire a first state parameter of a concentration tank corresponding to a mixture to be treated at a current time and historical dosages of medicaments added to the mixture to be treated at each time in a previous dosing cycle from the system, wherein the first state parameter includes a first supernatant depth, a concentration tank inlet flow rate, and a particle size of the mixture to be treated;

[0012] The first determination module is configured to determine an efficacy of the medicament in the concentration tank at the current time according to the historical dosages of medicaments, and determine a first dosage of medicament to be added to the concentration tank at the current time by weighting the efficacy and the first state parameter;

[0013] The second determination module is configured to input the efficacy and the first state parameter into a preset network model to obtain a second dosage of medicament output by the network model;

[0014] The second determination module is configured to input the efficacy and a second state parameter of the concentration tank in the previous dosing cycle into the network model to obtain a second supernatant depth output by the network model; wherein the second state parameter is acquired from the system when determining the dosage of medicament, and the second state parameter is consistent with the type of the first state parameter;

[0015] The third determination module is configured to determine a target dosage of medicament to be added to the concentration tank at the current time according to a difference between the second supernatant depth and the first supernatant depth, the first dosage of medicament, and the second dosage of medicament; wherein when the difference is greater than a threshold value, the first dosage of medicament is determined as the target dosage of medicament, and when the difference is less than the threshold value, the second dosage of medicament is determined as the target dosage of medicament.

[0016] Another aspect of the present application provides a computer device, comprising a processor and a memory;

[0017] The processor runs a program corresponding to executable program code stored in the memory by reading the executable program code, so as to implement the method of the above-mentioned embodiments.

[0018] Another aspect of the present application provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the method of the above-mentioned embodiments.

[0019] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0020] The above-mentioned and / or additional aspects and advantages of the present application will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

[0021] Figure 1A flowchart illustrating a method for determining the dosage of a drug as provided in an embodiment of this application;

[0022] Figure 2 A flowchart illustrating a method for determining the dosage of a drug as provided in an embodiment of this application;

[0023] Figure 3 A flowchart illustrating a method for determining the dosage of a drug as provided in an embodiment of this application;

[0024] Figure 4 A flowchart illustrating a method for determining the dosage of a drug as provided in an embodiment of this application;

[0025] Figure 5 This is a schematic diagram of a drug addition determination device provided in an embodiment of this application. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0027] The method for determining the dosage of the drug according to embodiments of this application is described below with reference to the accompanying drawings.

[0028] Figure 1 This is a flowchart illustrating a method for determining the dosage of a drug as provided in an embodiment of this application.

[0029] The method for determining the dosage of the drug in this application embodiment is executed by the drug dosage determination device (hereinafter referred to as the determination device) provided in this application embodiment. The device can be configured in a computer device to improve the accuracy of the dosage of the drug.

[0030] like Figure 1 As shown, the method for determining the dosage of this agent includes:

[0031] Step 101: Obtain the first state parameters of the thickener corresponding to the mixture to be treated at the current time, and the historical dosage of the drug added to the mixture to be treated at each time in the previous dosing cycle. The first state parameters include the first supernatant depth, the inlet flow rate of the thickener, and the particle size of the mixture to be treated.

[0032] Optionally, the first state parameters may also include factors that affect the efficacy of the agent, such as the temperature and pH of the mixture to be treated; this application does not impose any restrictions on this.

[0033] The mixture to be treated can include solid particles and water, and the solid particles can be coal slime or dust, which are not limited in the present application. In the present application, the mixture can be first input into the buffer pool, and then gradually transported from the buffer pool to one or more concentration pools. Then, the mixture can be separated into solid particles and water in the concentration pool. When the mixture is transported to the concentration pool, an agent (such as a flocculating agent) that helps the aggregation of solid particles can be added to the mixture to improve the speed and effect of the separation of solid particles and water in the mixture.

[0034] For example, the coal slime water in the concentration pool gradually aggregates under the action of the flocculating agent, thereby forming a precipitate with a larger density, and the settling speed of the coal slime particles in the coal slime water is greatly improved.

[0035] In the present application, the mixture in the buffer pool can be periodically transported to the concentration pool. At the start of each dosing period, the target agent amount required in each dosing period is determined. During the transportation of the mixture from the buffer pool to the concentration pool through the conveying pipeline in each dosing period, the agent of the target agent amount corresponding to each dosing period can be added to the mixture at different times in each dosing period to ensure the uniformity of the agent in the mixture, thereby improving the separation effect of the solid particles and water in the mixture. In addition, the agent amount of the agent added at each time can be saved in the system. Thus, the historical agent amount added to the mixture to be treated at each time in the previous dosing period at the current time can be obtained from the system.

[0036] The current time is the start time of the current dosing period.

[0037] In the present application, the supernatant depth of the concentration pool at each time can be obtained by the sludge interface instrument deployed in the concentration pool, the flow of the mixture flowing into the concentration pool at each time can be obtained by the flow detector, and the particle size of the solid particles in the mixture can be determined by measurement in advance. Then, the state parameters of the concentration pool at each time can be saved in the data storage module. Thus, when determining the agent addition amount, the first state parameter of the concentration pool can be obtained from the system.

[0038] In step 102, the effect of the agent in the concentration pool at the current time is determined according to the historical agent amounts, and the first state parameter is weighted to determine the first agent amount added to the concentration pool at the current time.

[0039] In the present application, the effect of the agent has the characteristic of short-term duration, and the effect changes nonlinearly with time during the duration. The effect of the agent can be determined by the following formula:

[0040]

[0041] wherein, is a preset weight vector, denotes the concentration of the medicament at the time point, is each time point in the previous dosing cycle, and have the same dimension. The medicament efficacy at the time point can be determined based on the historical medicament amount added at the time point.

[0042] Alternatively, when the medicament includes a cationic medicament and an anionic medicament, the efficacy of the cationic medicament in the concentration tank at the current time point can be determined according to the historical cationic medicament amount added to the mixture to be treated at each time point in the previous dosing cycle by the above formula. The efficacy of the anionic medicament in the concentration tank at the current time point can be determined according to the historical anionic medicament amount added to the mixture to be treated at each time point in the previous dosing cycle by the above formula.

[0043] In the present application, after determining the efficacy of the medicament in the concentration tank at the current time point, the efficacy of the medicament in the concentration tank at the current time point and the first state parameter can be weighted based on a preset weight, and the weighted sum of the efficacy and the first state parameter is used to determine the first medicament amount to be added to the concentration tank at the current time point. The preset weight can be determined and adjusted in advance according to human experience.

[0044] When the medicament includes a cationic medicament and an anionic medicament, after determining the efficacy of the cationic medicament in the concentration tank at the current time point, the efficacy of the cationic medicament in the concentration tank at the current time point and the first state parameter can be weighted based on a preset weight, and the weighted sum of the efficacy and the first state parameter is used to determine the first medicament amount of the cationic medicament to be added to the concentration tank at the current time point. After determining the efficacy of the anionic medicament in the concentration tank at the current time point, the efficacy of the anionic medicament in the concentration tank at the current time point and the first state parameter can be weighted based on a preset weight, and the weighted sum of the efficacy and the first state parameter is used to determine the first medicament amount of the anionic medicament to be added to the concentration tank at the current time point.

[0045] Step 103, input the efficacy and the first state parameter into a preset network model to obtain a second medicament amount output by the network model.

[0046] In the present application, a network model for predicting the medicament addition amount required to add medicament in the next dosing cycle can be trained in advance, and the network model is also used to output the supernatant depth of the post-concentration tank of the predicted medicament addition amount in the next dosing cycle.

[0047] After training the network model, the efficacy of the medicament in the concentration tank at the current time point and the first state parameter can be input into the network model to obtain the second medicament amount to be added at the current time point output by the network model.

[0048] In addition, the training manner of the network model is as follows: obtaining the third state parameters of each historical dosing cycle concentration tank, the historical medicament amount added to the mixture to be treated at each time in each historical dosing cycle, and the labeled supernatant depth after treatment corresponding to each historical dosing cycle, wherein the third state parameters include the third supernatant depth, the concentration tank inlet flow, and the particle size of the mixture to be treated. Then, according to the historical medicament amount of the previous dosing cycle of each historical dosing cycle, the efficacy of the medicament in the concentration tank at the initial time of each historical dosing cycle is determined, and the efficacy at the initial time of each historical dosing cycle and the third state parameters are input into the initial network model to obtain the predicted medicament amount and the predicted supernatant depth output by the initial network model. Then, according to the difference between the predicted supernatant depth and the labeled supernatant depth corresponding to each historical dosing cycle, a loss value is determined, and the initial network model is adjusted according to the loss value until the difference between the predicted supernatant depth output by the initial network model and the corresponding labeled supernatant depth tends to a preset value, and a final network model is obtained.

[0049] Step 104, input the efficacy and the second state parameters of the concentration tank of the previous dosing cycle into the network model to obtain the second supernatant depth output by the network model.

[0050] It can be understood that the second supernatant depth is the supernatant depth at the current time predicted by the network model.

[0051] Step 105, according to the difference between the second supernatant depth and the first supernatant depth, the first medicament amount and the second medicament amount, the target medicament amount added to the concentration tank at the current time is determined.

[0052] In this application, when the difference between the second supernatant depth and the first supernatant depth is greater than a threshold value, it indicates that the accuracy of the network model is not high. Therefore, the first medicament amount can be determined as the target medicament amount. Thus, the accuracy of the medicament addition amount is ensured.

[0053] When the difference between the second supernatant depth and the first supernatant depth is less than a threshold value, it indicates that the accuracy of the network model is high. Therefore, the second medicament amount can be determined as the target medicament amount. Thus, the accuracy of the medicament addition amount is ensured.

[0054] Optionally, the weight values corresponding to the first medicament amount and the second medicament amount can also be determined according to the difference between the second supernatant depth and the first supernatant depth. Then, the first medicament amount and the second medicament amount are weighted and summed by using the weight values to determine the target medicament amount added to the concentration tank at the current time.

[0055] When the medicament includes the cationic medicament and the anionic medicament, the efficacy of the cationic medicament and the second state parameter of the concentration tank in the previous medicament adding cycle can be input into the network model to obtain the second supernatant depth output by the network model. Then, according to the difference between the second supernatant depth and the first supernatant depth, the first medicament amount and the second medicament amount, the target medicament amount of the cationic medicament added into the concentration tank at the current time is determined. The efficacy of the anionic medicament and the second state parameter of the concentration tank in the previous medicament adding cycle can be input into the network model to obtain the second supernatant depth output by the network model. Then, according to the difference between the second supernatant depth and the first supernatant depth, the first medicament amount and the second medicament amount, the target medicament amount of the anionic medicament added into the concentration tank at the current time is determined.

[0056] It can be understood that the first medicament amount is determined based on artificial experience, and the reliability of the first medicament amount is relatively high, and there is no phenomenon of large error. However, since the medicament adding amount is affected by many factors, the accuracy of the determined first medicament amount can not be high. In addition, the accuracy of the second medicament amount predicted by the network model output is greatly related to the number and quality of the training data of the network model. When the training data sufficiently contains data in various scenes, the accuracy of the second medicament amount predicted by the network model trained based on the training data is relatively high. When the training data does not sufficiently contain data in various scenes, the accuracy of the second medicament amount predicted by the network model trained based on the training data can be relatively low. Due to the difference between medicaments and the difference between mixed substance processing environments, the network model trained by the training data collected in other mixed substance processing environments can not accurately predict the second medicament amount of the current concentration tank.

[0057] In the present application, the accuracy of the network model can be evaluated first, and then the first medicament amount and the second medicament amount are fused based on the accuracy of the network model. The influence of the inaccurate second medicament amount output by the network model on determining the target medicament amount is reduced. Thus, the accuracy of determining the medicament adding amount is improved.

[0058] In the present application, when the difference between the second supernatant depth and the first supernatant depth is greater than a threshold value, it indicates that the accuracy of the network model is relatively low. At this time, the efficacy of the medicament in the concentration tank at the current time, the first state parameter and the actual fourth supernatant depth of the concentration tank after the current medicament adding cycle is completed can be used to continue training the network model to adjust the network model and improve the accuracy of the network model.

[0059] In the present application, after obtaining the first state parameters of the concentration tank corresponding to the mixture to be treated at the current time, including the first supernatant depth, the concentration tank inlet flow, the particle size of the mixture to be treated, and the historical reagent amount added to the mixture to be treated at each time in the previous reagent adding period, the efficacy of the reagent in the concentration tank at the current time is determined according to each historical reagent amount, and the efficacy and the first state parameters are weighted to determine the first reagent amount added to the concentration tank at the current time. Then, the efficacy and the first state parameters are input into a preset network model to obtain the second reagent amount output by the network model, and the efficacy and the second state parameters of the concentration tank in the previous reagent adding period are input into the network model to obtain the second supernatant depth output by the network model. The target reagent amount added to the concentration tank at the current time is determined according to the difference between the second supernatant depth and the first supernatant depth, the first reagent amount, and the second reagent amount. Thus, the first reagent amount output by the artificial model is fused with the second reagent amount output by the network model to reduce the influence of the inaccurate first reagent amount output by the network model on the determination of the target reagent amount, thereby improving the accuracy of the reagent adding amount.

[0060] Figure 2 A flowchart of a reagent adding amount determination method provided by an embodiment of the present application is shown.

[0061] As shown in Figure 2 , the reagent adding amount determination method comprises:

[0062] Step 201: obtaining the first state parameters of the concentration tank corresponding to the mixture to be treated at the current time, and the historical reagent amount added to the mixture to be treated at each time in the previous reagent adding period, wherein the first state parameters include the first supernatant depth, the concentration tank inlet flow, and the particle size of the mixture to be treated.

[0063] Step 202: determining the efficacy of the reagent in the concentration tank at the current time according to each historical reagent amount, and weighting the efficacy and the first state parameters to determine the first reagent amount added to the concentration tank at the current time.

[0064] Step 203: inputting the efficacy and the first state parameters into a preset network model to obtain the second reagent amount output by the network model.

[0065] Step 204: inputting the efficacy and the second state parameters of the concentration tank in the previous reagent adding period into the network model to obtain the second supernatant depth output by the network model.

[0066] In the present application, the specific implementation process of steps 201-204 can be referred to the detailed description of any embodiment of the present application, which will not be repeated here.

[0067] Step 205: querying the weight mapping relationship table to determine the first weight of the first reagent amount corresponding to the difference and the second weight of the second reagent amount.

[0068] In the present application, the weight mapping relationship table can be pre-set in the system. When the difference value is larger, the second weight of the second medicament amount corresponding to the difference value in the weight mapping relationship table is smaller, and the first weight of the first medicament amount corresponding to the difference value is larger. Thus, the weights of the first medicament amount and the second medicament amount are determined by querying the weight mapping relationship table based on the difference value, which reduces the complexity of determining the medicament addition amount. Therefore, the accuracy of determining the medicament addition amount is improved, and the efficiency of determining the medicament addition amount is also improved.

[0069] In step 206, the first medicament amount and the second medicament amount are weighted by using the first weight and the second weight respectively, and the weighted sum of the first medicament amount and the second medicament amount is determined as the target medicament amount.

[0070] In the present application, after the first state parameters of the concentration tank corresponding to the mixture to be treated at the current time, including the first supernatant depth, the concentration tank inlet flow, the particle size of the mixture to be treated, and the historical medicament amounts added to the mixture to be treated at each time in the previous medicament addition period are obtained, the efficacy of the medicament in the concentration tank at the current time can be determined according to each historical medicament amount, and the efficacy and the first state parameters are weighted to determine the first medicament amount added to the concentration tank at the current time. Then, the efficacy and the second state parameters of the concentration tank in the previous medicament addition period are input into the preset network model to obtain the second medicament amount output by the network model, and the efficacy and the second supernatant depth output by the network model are input into the network model to obtain the second medicament amount output by the network model. Then, the weight mapping relationship table is queried to determine the first weight of the first medicament amount and the second weight of the second medicament amount corresponding to the difference value, so as to weight the first medicament amount and the second medicament amount by using the first weight and the second weight respectively, and determine the weighted sum of the first medicament amount and the second medicament amount as the target medicament amount. Thus, the weights of the first medicament amount and the second medicament amount are determined by querying the weight mapping relationship table based on the difference value, which reduces the complexity of determining the medicament addition amount. Therefore, the accuracy of determining the medicament addition amount is improved, and the efficiency of determining the medicament addition amount is also improved.

[0071] Figure 3 A flowchart of a medicament addition amount determination method provided by an embodiment of the present application is shown.

[0072] As shown in Figure 3 , the medicament addition amount determination method comprises the following steps.

[0073] In step 301, the first state parameters of the concentration tank corresponding to the mixture to be treated at the current time, and the historical medicament amounts added to the mixture to be treated at each time in the previous medicament addition period are obtained, wherein the first state parameters include the first supernatant depth, the concentration tank inlet flow, and the particle size of the mixture to be treated.

[0074] At step 302, the drug efficacy of the drug in the concentration tank at the current time is determined according to each historical drug amount, and the drug efficacy and the first state parameter are weighted to determine the first drug amount added to the concentration tank at the current time.

[0075] At step 303, the drug efficacy and the first state parameter are input into a preset network model to obtain a second drug amount output by the network model.

[0076] At step 304, the drug efficacy and the second state parameter of the concentration tank in the previous dosing cycle are input into the network model to obtain a second supernatant depth output by the network model.

[0077] In the present application, the specific implementation process of steps 301-304 can be referred to the detailed description of any embodiment of the present application, and will not be repeated here.

[0078] At step 305, the difference is input into a preset weight function to determine a first weight corresponding to the first drug amount and a second weight corresponding to the second drug amount.

[0079] In the present application, the target drug amount, the first drug amount and the second drug amount determined in each historical dosing cycle can be fitted to determine the weight function. Then, the difference can be input into the weight function to determine the first weight corresponding to the first drug amount and the second weight corresponding to the second drug amount. Thus, the accuracy of determining the first weight and the second weight is further improved.

[0080] At step 306, the first weight and the second weight are used to weight the first drug amount and the second drug amount respectively, and the weighted sum of the first drug amount and the second drug amount is determined as the target drug amount.

[0081] In the present application, the specific implementation process of step 306 can be referred to the detailed description of any embodiment of the present application, and will not be repeated here.

[0082] In the present application, after the first state parameters of the concentration tank corresponding to the mixture to be treated at the current time, including the first supernatant depth, the concentration tank inlet flow, the particle size of the mixture to be treated, and the historical reagent amount added to the mixture to be treated at each time in the previous reagent adding period are obtained, the efficacy of the reagent in the concentration tank at the current time can be determined according to each historical reagent amount, and the efficacy and the first state parameters are weighted to determine the first reagent amount added to the concentration tank at the current time. Then, the efficacy and the first state parameters are input into a preset network model to obtain a second reagent amount output by the network model, and the efficacy and the second state parameters of the concentration tank in the previous reagent adding period are input into the network model to obtain a second supernatant depth output by the network model. Then, the difference between the second supernatant depth and the first supernatant depth, the first reagent amount and the second reagent amount are input into a preset weight function to determine a first weight corresponding to the first reagent amount and a second weight corresponding to the second reagent amount, so as to weight the first reagent amount and the second reagent amount by using the first weight and the second weight respectively, and the weighted sum of the first reagent amount and the second reagent amount is determined as the target reagent amount. Thus, the accuracy of determining the reagent adding amount is improved.

[0083] Figure 4 A flowchart of a reagent adding amount determination method provided by an embodiment of the present application is shown.

[0084] As shown in the flowchart, the reagent adding amount determination method comprises the following steps. Figure 4

[0085] In step 401, the first state parameters of the concentration tank corresponding to the mixture to be treated at the current time, and the historical reagent amount added to the mixture to be treated at each time in the previous reagent adding period are obtained, wherein the first state parameters include the first supernatant depth, the concentration tank inlet flow, and the particle size of the mixture to be treated.

[0086] In step 402, the efficacy of the reagent in the concentration tank at the current time is determined according to each historical reagent amount, and the efficacy and the first state parameters are weighted to determine the first reagent amount added to the concentration tank at the current time.

[0087] In step 403, the efficacy and the first state parameters are input into a preset network model to obtain a second reagent amount output by the network model.

[0088] In step 404, the efficacy and the second state parameters of the concentration tank in the previous reagent adding period are input into the network model to obtain a second supernatant depth output by the network model.

[0089] In step 405, the target reagent amount added to the concentration tank at the current time is determined according to the difference between the second supernatant depth and the first supernatant depth, the first reagent amount and the second reagent amount.

[0090] In the present application, the specific implementation process of steps 401-405 can be referred to the detailed description of any embodiment of the present application, which will not be repeated here.​

[0091] In step 406, a dosing instruction is sent to the dosing execution component to control the dosing execution component to release the target drug dose of the drug.

[0092] In this application, the dosing instruction can be generated based on the target drug dose, and the dosing instruction is sent to the dosing execution component to control the dosing execution component to release the target drug dose of the drug in the current dosing period. The dosing instruction can be a start drug preparation instruction, and can include information such as the amount of drug to be released.

[0093] The dosing execution component can include a cation drug preparation component and an anion drug preparation component. The cation drug preparation component is deployed in the buffer tank for releasing cation drugs. The anion drug preparation component is deployed on the pipeline between the buffer tank and the concentration tank for releasing anion drugs. Thus, multi-point addition is achieved to ensure that the drug is fully mixed with the mixture to be treated.

[0094] In this application, after obtaining the first state parameters of the concentration tank corresponding to the mixture to be treated at the current time, including the first supernatant depth, the concentration tank inlet flow, the particle size of the mixture to be treated, and the historical drug dose added to the mixture to be treated at each time in the previous dosing period, the drug efficacy of the drug in the concentration tank at the current time is determined according to each historical drug dose, and the drug efficacy and the first state parameters are weighted to determine the first drug dose added to the concentration tank at the current time. Then, the drug efficacy and the second state parameters of the concentration tank in the previous dosing period are input into the preset network model to obtain the second drug dose output by the network model, and the drug efficacy and the second state parameters of the concentration tank in the previous dosing period are input into the network model to obtain the second supernatant depth output by the network model. The target drug dose added to the concentration tank at the current time is determined according to the difference between the second supernatant depth and the first supernatant depth, the first drug dose, and the second drug dose. Then, a dosing instruction is sent to the dosing execution component to control the dosing execution component to release the target drug dose of the drug. Thus, the accuracy of the drug addition amount is improved.

[0095] To achieve the above-mentioned embodiments, an application embodiment also provides a drug addition amount determination device. Figure 5 A structure diagram of a drug addition amount determination device provided by an application embodiment is shown.

[0096] As shown in Figure 5 The drug addition amount determination device 500 includes:

[0097] The acquisition module 510 is configured to acquire the first state parameters of the concentration tank corresponding to the mixture to be treated at the current time, and the historical drug dose added to the mixture to be treated at each time in the previous dosing period. The first state parameters include the first supernatant depth, the concentration tank inlet flow, and the particle size of the mixture to be treated.

[0098] The first determining module 520 is configured to determine the drug effect of the drug in the concentration tank at the current moment according to the historical drug amounts, and determine the first drug amount added to the concentration tank at the current moment by weighting the drug effect and the first state parameter.

[0099] The second determining module 530 is configured to input the drug effect and the first state parameter into a preset network model, and obtain a second drug amount output by the network model.

[0100] The second determining module 530 is configured to input the drug effect and the second state parameter of the concentration tank in the previous dosing cycle into the network model, and obtain a second supernatant depth output by the network model.

[0101] The third determining module 540 is configured to determine a target drug amount added to the concentration tank at the current moment according to the difference between the second supernatant depth and the first supernatant depth, the first drug amount and the second drug amount.

[0102] In a possible implementation of the embodiment of the present application, the third determining module 540 is configured to:

[0103] query a weight mapping relationship table to determine a first weight corresponding to the first drug amount and a second weight corresponding to the second drug amount according to the difference;

[0104] weight the first drug amount and the second drug amount by using the first weight and the second weight respectively, and determine the target drug amount as a weighted sum of the first drug amount and the second drug amount.

[0105] In a possible implementation of the embodiment of the present application, the third determining module 540 is configured to:

[0106] input the difference into a preset weight function to determine a first weight corresponding to the first drug amount and a second weight corresponding to the second drug amount according to the difference;

[0107] weight the first drug amount and the second drug amount by using the first weight and the second weight respectively, and determine the target drug amount as a weighted sum of the first drug amount and the second drug amount.

[0108] In a possible implementation of the embodiment of the present application, the method further includes:

[0109] The dosing module is configured to send a dosing instruction to the dosing execution assembly to control the dosing execution assembly to release the target drug amount of the drug.

[0110] In a possible implementation of the embodiment of the present application, the dosing execution assembly includes a cation pharmaceutical component and an anion pharmaceutical component, the cation pharmaceutical component is arranged in the buffer tank, and the anion pharmaceutical component is arranged on a pipeline between the buffer tank and the concentration tank.

[0111] In a possible implementation manner of the embodiment of the present application, the first determining module 520, the second determining module 530, and the third determining module 540 can be deployed in the intelligent dosing server; and the dosing module can include a programmable logic controller (PLC). The method is used for automatically controlling the dosing assembly.

[0112] It should be noted that the above explanation and description of the method for determining the dosing amount also applies to the dosing amount determining device of the embodiment, and thus will not be described herein again.

[0113] In the present application, after the first state parameters including the first supernatant depth, the concentration pool inlet flow, the particle size of the to-be-processed mixture, and the historical dosing amount of the dosing agent added to the to-be-processed mixture at each time point in the previous dosing period of the concentration pool corresponding to the to-be-processed mixture at the current time point are acquired, the efficacy of the dosing agent in the concentration pool at the current time point is determined according to the historical dosing amount, and the efficacy and the first state parameters are weighted to determine the first dosing amount of the dosing agent added to the concentration pool at the current time point. Then, the efficacy and the first state parameters are input into a preset network model to acquire the second dosing amount output by the network model, and the efficacy and the second state parameters of the concentration pool in the previous dosing period are input into the network model to acquire the second supernatant depth output by the network model, so as to determine the target dosing amount of the dosing agent added to the concentration pool at the current time point according to the difference between the second supernatant depth and the first supernatant depth, the first dosing amount, and the second dosing amount. In this way, the first dosing amount output by the artificial model and the second dosing amount output by the network model are fused to reduce the influence of the inaccurate first dosing amount output by the network model on the determination of the target dosing amount, thereby improving the accuracy of the dosing amount.

[0114] To implement the above-mentioned embodiments, the present application further provides a computer device, comprising a processor and a memory;

[0115] The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the method for determining the dosing amount of the above-mentioned embodiments.

[0116] To implement the above-mentioned embodiments, the present application further provides a computer readable storage medium, which stores a computer program. The program is executed by the processor to implement the method for determining the dosing amount of the above-mentioned embodiments.

[0117] Although the embodiments of the present application have been shown and described above, it should be understood that the above-mentioned embodiments are exemplary and cannot be understood as limiting the present application. Those skilled in the art can make changes, modifications, replacements, and variations to the above-mentioned embodiments within the scope of the present application.

Claims

1. A method for determining the dosage of a drug, characterized in that, The method comprises the following steps: obtaining a first state parameter of a concentration tank corresponding to a mixture to be treated at a current time from a system, and a historical reagent amount added to the mixture to be treated at each time in a previous reagent adding period, wherein the first state parameter comprises a first supernatant depth, a concentration tank inlet flow rate, and a particle size of the mixture to be treated; determining a reagent effect of the concentration tank at the current time according to each historical reagent amount, and determining a first reagent amount added to the concentration tank at the current time by weighting the reagent effect and the first state parameter; inputting the reagent effect and the first state parameter into a preset network model to obtain a second reagent amount output by the network model; inputting the reagent effect and a second state parameter of the concentration tank in the previous reagent adding period into the network model to obtain a second supernatant depth output by the network model; wherein the second state parameter is obtained from the system when determining the reagent adding amount, and the second state parameter is consistent with the type of the first state parameter; determining a target reagent amount added to the concentration tank at the current time according to a difference between the second supernatant depth and the first supernatant depth, the first reagent amount, and the second reagent amount; wherein when the difference is greater than a threshold value, the first reagent amount is determined as the target reagent amount, and when the difference is less than a threshold value, the second reagent amount is determined as the target reagent amount.

2. The method of claim 1, wherein, The method of determining the target reagent amount added to the concentration tank at the current time according to the difference between the second supernatant depth and the first supernatant depth, the first reagent amount, and the second reagent amount comprises: querying a weight mapping relationship table to determine a first weight corresponding to the first reagent amount and a second weight corresponding to the second reagent amount corresponding to the difference; weighting the first reagent amount and the second reagent amount by using the first weight and the second weight respectively, and determining a weighted sum of the first reagent amount and the second reagent amount as the target reagent amount.

3. The method of claim 1, wherein, The method of determining the target reagent amount added to the concentration tank at the current time according to the difference between the second supernatant depth and the first supernatant depth, the first reagent amount, and the second reagent amount comprises: inputting the difference into a preset weight function to determine a first weight corresponding to the first reagent amount and a second weight corresponding to the second reagent amount corresponding to the difference; weighting the first reagent amount and the second reagent amount by using the first weight and the second weight respectively, and determining a weighted sum of the first reagent amount and the second reagent amount as the target reagent amount.

4. The method of claim 1, wherein, The method further comprises: sending a reagent adding instruction to a reagent adding execution component to control the reagent adding execution component to release the target reagent amount of reagent.

5. The method of claim 4, wherein, The reagent adding execution component comprises a cation reagent preparation component and an anion reagent preparation component, the cation reagent preparation component is arranged in a buffer tank, and the anion reagent preparation component is arranged on a pipeline between the buffer tank and the concentration tank.

6. A medicament additive amount determining apparatus characterized by comprising: The method comprises the following steps: The acquisition module is configured to acquire, from a system, a first state parameter of a concentration tank corresponding to a to-be-processed mixture at a current time and historical dosages of a medicament added to the to-be-processed mixture at each time in a previous medicament adding period, wherein the first state parameter includes a first supernatant depth, a concentration tank inlet flow rate, and a particle size of the to-be-processed mixture; The first determination module is configured to determine a drug effect of the medicament in the concentration tank at the current time according to each of the historical dosages of the medicament, and determine a first dosage of the medicament to be added to the concentration tank at the current time by weighting the drug effect and the first state parameter. The second determination module is configured to input the drug effect and the first state parameter into a preset network model to obtain a second dosage of the medicament output by the network model. The second determination module is configured to input the drug effect and a second state parameter of the concentration tank in the previous medicament adding period into the network model to obtain a second supernatant depth output by the network model; wherein the second state parameter is acquired from the system when determining the dosage of the medicament, and the second state parameter is consistent with the type of the first state parameter. The third determination module is configured to determine a target dosage of the medicament to be added to the concentration tank at the current time according to a difference between the second supernatant depth and the first supernatant depth, the first dosage of the medicament, and the second dosage of the medicament; wherein when the difference is greater than a threshold value, the first dosage of the medicament is determined as the target dosage of the medicament, and when the difference is less than the threshold value, the second dosage of the medicament is determined as the target dosage of the medicament.

7. The apparatus of claim 6, wherein, The third determination module is configured to: query a weight mapping relationship table to determine a first weight corresponding to the first dosage of the medicament and a second weight corresponding to the second dosage of the medicament corresponding to the difference; weight the first dosage of the medicament and the second dosage of the medicament by using the first weight and the second weight respectively, and determine a weighted sum of the first dosage of the medicament and the second dosage of the medicament as the target dosage of the medicament.

8. The apparatus of claim 6, wherein, The third determination module is configured to: input the difference into a preset weight function to determine a first weight corresponding to the first dosage of the medicament and a second weight corresponding to the second dosage of the medicament corresponding to the difference; weight the first dosage of the medicament and the second dosage of the medicament by using the first weight and the second weight respectively, and determine a weighted sum of the first dosage of the medicament and the second dosage of the medicament as the target dosage of the medicament.

9. The apparatus of claim 6, wherein, The medicament adding module is configured to send a medicament adding instruction to a medicament adding execution assembly to control the medicament adding execution assembly to release the target dosage of the medicament. The medicament adding execution assembly includes a cation medicament preparation assembly and an anion medicament preparation assembly, the cation medicament preparation assembly is disposed in a buffer tank, and the anion medicament preparation assembly is disposed on a pipeline between the buffer tank and the concentration tank.

10. The apparatus of claim 9, wherein, ​

Images