Chemical adding control method, device, product and equipment in water treatment process

Abstract

The invention relates to the technical field of water treatment, and provides a chemical adding control method and device in a water treatment process, a product and equipment. The method comprises the steps that characteristic values of dosage prediction characteristics in the water treatment process are obtained, and the dosage prediction characteristics comprise floc characteristics, water inlet parameter characteristics and water outlet parameter characteristics; inputting the characteristic value of each dosage prediction characteristic into a dosage regression prediction model taking the dosage prediction characteristic as single input to obtain a first dosage prediction value corresponding to each dosage prediction characteristic; inputting the characteristic value of each dosage prediction characteristic into a dosage classification prediction model taking the dosage prediction characteristic as single input to obtain a first dosage prediction grade corresponding to each dosage prediction characteristic; and controlling the dosage in the water treatment process based on the first dosage prediction value and the first dosage prediction grade corresponding to each dosage prediction feature. The control accuracy of the dosage can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of water treatment technology, and in particular to a dosing control method, a dosing control device, a computer program product, and an electronic device in a water treatment process. Background Art

[0002] Water treatment is the process of regulating raw water through physical, chemical, or biological means to achieve the required water quality standards for a specific use. For example, raw water can be treated with chemicals to remove impurities and make it suitable for drinking. Chemical addition is a crucial step in the water treatment process, determining the final effluent quality.

[0003] Related technologies can use machine learning models to process influent parameters, floc images, and other data to predict and control dosage. However, these technologies rely on a single model, and the output of this model cannot be verified, resulting in low prediction accuracy.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the Invention

[0005] The purpose of the present disclosure is to provide a method, device, computer program product and electronic equipment for controlling dosing in a water treatment process, thereby improving the accuracy and reliability of dosing control in the water treatment process at least to a certain extent.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by practice of the present disclosure.

[0007] According to a first aspect of the present disclosure, there is provided a chemical dosing control method in a water treatment process, including: obtaining characteristic values of chemical dosing prediction characteristics in the water treatment process, where the chemical dosing prediction characteristics include floc characteristics, influent parameter characteristics, and effluent parameter characteristics; respectively inputting the characteristic values of each chemical dosing prediction characteristic into a preset chemical dosing prediction model, a chemical dosing regression prediction model with the chemical dosing prediction characteristic as a single input, and respectively obtaining first chemical dosing prediction values corresponding to each chemical dosing prediction characteristic according to the output of each chemical dosing regression prediction model; respectively inputting the characteristic values of each chemical dosing prediction characteristic into the preset chemical dosing prediction model, a chemical dosing classification prediction model with the chemical dosing prediction characteristic as a single input, and respectively obtaining first chemical dosing prediction grades corresponding to each chemical dosing prediction characteristic according to the output of each chemical dosing classification prediction; determining a second chemical dosing prediction value based on the first chemical dosing prediction value corresponding to each chemical dosing prediction characteristic, and determining a second chemical dosing prediction grade based on the first chemical dosing grade corresponding to each chemical dosing prediction characteristic; determining a target chemical dosing prediction value according to the overlapping relationship between the second chemical dosing prediction value and the second chemical dosing prediction grade, and controlling the chemical dosing amount in the water treatment process based on the target chemical dosing prediction value.

[0008] According to a second aspect of the present disclosure, there is provided a chemical dosing control device in a water treatment process, including: a characteristic value acquisition module configured to obtain characteristic values of chemical dosing prediction characteristics in the water treatment process, where the chemical dosing prediction characteristics include floc characteristics, influent parameter characteristics, and effluent parameter characteristics; a first prediction module configured to respectively input the characteristic values of each chemical dosing prediction characteristic into a preset chemical dosing prediction model, a chemical dosing regression prediction model with the chemical dosing prediction characteristic as a single input, and respectively obtain first chemical dosing prediction values corresponding to each chemical dosing prediction characteristic according to the output of each chemical dosing regression prediction model; a second prediction module configured to respectively input the characteristic values of each chemical dosing prediction characteristic into the preset chemical dosing prediction model, a chemical dosing classification prediction model with the chemical dosing prediction characteristic as a single input, and respectively obtain first chemical dosing prediction grades corresponding to each chemical dosing prediction characteristic according to the output of each chemical dosing classification prediction; a target prediction module configured to determine a second chemical dosing prediction value based on the first chemical dosing prediction value corresponding to each chemical dosing prediction characteristic, and determine a second chemical dosing prediction grade based on the first chemical dosing grade corresponding to each chemical dosing prediction characteristic; a chemical dosing control module configured to determine a target chemical dosing prediction value according to the overlapping relationship between the second chemical dosing prediction value and the second chemical dosing prediction grade, and control the chemical dosing amount in the water treatment process based on the target chemical dosing prediction value.

[0009] According to a third aspect of the present disclosure, there is provided a computer program product comprising instructions which, when run on a computer, cause the computer to execute the steps of the chemical dosing control method in the water treatment process as described in the first aspect.

[0010] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the chemical dosing control method in the water treatment process as described in the first aspect in the above embodiments.

[0011] According to a fifth aspect of the embodiments of the present disclosure, there is provided an electronic device comprising: a processor; and a storage device for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the chemical dosing control method in the water treatment process as described in the first aspect in the above embodiments.

[0012] As can be seen from the above technical solutions, the chemical dosing control method in the water treatment process, the chemical dosing control device in the water treatment process, and the computer program product and electronic device for implementing the chemical dosing control method in the water treatment process in the exemplary embodiments of the present disclosure at least have the following advantages and positive effects: In the technical solutions provided in some embodiments of the present disclosure, the chemical dosing amount value and the chemical dosing amount level are respectively predicted by multiple chemical dosing amount prediction features, and the final chemical dosing amount prediction value is determined according to the overlapping relationship between the predicted chemical dosing amount value and the chemical dosing amount level, and the chemical dosing amount in the water treatment process is controlled according to the chemical dosing amount prediction value. Compared with the related art, on the one hand, in the present disclosure, the chemical dosing amount level is predicted by a classification prediction model, the specific chemical dosing amount value is predicted by a regression model, and the target chemical dosing amount prediction value is determined by the comprehensive result of the chemical dosing amount level and the specific chemical dosing amount value. In this way, the predicted chemical dosing amount value can be verified based on the predicted chemical dosing amount level, and the reliability of the prediction of the specific chemical dosing amount value can be improved; on the other hand, in the present disclosure, the chemical dosing amount is predicted by a classification prediction model and a regression prediction model respectively based on multiple chemical dosing amount prediction features, and the multiple chemical dosing amount prediction features can adapt to the chemical dosing amount prediction requirements in different scenarios, and the accuracy of the chemical dosing amount prediction can be improved.

[0013] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0015] Figure 1 A schematic flowchart showing a chemical dosing control method in a water treatment process according to an exemplary embodiment of the present disclosure; Figure 2 A schematic diagram showing a water treatment system according to an exemplary embodiment of the present disclosure; Figure 3 A schematic flowchart showing a method for determining a preset chemical dosing prediction model according to an exemplary embodiment of the present disclosure; Figure 4 A schematic flowchart showing a method for determining different chemical dosing levels according to an exemplary embodiment of the present disclosure; Figure 5 A schematic flowchart showing a method for determining a target chemical dosing prediction value according to an overlapping relationship according to an exemplary embodiment of the present disclosure; Figure 6 A schematic flowchart showing a method for chemical dosing control according to a target chemical dosing prediction value according to an exemplary embodiment of the present disclosure; Figure 7 A schematic diagram showing the composition of a chemical dosing control device in a water treatment process according to an exemplary embodiment of the present disclosure; Figure 8 A schematic diagram showing the structure of an electronic device according to an exemplary embodiment of the present disclosure. Detailed implementation manners

[0016] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art. The features, structures, or characteristics described can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art will realize that the technical solutions of the present disclosure can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. can be used. In other cases, well-known technical solutions are not shown or described in detail to avoid obscuring the various aspects of the present disclosure.

[0017] As used in this specification, the terms "a", "an", "the", and "said" are used to indicate the presence of one or more elements / components / etc.; the terms "comprising" and "having" are used to mean an open inclusion and mean that there may be additional elements / components / etc. in addition to the listed elements / components / etc.; the terms "first", "second", etc. are used only as labels and are not a limitation on the quantity of their objects.

[0018] In addition, the drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0019] In the related art, although the chemical dosage prediction in the water treatment process can be performed by a machine learning model, it performs the chemical dosage prediction through a single type of model. Even if the final prediction result is determined through multiple models, since the types of the multiple models are the same, the mutual verification of the prediction results cannot be performed, resulting in the reliability and accuracy of the prediction results being difficult to guarantee.

[0020] Based on this, the present disclosure provides a chemical dosage control method in the water treatment process to at least solve one or more of the above problems in the related art to a certain extent.

[0021] Exemplarily, Figure 1 shows a schematic flowchart of a chemical dosage control method in an exemplary embodiment of the present disclosure. Refer to Figure 1 , the method may include: Step S110, obtaining the characteristic values of the chemical dosage prediction characteristics in the water treatment process, where the chemical dosage prediction characteristics include floc characteristics, influent parameter characteristics, and effluent parameter characteristics; Step S120, respectively inputting the characteristic values of each chemical dosage prediction characteristic into a preset chemical dosage prediction model, a chemical dosage regression prediction model with the chemical dosage prediction characteristic as a single input, and respectively obtaining the first chemical dosage prediction value corresponding to each chemical dosage prediction characteristic according to the output of each chemical dosage regression prediction model; Step S130, respectively inputting the characteristic values of each chemical dosage prediction characteristic into the preset chemical dosage prediction model, a chemical dosage classification prediction model with the chemical dosage prediction characteristic as a single input, and respectively obtaining the first chemical dosage prediction level corresponding to each chemical dosage prediction characteristic according to the output of each chemical dosage classification prediction; Step S140: Determine a second chemical dosage prediction value based on the first chemical dosage prediction value corresponding to each chemical dosage prediction feature, and determine a second chemical dosage prediction level based on the first chemical dosage level corresponding to each chemical dosage prediction feature; Step S150: Determine a target chemical dosage prediction value according to the overlapping relationship between the second chemical dosage prediction value and the second chemical dosage prediction level, and control the chemical dosage in the water treatment process based on the target chemical dosage prediction value.

[0022] In Figure 1 In the technical solution provided by the embodiment shown, the chemical dosage value and the chemical dosage level are predicted respectively through multiple chemical dosage prediction features. The final chemical dosage prediction value is determined according to the overlapping relationship between the predicted chemical dosage value and the chemical dosage level, and the chemical dosage in the water treatment process is controlled according to the chemical dosage prediction value. Compared with the related art, on the one hand, the chemical dosage level is predicted through a classification prediction model, the specific chemical dosage value is predicted through a regression model, and the target chemical dosage prediction value is determined according to the comprehensive result of the chemical dosage level and the specific chemical dosage value. In this way, the predicted chemical dosage value can be verified based on the predicted chemical dosage level, and the reliability of the prediction of the specific chemical dosage value can be improved; on the other hand, the chemical dosage is predicted through a classification prediction model and a regression prediction model respectively based on multiple chemical dosage prediction features. The multiple chemical dosage prediction features can meet the chemical dosage prediction requirements in different scenarios and improve the accuracy of the chemical dosage prediction.

[0023] Next, the specific implementation of "Step S110: Obtain the feature value of the chemical dosage prediction feature in the water treatment process" will be described in detail first.

[0024] To more clearly illustrate the water treatment process of the present disclosure, first combine Figure 2 for illustration. Exemplarily, Figure 2 shows a schematic diagram of a water treatment system in an exemplary embodiment of the present disclosure. Refer to Figure 2, the water treatment system may include an inlet tank 21, a flocculation tank 22, a sedimentation tank 23, a filtration tank 24, and an outlet tank 25. The inlet tank 21 can regulate the water volume of the incoming water and play a certain buffering role in the sudden change of the raw water quality. The flocculation tank 22 is used to add flocculants, such as polyaluminum chloride, etc., to the water to destabilize the fine suspended particles, colloids and other impurities in the water and make them aggregate with each other to form larger flocs for subsequent sedimentation separation. The sedimentation tank 23 makes the large flocs formed in the flocculation tank settle to the bottom of the tank by gravity, realizes solid-liquid separation, and removes most of the suspended solids, sediment and other impurities in the water. The filtration tank 24 further removes the remaining fine suspended particles, colloids, bacteria and other impurities in the water through filtration media, such as quartz sand, activated carbon, etc., improves the clarity and transparency of the water, and makes the water quality reach a higher standard. The outlet tank 25 is used to store the treated clear water and provide a stable water supply guarantee for the water use link. At the same time, the water quality of the treated water can also be monitored in the outlet tank 25, such as detecting indexes such as pH, turbidity, suspended solids, etc., to ensure that the outlet water quality meets the relevant standards and usage requirements.

[0025] Exemplarily, water inlet parameter detection devices, such as a water inlet pH detection device, a water inlet temperature detection device, a water inlet turbidity detection device, and a water inlet flow detection device, can be arranged between the inlet tank 21 and the flocculation tank 22. An underwater camera can be arranged in the flocculation tank 22 to capture floc images and obtain floc characteristics. An underwater camera can also be arranged in the sedimentation tank 23 to capture floc sedimentation videos. A chemical dosing pump can be arranged between the inlet tank 21 and the flocculation tank 22, and the flocculant can be added to the water through the chemical dosing pump.

[0026] In an exemplary embodiment, the chemical dosing amount prediction feature includes floc characteristics, water inlet parameter characteristics, and water outlet parameter characteristics. The chemical dosing amount prediction feature can be understood as being associated with the chemical dosing amount and can be used to reflect or affect the chemical dosing amount.

[0027] In an exemplary embodiment, the floc characteristics can be determined by the floc images captured by the camera. For example, an underwater camera is arranged in the flocculation tank, and the underwater camera can be installed on the inner wall of the flocculation tank. The floc images of the water in the flocculation tank can be captured through the underwater camera. The camera can send the captured floc images to the server. After receiving the floc images, the server performs feature extraction processing on the floc images to obtain the floc characteristics. Among them, the floc characteristics can include floc area distribution characteristics, the number of flocs per unit area, the average particle size of flocs, etc.

[0028] For example, a floc recognition model can be pre-trained to extract floc features through the floc recognition model. For example, a large number of historical floc images can be collected in advance, and then the historical floc images are manually annotated to mark the contour or boundary of each floc in the floc image, the floc area and floc particle size corresponding to the contour or boundary, and the number of flocs in the floc image as training data.

[0029] A machine learning model is trained with the training data to enable the machine learning model to recognize the contour or boundary of the flocs in the image, as well as the area of each floc and count the number of flocs in the image according to the recognition situation. It is also possible to fine-tune an existing object detection model with the training data to train the object detection model to detect and recognize flocs, thereby obtaining a floc recognition model. Through the floc recognition model, flocs can be accurately recognized in an image containing flocs, and the area and floc particle size of the recognized flocs, as well as the number of flocs in the image, can be determined. Then, the recognition results of the floc recognition model are analyzed and processed to obtain floc features. For example, the area distribution of the recognized flocs is statistically analyzed (such as the proportion of the number of different areas, etc.), the average of all floc particle sizes is obtained to get the average floc particle size, and the number of flocs per unit area feature is determined according to the number of flocs in the image and the actual scene area indicated by the image. Of course, the floc features can include other features, such as floc morphology, edge sharpness, etc., and this exemplary embodiment does not make special limitations on this.

[0030] In an exemplary embodiment, the influent parameter features include influent pH value, influent flow rate, influent turbidity, and influent temperature. As described above, influent parameter detection devices such as an influent pH value detection device, an influent flow rate detection device, an influent turbidity detection device, and an influent temperature detection device can be installed between the influent tank and the flocculation tank. These influent parameter detection devices can send the detected influent parameters to the server, and the server can obtain the influent parameter features based on the data sent by these influent parameter detection devices.

[0031] In an exemplary embodiment, the effluent parameter features include effluent turbidity, effluent pH, and effluent suspended solid content. For example, effluent parameter detection devices such as an effluent turbidity detection device, an effluent pH detection device, and an effluent suspended solid content detection device can be configured in the effluent tank to detect the effluent parameter features through these effluent parameter detection devices, and these effluent parameter detection devices can send the detected effluent parameter features to the server.

[0032] After the server obtains the feature values of the chemical dosage prediction features, it can determine the target chemical dosage prediction value according to the method of subsequent step S120 to step S150, and then send a control instruction to the chemical dosing pump according to the target chemical dosage prediction value to control the dosage of the chemical dosing pump.

[0033] In an exemplary embodiment, the water inlet parameters can be detected at regular intervals, and the most recently detected water inlet parameters are compared with the currently detected water inlet parameters. When the change in any water inlet parameter exceeds the corresponding preset threshold, the camera is started to capture the floc image, and the current water inlet parameter characteristics, the captured floc image, and the water outlet parameter characteristics are sent to the server, so that the server performs the chemical dosing prediction control in the manner of the present disclosure.

[0034] In other words, in the present disclosure, the water inlet parameters can be collected at regular intervals, and then it is determined whether to re-perform the chemical dosing prediction according to the fluctuation of the water inlet parameters. For example, when the water inlet parameters change greatly, the chemical dosing is re-predicted and controlled. When the water inlet parameters do not change or the change in the water inlet parameters is not significant compared with the previous water inlet parameters, the chemical dosing corresponding to the previous water inlet parameters can be directly used for chemical dosing control. In this way, it is not necessary to re-perform the chemical dosing prediction at each sampling moment, which can improve the chemical dosing control efficiency and save computing resources at the same time.

[0035] Of course, the water inlet parameters, floc images, and water outlet parameters can also be collected at regular intervals, and then the relevant data is sent to the server to perform the chemical dosing prediction control at regular intervals. That is, without determining the fluctuation of the water inlet parameters, the chemical dosing prediction control is directly re-performed regularly each time. This exemplary embodiment does not make special limitations on this.

[0036] In an exemplary embodiment, the characteristic value of the chemical dosing prediction feature can be understood as the specific value of each feature determined at the sampling moment, such as the water inlet temperature value, the water inlet pH value, etc.

[0037] Next, a detailed description is given of the specific implementation manner of "Step S120, respectively inputting the characteristic values of each chemical dosing prediction feature into a preset chemical dosing prediction model, a chemical dosing regression prediction model with the chemical dosing prediction feature as a single input, and respectively obtaining the first chemical dosing prediction value corresponding to each chemical dosing prediction feature according to the output of each chemical dosing regression prediction model".

[0038] Exemplarily, Figure 3 The flowchart showing a method for determining a preset chemical dosing prediction model in an exemplary embodiment of the present disclosure is shown. Refer to Figure 3 , this method may include Step S310 to Step S370. Among them: In Step S310, respectively generate a first label data set corresponding to each chemical dosing prediction feature according to the historical data corresponding to each chemical dosing prediction feature and the historical chemical dosing value corresponding to the historical data.

[0039] For example, the eigenvalue of the historical influent parameter characteristics and the correct chemical dosage value corresponding to this eigenvalue can be collected, so as to obtain the first label dataset corresponding to the influent parameter characteristics. The historical floc images are collected, the floc characteristics are extracted from the historical floc images, and according to the extracted floc characteristics and the correct chemical dosage value corresponding to the historical floc images, the first label dataset corresponding to the floc characteristics is obtained. The eigenvalue of the historical effluent parameter characteristics and the correct chemical dosage value corresponding to this eigenvalue are collected, so as to generate the first label dataset corresponding to the effluent parameter characteristics. In other words, one chemical dosage prediction feature corresponds to one label dataset, that is, the number of types of chemical dosage prediction features corresponds to the number of the first label datasets.

[0040] Among them, the correct chemical dosage value can be understood as the chemical dosage that can make the effluent water quality reach the preset standard under this eigenvalue.

[0041] In an exemplary embodiment, the historical data of the chemical dosage prediction feature can be collected. If the chemical dosage corresponding to this historical data is not the correct chemical dosage value, it can be excluded, that is, the collected historical data is filtered according to whether it corresponds to a normal chemical dosage value, so as to obtain the first label dataset corresponding to each chemical dosage prediction feature.

[0042] In step S320, according to the historical data corresponding to each chemical dosage prediction feature and the chemical dosage level to which the historical chemical dosage value corresponding to the historical data belongs among different chemical dosage levels corresponding to the chemical dosage prediction feature, the second label dataset corresponding to each chemical dosage prediction feature is generated.

[0043] In an exemplary embodiment, different chemical dosage levels corresponding to each chemical dosage prediction feature can be determined in advance. Different chemical dosage levels correspond to different chemical dosage ranges.

[0044] For example, the number of levels of the chemical dosage levels corresponding to different chemical dosage prediction features is the same, and the levels representing the same chemical dosage degree have the same level identifier. For example, the chemical dosage levels corresponding to each chemical dosage prediction feature are all divided into 3 levels: low, medium, and high. However, the chemical dosage ranges indicated by the same level of chemical dosage levels corresponding to different chemical dosage prediction features may be the same or different. For example, the chemical dosage range indicated by the low level of the influent parameter characteristics may be [a1, b1], while the chemical dosage range indicated by the low level of the floc characteristics may be [a2, b2]. Among them, a1, a2, b1, and b2 are all different values.

[0045] Exemplarily, the determination method of different chemical dosage levels corresponding to any chemical dosage prediction feature can refer to Figure 4 . Such as Figure 4As shown, the method may include steps S410 to S420. Among them: In step S410, historical data of the chemical dosage prediction feature is collected, and the collected historical data is clustered.

[0046] For example, for each chemical dosage prediction feature, historical data of this chemical dosage prediction feature can be collected. Similarly, the historical data can also be filtered according to whether there is a correct chemical dosage value, so as to obtain the historical data of this chemical dosage prediction feature. For example, the influent parameter feature corresponds to the first historical data, the floc feature corresponds to the second historical data, and the effluent parameter feature corresponds to the third historical data. Then, the first clustering is performed on the first historical data to obtain a first clustering result, the second clustering is performed on the second historical data to obtain a second clustering result, and the third clustering is performed on the third historical data to obtain a third clustering result.

[0047] Among them, the number of clustering categories of each chemical dosage prediction feature is the same, that is, the K value in the clustering algorithm is the same, that is, the K values of the first clustering, the second clustering, and the third clustering are the same, such as all being 3. The specific K value can be determined according to experience or test results, and this exemplary embodiment does not make special limitations on this. For example, different K values are taken respectively, and the clustering results of different K values are tested, and the K value with the best test effect is selected as the final K value.

[0048] In step S420, different chemical dosage prediction levels are determined according to the chemical dosages indicated by the historical data in each clustering category in the clustering result.

[0049] Exemplarily, the magnitude relationship of the chemical dosage degrees between different chemical dosage levels corresponding to different clustering categories can be determined according to the magnitude relationship of the means of the chemical dosages indicated by the historical data in each clustering category in the clustering result; the chemical dosage range indicated by the chemical dosage level corresponding to the clustering category is determined according to the minimum value and the maximum value of the chemical dosages indicated by the historical data in the clustering category.

[0050] Taking the K value as 3 as an example, each clustering category corresponds to a level, and there are a total of 3 levels. Among them, the average value of the correct dosing amount corresponding to the historical data in clustering category 1 is c1, the average value of the correct dosing amount corresponding to the historical data in clustering category 2 is c2, and the average value of the correct dosing amount corresponding to the historical data in clustering category 3 is c3. The magnitude relationship of c1, c2, and c3 is c1 < c2 < c3. Then, clustering category 1 corresponds to the first dosing level, clustering category 2 corresponds to the second dosing level, and clustering category 3 corresponds to the third dosing level. The dosing degree of the first dosing level is less than that of the second dosing level, and the dosing degree of the second dosing level is less than that of the third dosing level. For example, the first dosing level is the low dosing level, the second dosing level is the medium dosing level, and the third dosing level is the high dosing level.

[0051] For example, according to the minimum and maximum values of the dosing amounts indicated by the historical data in the clustering category corresponding to each dosing level, the dosing range indicated by the dosing level can be determined. For example, there are 100 historical data in a certain dosing level, and these 100 historical data correspond to 100 dosing values. The interval range composed of the minimum and maximum values of these 100 dosing values is the dosing range indicated by the dosing level.

[0052] Taking the above-mentioned first dosing level, second dosing level, and third dosing level as examples, there may be a situation where there is an overlapping interval between the dosing ranges indicated by different dosing levels corresponding to the dosing levels determined according to the clustering results. At this time, manual verification and adjustment can be performed to make the dosing ranges of adjacent dosing levels continuous and non-overlapping. It is also possible to use the dosing range corresponding to the dosing level obtained according to the clustering result as the initial dosing range, and adjust the initial dosing range according to the first preset rule to obtain the final dosing range corresponding to the dosing level.

[0053] Exemplarily, the first preset rule may include: in the case where there is an overlapping interval between the dosing ranges indicated by adjacent dosing levels, taking the overlapping interval as a buffer interval, that is, the dosing amount in the buffer interval belongs to both adjacent dosing levels at the same time, and the dosing amount in the non-overlapping part belongs to their respective dosing levels. That is, the dosing range indicated by each dosing level is composed of a non-overlapping interval plus a buffer interval. In this way, when performing data annotation, the dosing level of the historical data with the dosing amount falling in the buffer interval has two labels. Subsequently, when making predictions, the final output result is also selected through the category confidence. For example, when the confidence is greater than 0.9, it is the finally predicted category. In this way, the problem of low prediction accuracy caused by unreasonable division of the interval range of the dosing level can be avoided to the greatest extent.

[0054] Exemplarily, the first preset rule may also include: evenly dividing the overlapping interval and allocating it to the dosing ranges indicated by adjacent dosing amount levels, that is, dividing the overlapping interval into two at the median of the overlapping interval. The final dosing range corresponding to the lower level in adjacent dosing amount levels is the minimum value in its corresponding initial dosing range to the median of the overlapping interval, and the final dosing range corresponding to the higher level in adjacent dosing amount levels is the median of the overlapping interval to the maximum value in its corresponding initial dosing range.

[0055] Similarly, for the case where the initial dosing ranges corresponding to adjacent dosing amount levels are discontinuous, it can also be adjusted by the second preset rule to obtain the final dosing range.

[0056] Exemplarily, the second preset rule may include: regarding the range interval between the initial dosing ranges corresponding to adjacent dosing amount levels as a buffer interval, and the buffer interval belongs to these adjacent dosing amount levels at the same time. Similarly, in this way, when performing data annotation, there will be two labels for the dosing amount level of historical data with dosing amount values falling within the buffer interval.

[0057] Exemplarily, the second preset rule may also include: regarding the interval between the range intervals between the initial dosing ranges corresponding to adjacent dosing amount levels as a buffer interval, and dividing the buffer interval into two at the median of the buffer interval and allocating it to these adjacent dosing amount levels. That is, the final dosing range corresponding to the lower level in adjacent dosing amount levels is the minimum value in its corresponding initial dosing range to the median of the buffer interval, and the final dosing range corresponding to the higher level in adjacent dosing amount levels is the median of the buffer interval to the maximum value in the initial dosing range corresponding to this dosing amount level.

[0058] In an exemplary implementation manner, if there is an overlapping interval in the dosing ranges indicated by non - adjacent dosing amount levels, such as an overlapping interval between the dosing ranges corresponding to the above - mentioned low dosing amount level and high dosing amount level, then after adjusting the K value, clustering is performed again, and the dosing amount level is determined again according to the clustering result. It is also possible to adjust the dosing ranges indicated by non - adjacent dosing amount levels based on the dosing ranges indicated by the intermediate dosing amount levels between non - adjacent dosing amount levels, so as to ensure that there is no overlapping interval in the dosing ranges indicated by non - adjacent dosing amount levels. For example, taking the minimum value of the intermediate dosing amount level as the upper limit value of the lower dosing amount level among non - adjacent dosing amount levels, and taking the maximum value of the intermediate dosing amount level as the lower limit value of the higher dosing amount level among non - adjacent dosing amount levels.

[0059] Adjusting the initial chemical dosing amount range based on the above method can ensure the rationality of the initial chemical dosing amount range, and ensure that the initial chemical dosing amount range is continuous without any gaps. In this way, the target chemical dosing amount prediction value can be determined according to the overlapping relationship between the second chemical dosing amount prediction value and the second chemical dosing amount prediction level in the subsequent steps.

[0060] Through the above steps S410 to S420, the chemical dosing amount ranges corresponding to different chemical dosing amount levels indicated by different chemical dosing amount prediction features can be automatically determined based on historical data. Compared with the method of manually determining the chemical dosing amount level, this method of automatically determining the chemical dosing amount level by clustering not only improves the efficiency of determining the chemical dosing amount level, but also provides a basis for determining the chemical dosing amount level, making the determined chemical dosing amount level highly interpretable and reliable.

[0061] For example, after obtaining the chemical dosing amount ranges corresponding to different chemical dosing amount levels indicated by each chemical dosing amount prediction feature, the first label data set corresponding to each chemical dosing amount prediction feature can be copied to generate a first label copy data set. For the first label copy data set corresponding to any chemical dosing amount prediction feature, the correct chemical dosing amount value label in this first label copy data set is modified to the specific chemical dosing amount level to which the correct chemical dosing amount value belongs in the different chemical dosing amount levels corresponding to this chemical dosing amount prediction feature, so as to generate a second label data set.

[0062] Taking the influent parameter feature in the chemical dosing amount prediction feature as an example, the first label data set A corresponding to the influent parameter feature is copied to generate the first label copy data set A1 corresponding to the influent parameter feature. According to the chemical dosing amount ranges indicated by different chemical dosing amount levels corresponding to the influent parameter feature, it is determined which range in the chemical dosing amount ranges indicated by different chemical dosing amount levels corresponding to the influent parameter feature the label of the training data in A1, that is, the specific correct chemical dosing amount value, belongs to. Thus, it is obtained which level in the different chemical dosing amount levels corresponding to the influent parameter feature the training data in A1 belongs to, and this level is used as the new training label, so as to generate a second label data set.

[0063] In other words, the training data in the first label data set and the second label data set can be the same, but their labels are different. The label of each training data in the first label data set is a specific chemical dosing amount value, and the label of each training data in the second label data set is a chemical dosing amount level.

[0064] In step S330, a first number of candidate regression prediction models are respectively trained based on the first label data set corresponding to each chemical dosing amount prediction feature to obtain a second number of chemical dosing regression prediction models.

[0065] In an exemplary embodiment, the first quantity is an integer greater than 1. The candidate regression prediction model can include any machine learning model capable of performing regression prediction, such as a neural network model, a regression model based on support vector machines, etc. This exemplary embodiment does not make special limitations in this regard.

[0066] Taking the first quantity as an example of 3, three candidate regression prediction models are trained based on the first labeled dataset corresponding to the influent parameter features, and three chemical dosing regression prediction models can be obtained according to the training results. Three candidate regression prediction models are trained based on the first labeled dataset corresponding to the floc characteristics, and three chemical dosing regression prediction models can be obtained according to the training results. Three candidate regression models are trained based on the first labeled dataset corresponding to the effluent parameter features, and three chemical dosing regression prediction models can be obtained according to the training results. That is, a total of nine chemical dosing regression prediction models can be obtained. That is, the second quantity and the first quantity are in a multiple relationship, and the second quantity is three times the first quantity.

[0067] In step S340, a third quantity of candidate classification prediction models are respectively trained based on the second labeled dataset corresponding to each chemical dosing prediction feature, and a fourth quantity of chemical dosing classification prediction models are obtained.

[0068] In an exemplary embodiment, the third quantity is an integer greater than 1. The candidate classification prediction model can include any machine learning model capable of performing classification prediction, such as a neural network classification model, a decision tree classification model, etc. This exemplary embodiment does not make special limitations in this regard.

[0069] Taking the third quantity as 4 as an example, similarly, four chemical dosing classification prediction models can be obtained respectively based on each chemical dosing prediction feature, and a total of twelve chemical dosing classification prediction models can be obtained. That is, the fourth quantity and the third quantity are also in a multiple relationship, and the fourth quantity is also three times the third quantity.

[0070] In step S350, model testing is performed on the second quantity of chemical dosing regression prediction models, and a target chemical dosing regression prediction model is determined from the second quantity of chemical dosing regression prediction models according to the test results.

[0071] For example, the second quantity of chemical dosing prediction models can be tested through a test dataset, such as testing the prediction accuracy of the models, and the chemical dosing regression prediction models with a prediction accuracy greater than a preset threshold are determined as the target chemical dosing regression prediction models. Among them, the preset threshold can be determined according to requirements or experience, and this exemplary embodiment does not make special limitations in this regard.

[0072] For each dosing prediction model corresponding to a dosing amount prediction feature, the top N dosing prediction models ranked by test accuracy can also be selected as candidate target dosing prediction models, and the target dosing prediction model can be obtained according to the collection of candidate target dosing prediction models corresponding to each dosing amount prediction feature. Among them, N is greater than or equal to 1 and less than the first quantity.

[0073] In step S360, model testing is performed on the fourth quantity of dosing classification prediction models, and a target dosing classification prediction model is determined from the fourth quantity of dosing classification prediction models according to the test results.

[0074] Exemplarily, the specific implementation manner of step S360 can refer to step S350, and details are not described herein again.

[0075] In step S370, the preset dosing amount prediction model is determined according to the target dosing regression prediction model and the target dosing classification prediction model.

[0076] Exemplarily, the preset dosing amount prediction model can be obtained according to the collection of the target dosing regression prediction model and the target dosing classification prediction model. In other words, the preset dosing amount prediction model can be composed of multiple target dosing regression prediction models and multiple target dosing classification prediction models.

[0077] Through the above steps S310 to S370, in the process of generating the preset dosing amount prediction model, data annotation and training are respectively performed based on various dosing amount prediction features, so as to obtain multiple target dosing regression prediction models and multiple target dosing classification prediction models. The multiple models can adapt to various different water quality environments, thereby ensuring the accuracy and reliability of dosing amount prediction under different water quality environments.

[0078] Exemplarily, as described above, each model in the preset dosing amount prediction model takes a single dosing amount prediction feature as input. Therefore, for each dosing amount prediction feature, the dosing amount prediction feature can be input into the dosing regression prediction model in the preset dosing amount prediction model that takes the dosing amount prediction feature as a single input, and the first dosing amount prediction value corresponding to the dosing amount prediction feature can be obtained according to the output of the dosing regression prediction model.

[0079] Among them, the number of first chemical dosage prediction values that each chemical dosage prediction feature can predict is determined according to the number of chemical dosage regression prediction models in the preset chemical dosage prediction model that use this chemical dosage prediction feature as a single input. For example, if the influent parameter feature corresponds to 2 chemical dosage regression prediction models, then 2 first chemical dosage prediction values can be predicted through the influent parameter feature. In other words, the number of first chemical dosage prediction values that each chemical dosage prediction feature can predict is the same as the number of target chemical dosage regression prediction models corresponding to this chemical dosage prediction feature in the preset chemical dosage regression prediction model.

[0080] Next, a detailed description will be given of the specific implementation manner of "step S130: respectively input the feature values of each chemical dosage prediction feature into the chemical dosage classification prediction model in the preset chemical dosage prediction model that uses this chemical dosage prediction feature as a single input, and respectively obtain the first chemical dosage prediction level corresponding to each chemical dosage prediction feature according to the output of each chemical dosage classification prediction".

[0081] Exemplarily, similarly, for each chemical dosage prediction feature, this chemical dosage prediction feature can be input into the chemical dosage classification prediction model in the preset chemical dosage prediction model that uses this chemical dosage prediction feature as a single input, and the first chemical dosage prediction level corresponding to this chemical dosage prediction feature is obtained according to the output of this chemical dosage classification prediction model.

[0082] Exemplarily, other specific implementation manners of step S130 can refer to the relevant content in step S120, and will not be elaborated here.

[0083] Next, a detailed description will be given of the specific implementation manner of "step S140: determine the second chemical dosage prediction value based on the first chemical dosage prediction value corresponding to each chemical dosage prediction feature, and determine the second chemical dosage prediction level based on the first chemical dosage level corresponding to each chemical dosage prediction feature".

[0084] Exemplarily, the specific implementation manner of step S140 may include: obtaining the second chemical dosage prediction value according to the mean value of the first chemical dosage prediction values corresponding to each chemical dosage prediction feature, and obtaining the second chemical dosage prediction level according to the first chemical dosage level with the largest quantity among the first chemical dosage levels corresponding to each chemical dosage prediction feature.

[0085] For example, all the first chemical dosage prediction values obtained in step S120 can be averaged to obtain the second chemical dosage prediction value, and the second chemical dosage prediction level can be obtained according to the first chemical dosage level with the highest frequency of occurrence among the first chemical dosage levels in step S130.

[0086] Next, a detailed description will be given of the specific implementation manner of "Step S150: Determine the target chemical addition amount prediction value according to the overlapping relationship between the second chemical addition amount prediction value and the second chemical addition amount prediction level, and control the chemical addition amount in the water treatment process based on the target chemical addition amount prediction value".

[0087] Exemplarily, the overlapping relationship between the second chemical addition amount prediction value and the second chemical addition amount prediction level may include that the second chemical addition amount prediction value belongs to the chemical addition amount range indicated by the second chemical addition amount prediction level or the second chemical addition amount prediction value does not belong to the chemical addition amount range indicated by the second chemical addition amount prediction level.

[0088] Taking the medium chemical addition amount level as the second chemical addition amount prediction level as an example, the chemical addition amount ranges indicated by the medium chemical addition amount level corresponding to each chemical addition amount prediction feature may be different. In an exemplary implementation manner, as long as the second chemical addition amount prediction value falls within the chemical addition amount range indicated by the medium chemical addition amount level corresponding to any target chemical addition amount prediction feature, it is considered that the second chemical addition amount prediction value belongs to the chemical addition amount range indicated by the medium chemical addition amount level. If the second chemical addition amount prediction value does not fall within the chemical addition amount range indicated by the medium chemical addition amount level corresponding to any target chemical addition amount prediction feature, it is considered that the second chemical addition amount prediction value does not belong to the chemical addition amount range indicated by the medium chemical addition amount level. Among them, the target chemical addition amount prediction feature includes the chemical addition amount prediction feature whose corresponding first chemical addition amount prediction level and the finally determined second chemical addition amount prediction level are the same. For example, there are 6 models in the target chemical addition classification prediction model. Among these 6 models, the prediction results of 4 models are the medium chemical addition amount level, and the prediction results of 2 models are the low chemical addition amount level, that is, the second predicted chemical addition amount level is the medium chemical addition amount level. Then, the chemical addition amount prediction features input by the 4 target chemical addition classification prediction models with the prediction result of the medium chemical addition amount level are the target chemical addition amount prediction features.

[0089] Next, in combination with Figures 5 to 6 a detailed description will be given of the specific implementation manner of Step S150.

[0090] Exemplarily, Figure 5 FIG. shows a flowchart of a method for determining a target chemical addition amount prediction value according to an overlapping relationship in an exemplary embodiment of the present disclosure. Referring to Figure 5 , this method may include Step S510 to Step S540.

[0091] In Step S510, it is judged whether the second chemical addition amount prediction value is within the first chemical addition amount range indicated by the second chemical addition amount prediction level. If so, go to Step S520; otherwise, go to Step S530.

[0092] In an exemplary embodiment, the database pre-stores the dosing ranges indicated by each dosing level corresponding to different dosing prediction features. For the sake of distinct expression, the dosing ranges stored in the database are the first dosing ranges, and the dosing ranges obtained after adjusting the dosing ranges in the database are the second dosing ranges.

[0093] In step S520, the second dosing prediction value is determined as the target dosing prediction value.

[0094] For example, as described above, as long as the second dosing prediction value falls within the first dosing range indicated by the second dosing prediction level corresponding to any target dosing prediction feature, it is considered that the second dosing prediction value is within the first dosing range indicated by the second dosing prediction level. When the second dosing prediction value is within the first dosing range indicated by the second dosing prediction level, it indicates that the second dosing prediction value is consistent with the prediction result of the second dosing prediction level, that is, the dosing values predicted by different types of models are consistent, which means that the second dosing prediction value is reliable. Therefore, the second dosing prediction value can be directly determined as the target dosing prediction value.

[0095] In step S530, the water flow rate and water turbidity detected at multiple detection points during the process from the raw water point to the inlet point are obtained. When the water flow rate is greater than the first preset value and / or the water turbidity is greater than the second preset value, the first dosing range indicated by the second dosing prediction level is adjusted according to historical experience to obtain the second dosing range.

[0096] Taking the medium dosing level as the above-mentioned second dosing prediction level as an example, if the second dosing prediction value does not fall within the first dosing range indicated by the medium dosing level corresponding to any target dosing prediction value, it is considered that the second dosing prediction value is not within the first dosing range indicated by the second dosing prediction level. Under normal circumstances, the prediction results of different types of models should be consistent. When the prediction results of different types of models are inconsistent, it is necessary to re-evaluate the prediction results of the models.

[0097] For example, in the present disclosure, the predicted chemical dosage value is constrained and verified by the chemical dosage prediction level, so as to ensure the reliability and accuracy of the determined target predicted chemical dosage value. Therefore, the division of the chemical dosage level is crucial for the prediction result. When the prediction results of the classification prediction model and the regression prediction model are inconsistent, it is possible to first consider whether there is a sudden change in the current water quality, such as a sudden change in the influent turbidity, and the chemical dosage ranges of each previously determined chemical dosage level may not cover the current water quality mutation situation, resulting in the problem that the prediction results of the classification prediction model and the regression prediction model are inconsistent. Therefore, in the present disclosure, when the second predicted chemical dosage value does not belong to the first chemical dosage range indicated by the second chemical dosage prediction level of any target predicted chemical dosage feature, the first chemical dosage range indicated by the previously determined second chemical dosage prediction level can be dynamically adjusted to meet as many complex emergencies as possible and improve the accuracy of the chemical dosage prediction.

[0098] In an exemplary embodiment, a plurality of detection points may be set between the raw water point and the influent point to determine whether there is a water quality mutation due to sudden weather changes or other emergencies. Among them, the raw water point refers to the source point of the water to be treated in the water treatment system. For example, if the water treatment system treats the water in a reservoir, then the reservoir is the raw water point, and the influent point refers to the inlet of the influent tank in the water treatment system.

[0099] For example, a plurality of detection points may be intermittently set between the raw water point and the influent point, and each detection point is equipped with a turbidimeter and a flowmeter. Whether there is a sudden change in turbidity or flow can be detected by the turbidimeter and the flowmeter. When the flow detected by the flowmeter is greater than the first preset value and / or the turbidity detected by the turbidimeter is greater than the second preset value, it is determined that there is a water quality mutation situation, and then the first chemical dosage range indicated by the second chemical dosage prediction level can be adjusted.

[0100] Among them, the first preset value and the second preset value can be determined according to requirements, and this exemplary embodiment does not make special limitations on this. For example, the first preset value can be the maximum value of the influent flow rate of all influent parameter features in the training data multiplied by a coefficient greater than 1.

[0101] For example, the first chemical dosage range indicated by the second chemical dosage prediction level corresponding to the target predicted chemical dosage feature can be adjusted according to historical experience. Taking the target predicted chemical dosage feature including influent parameter features as an example, the similarity between the current influent parameter feature value and the historical influent parameter feature values stored in the database can be calculated, and multiple historical influent parameter feature values similar to the current influent parameter feature value are selected. The interval corresponding to the minimum and maximum historical chemical dosage values of the multiple historical influent parameter feature values is used as the second chemical dosage range indicated by the second chemical dosage prediction level corresponding to the influent parameter feature.

[0102] In step S540, according to the overlapping relationship between the second chemical dosage range and the second chemical dosage prediction value, the target chemical dosage prediction value is determined.

[0103] Exemplarily, an implementation manner of step S540 may include: when the second chemical dosage prediction value is within the second chemical dosage range, determining the second chemical dosage prediction value as the target chemical dosage prediction value; when the second chemical dosage prediction value is not within the second chemical dosage range, determining the target chemical dosage prediction value according to the second chemical dosage prediction value and the minimum value in the first chemical dosage range indicated by the second chemical dosage prediction level.

[0104] For example, when the second chemical dosage prediction value is within the second chemical dosage range indicated by the second chemical dosage prediction level corresponding to any target chemical dosage prediction feature, the second chemical dosage prediction value may be determined as the target chemical dosage prediction value; otherwise, the minimum value between the first chemical dosage range indicated by the second chemical dosage prediction level corresponding to the target chemical dosage prediction feature and the second chemical dosage prediction value is determined as the target chemical dosage prediction value. In this way, the chemical dosage can be initially adjusted and controlled through the target chemical dosage prediction value, and then it can be determined whether further adjustment is needed through the method shown later, so as to adjust the chemical dosage in a timely manner while avoiding the situation of secondary pollution of water caused by excessive chemical dosage. Figure 6 It should be noted that the adjustment of the first chemical dosage range indicated by the second chemical dosage prediction level to the second chemical dosage range is only for using the second chemical dosage range in this comparison. The first chemical dosage range is still stored in the database. When predicting the chemical dosage value next time, the first chemical dosage range in the database is still used for determination.

[0105] In the present disclosure, chemical dosage prediction can be performed through different types of models. The prediction results of different types of models can be mutually verified and restricted, so as to improve the accuracy of the predicted chemical dosage prediction value. At the same time, by taking each chemical dosage prediction feature as an input separately, the chemical dosage can be predicted from multiple angles through multiple chemical dosage prediction features, enabling the model to adapt to a variety of different water quality environments and further improving the accuracy and reliability of the predicted chemical dosage.

[0106] Exemplarily,

[0107] Exemplarily, Figure 6 shows a schematic flowchart of a method for chemical dosage control according to a target chemical dosage prediction value in an exemplary embodiment of the present disclosure. Referring to Figure 6 , the method may include step S610 to step S660. Among them: In step S610, control the dosing pump to perform the dosing operation in the water treatment process according to the first target dosing amount indicated by the predicted value of the target dosing amount.

[0108] In an exemplary embodiment, the dosing pump can control the dosing amount by frequency. There is a mapping relationship between the frequency of the dosing pump and the dosing amount. Based on this mapping relationship, the frequency of the dosing pump corresponding to the predicted value of the target dosing amount can be determined, and then this frequency can be sent to the dosing pump through an instruction to control the dosing pump to perform the dosing operation according to this frequency.

[0109] In step S620, in response to monitoring that the turbidity of the water in the sedimentation tank decreases and the turbidity difference between adjacent sampling times is less than the third preset value for the first time, start the radar device and the video acquisition device acting on the sedimentation tank at the same time.

[0110] For example, after controlling the dosing pump to perform the dosing operation according to the predicted value of the target dosing amount, the dosing effect can be detected by the floc state in the sedimentation tank, so as to monitor whether the predicted value of the target dosing amount is appropriate, and determine whether the dosing amount needs to be adjusted according to the monitoring situation.

[0111] In an exemplary embodiment, while sending a dosing control instruction to the dosing pump, an opening control instruction can be sent to the turbidity detector in the sedimentation tank to turn on the turbidity detector in the sedimentation tank, and the turbidity in the sedimentation tank can be obtained through the turbidity detector.

[0112] When it is monitored by the turbidity meter that the turbidity of the water in the sedimentation tank decreases and the turbidity difference between adjacent sampling times is less than the third preset value for the first time, send an opening control instruction to the radar device and the video acquisition device acting on the sedimentation tank at the same time.

[0113] For example, when the turbidity of the water in the sedimentation tank shows a downward trend and begins to stabilize for the first time, it indicates that the flocs in the sedimentation tank have begun to settle. At this time, the dosing effect can be determined by the settling speed of the flocs.

[0114] Exemplarily, in the present disclosure, the settling speed of the flocs can be measured by the radar and the video at the same time to ensure the accuracy of the determination of the settling speed. Therefore, it is necessary for the radar and the video to measure the settling speed of the flocs in the same area within the same time period. So the radar device and the video acquisition device can be started at the same time. At the same time, it is necessary to layout the installation positions of the radar device and the video acquisition device to ensure that the action ranges of the radar device and the video acquisition device overlap to the greatest extent, so as to ensure as much as possible that the radar device and the video acquisition device measure the settling speed of the flocs in the same area, which is convenient for subsequent accurate fusion.

[0115] In step S630, the first average sedimentation velocity of the flocs in the sedimentation tank is measured by a radar device within a first preset time period.

[0116] In an exemplary embodiment, the first preset time period can be determined according to the radar turn-on time and a first preset duration, that is, the first preset time period is the time period within the first preset duration after the radar is turned on. Taking the first preset time period as 2 minutes as an example, the first preset time period is 2 minutes after the radar is turned on.

[0117] For example, the radar device can emit electromagnetic waves to the sedimentation tank. The flocs in the sedimentation tank will reflect the electromagnetic waves, and the radar will receive the reflected waves. According to the Doppler effect, when the target object moves relative to the radar, the frequency of the reflected wave will change. If the target object moves towards the radar, the frequency of the reflected wave will increase; if the target object moves away from the radar, the frequency of the reflected wave will decrease. The amount of frequency change is proportional to the moving speed of the target object. By measuring the frequency difference between the transmitted wave and the reflected wave, that is, the Doppler frequency shift, and combining the parameters of the radar system, the moving speed of the target object can be determined.

[0118] Exemplarily, the radar signal can be filtered to avoid the interference of water surface fluctuations on the radar signal. After the filtering process, according to the intensity distribution of the radar signal, the signals in different regions are subjected to gain adjustment, so as to facilitate subsequent analysis. Next, the radar signal in the time domain can be converted into a signal in the frequency domain, and the frequency spectrum distribution of the signal is obtained through Fourier transform. Extreme points are searched in the frequency spectrum, and the reflection signals of floc groups with different speeds correspond to different extreme points. The speed of each extreme point is determined according to the reflection frequency of each extreme point, and the speeds of all extreme points are weighted and averaged to obtain the average sedimentation velocity of the flocs within the detection range during this measurement process. Among them, the weight of the speed of each extreme point can be determined according to the energy of each extreme point. The greater the energy of the extreme point, the greater the weight of the speed corresponding to the extreme point. Thus, the speeds of different extreme points are weighted by the weight to obtain the average sedimentation velocity.

[0119] In an exemplary embodiment, the average sedimentation velocity of the flocs within the detection range can be measured multiple times through the radar signal within the first preset time period, and the average sedimentation velocities measured multiple times within the first preset time period are averaged again to obtain the first average sedimentation velocity of the flocs in the sedimentation tank within the first preset time period.

[0120] In an exemplary embodiment, the radar device can also be used to measure the separation interface between the water in the sedimentation tank and the mud at the bottom of the sedimentation tank, and determine whether to perform the sludge discharge operation of the sedimentation tank according to the height of the separation interface. For example, after continuous floc precipitation, sludge will gradually accumulate at the bottom of the sedimentation tank. The radar device can determine the stratification interface between the sludge and the water according to the different reflection characteristics of electromagnetic waves by different media, so as to obtain the height of the sludge interface. When the height of the sludge interface reaches a preset height, the alarm device can be triggered to remind to perform the sewage discharge operation on the sedimentation tank, avoiding excessive accumulation of sludge and affecting the precipitation effect.

[0121] In other words, in the present disclosure, the radar device installed in the sedimentation tank can be used to measure the sedimentation speed of the flocs and to remind of sewage discharge from the sedimentation tank. In this way, by reusing the radar device, while improving the water treatment effect, the hardware cost in the water treatment process can be reduced.

[0122] In step S640, the video in the sedimentation tank during the first preset period is collected by the video acquisition device, the floc target tracking and video depth estimation are performed on the video, and the second average sedimentation speed of the flocs in the sedimentation tank during the first preset period is determined according to the results of the floc target tracking and video depth estimation.

[0123] For example, the floc video can be sampled to obtain a sequence of floc image frames. Then, based on the object detection algorithm, all flocs are detected in the first frame of the floc image frame sequence, and then based on the multi-object tracking algorithm, the flocs detected in the first frame are tracked in the second and subsequent floc image frames of the floc image frame sequence. The two-dimensional pixel coordinates of the same floc in each floc image frame are determined through the object tracking result. At the same time, video depth estimation can be performed on the floc image frame sequence, and the depth value of the same floc in each floc image frame is determined according to the result of the video depth estimation. Through the camera calibration parameters, the pixel coordinates and depth values of the flocs are converted into three-dimensional spatial positions in the world coordinate system, the position component in the floc sedimentation direction in the three-dimensional spatial positions is extracted, and according to the absolute value of the difference between the position components of the current frame and the reference frame of the current frame, the sedimentation displacement of the flocs is determined. According to the sedimentation displacement and the time interval between the current frame and the reference frame, the sedimentation speed between the current frame and the reference frame is obtained. According to the mean value of the sedimentation speeds between each current frame and the reference frame, the mean value of the sedimentation speeds of the flocs during the first preset period is obtained, and according to the mean value of the sedimentation speeds of each successfully tracked floc during the first preset period, the second average sedimentation speed of the flocs during the first preset period is obtained.

[0124] Among them, the reference frame of the current frame can be the floc image frame located before the current frame and separated from the current frame by N frames. Here, N is determined according to experience or requirements. For example, N is 1, or N is 2, or N is 0, etc. This exemplary embodiment does not make special limitations on this. N being 0 means that the reference frame of the current frame is the previous frame of the current frame.

[0125] In step S650, the first average sedimentation velocity and the second average sedimentation velocity are fused to obtain the target average sedimentation velocity.

[0126] In an exemplary embodiment, the first average sedimentation velocity and the second average sedimentation velocity can be fused according to the first weight corresponding to the first average sedimentation velocity and the second weight corresponding to the second average sedimentation velocity to obtain the target average sedimentation velocity. Here, the first weight and the second weight can be determined according to requirements and experience. This exemplary embodiment does not make special limitations on this.

[0127] In step S660, according to the difference between the target average sedimentation velocity and the preset sedimentation velocity, the first target chemical dosage is adjusted to obtain the second target chemical dosage, and the chemical dosing pump is controlled to perform the chemical dosing operation in the water treatment process according to the second target chemical dosage.

[0128] In an exemplary embodiment, the preset sedimentation velocity is determined according to the floc sedimentation velocity corresponding to the target effluent water quality of the sedimentation tank. Here, the target effluent water quality can include the effluent water quality that meets the expected requirements, which is customarily determined according to requirements. This exemplary embodiment does not make special limitations on this.

[0129] Exemplarily, the preset sedimentation velocity can be determined according to the corresponding relationship between the historical sedimentation velocity and the effluent water quality. For example, the sedimentation velocity calculated each time and the corresponding effluent water quality can be recorded to obtain the first recorded data. Then, the second recorded data with the effluent water quality meeting the target effluent water quality is selected from the first recorded data, and then the preset sedimentation velocity is determined according to the sedimentation velocity in the second recorded data.

[0130] For example, the preset sedimentation rate can be determined based on the average value of the sedimentation rates in the second recorded data, or the preset sedimentation rate range can be determined based on the range formed by the minimum and maximum values of the sedimentation rates in the second recorded data. That is, the preset sedimentation rate can also be a range. When the preset sedimentation rate is a range, if the target average sedimentation rate falls within this range, it is considered that there is no difference between the target average sedimentation rate and the preset sedimentation rate, and in this case, the chemical dosing amount does not need to be adjusted. If the target average sedimentation rate is less than the minimum value of the preset sedimentation rate range and the absolute value of the difference from this minimum value is greater than the fourth preset value, the first adjustment strategy is executed; if the target average sedimentation rate is greater than the maximum value of the preset sedimentation rate range and the absolute value of the difference from this maximum value is greater than the fourth preset value, the second adjustment strategy is executed. When the preset sedimentation rate is a specific value, if the target average sedimentation rate is less than the preset sedimentation rate and the absolute value of the difference from the preset sedimentation rate is greater than the fourth preset value, the first adjustment strategy is executed; if the target average sedimentation rate is greater than the preset sedimentation rate and the absolute value of the difference from the preset sedimentation rate is greater than the fourth preset value, the second adjustment strategy is executed.

[0131] In an exemplary embodiment, the first adjustment strategy includes: increasing the current chemical dosing amount of the chemical dosing pump according to a preset increase ratio coefficient. For example, the preset increase ratio coefficient can be greater than 1, such as 1.05. In this way, each time the value obtained by multiplying the current chemical dosing amount by 1.05 is determined as the new chemical dosing amount. After re-dosing, the target average sedimentation rate is re-determined in the above manner, and then it is determined whether further adjustment is needed. This cycle continues until the difference between the newly determined target average sedimentation rate and the preset sedimentation rate meets the preset conditions, and then the adjustment stops.

[0132] In an exemplary embodiment, the second adjustment strategy includes: reducing the current chemical dosing amount of the chemical dosing pump according to a preset decrease ratio coefficient. For example, the preset decrease ratio coefficient can be less than 1, such as 0.95. In this way, each time the value obtained by multiplying the current chemical dosing amount by 0.95 is determined as the new chemical dosing amount. After re-dosing, the target average sedimentation rate is re-determined in the above manner, and then it is determined whether further adjustment is needed. This cycle continues until the difference between the newly determined target average sedimentation rate and the preset sedimentation rate meets the preset conditions, and then the adjustment stops.

[0133] In other words, if the sedimentation rate of the flocs is fast after chemical dosing, it indicates that the large chemical dosing amount causes the flocs to further aggregate into larger flocs, and the sedimentation rate may increase. If the sedimentation rate of the flocs is slow, it indicates that the chemical dosing amount is small.

[0134] Among them, the preset conditions include that the absolute value of the difference between the target average sedimentation rate and the preset sedimentation rate is less than the fourth preset value.

[0135] In an exemplary embodiment, the chemical dosage and influent parameter characteristics, floc characteristics, and effluent water quality parameter characteristics at the time of stopping adjustment can be associated to generate an updated database, and each model in the above-mentioned preset chemical dosage prediction model can be updated regularly according to the updated database.

[0136] Through the above steps S610 to S660, the sedimentation velocity of the flocs can be determined in various ways, improving the accuracy of determining the sedimentation velocity of the flocs. Based on the sedimentation velocity of the flocs, the dosing treatment effect of the target chemical dosage prediction value can be judged, and thus the chemical dosage can be adjusted according to the judgment result to more accurately control the chemical dosage. At the same time, adjusting the chemical dosage according to the floc condition in the sedimentation stage can improve the timeliness of adjusting the chemical dosage.

[0137] In the present disclosure, by using the method of taking multiple chemical dosage prediction characteristics as single inputs respectively for chemical dosage prediction, the chemical dosage can be predicted and verified from multiple perspectives respectively, so that the chemical dosage can be predicted under different water quality scenarios, improving the accuracy of chemical dosage prediction. At the same time, by performing chemical dosage prediction simultaneously with the chemical dosage classification prediction model and the chemical dosage regression prediction model, and verifying and constraining the prediction result of the chemical dosage regression prediction model based on the output result of the chemical dosage classification prediction model, not only can the specific chemical dosage value be predicted, but also the accuracy and reliability of the predicted specific chemical dosage value can be improved, thereby assisting in improving the stability and reliability of chemical dosing control in the water treatment process. In addition, since the accuracy of chemical dosage control is improved by the method of the present disclosure, the degree of manual participation in the chemical dosing process can be reduced, and the automation of the water treatment process can be realized to a higher degree.

[0138] In addition, it should be noted that the above-mentioned drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present invention, rather than for limiting purposes. It is easy to understand that the processes shown in the above-mentioned drawings do not indicate or limit the time sequence of these processes. Additionally, it is also easy to understand that these processes can be executed, for example, synchronously or asynchronously in multiple modules.

[0139] Furthermore, the exemplary embodiment of the present disclosure further provides a chemical dosing control device in the water treatment process. Refer to Figure 7As shown, the chemical dosing control device 700 in the water treatment process includes the following program modules: an eigenvalue acquisition module 710, configured to acquire the eigenvalues of the chemical dosing prediction features in the water treatment process, where the chemical dosing prediction features include floc characteristics, influent parameter characteristics, and effluent parameter characteristics; a first prediction module 720, configured to input the eigenvalues of each chemical dosing prediction feature into a preset chemical dosing prediction model, a chemical dosing regression prediction model with the chemical dosing prediction feature as a single input, and respectively obtain the first chemical dosing prediction values corresponding to each chemical dosing prediction feature according to the outputs of each chemical dosing regression prediction model; a second prediction module 730, configured to input the eigenvalues of each chemical dosing prediction feature into the preset chemical dosing prediction model, a chemical dosing classification prediction model with the chemical dosing prediction feature as a single input, and respectively obtain the first chemical dosing prediction levels corresponding to each chemical dosing prediction feature according to the outputs of each chemical dosing classification prediction; a target prediction module 740, configured to determine the second chemical dosing prediction value based on the first chemical dosing prediction values corresponding to each chemical dosing prediction feature, and determine the second chemical dosing prediction level based on the first chemical dosing levels corresponding to each chemical dosing prediction feature; a chemical dosing control module 750, configured to determine the target chemical dosing prediction value according to the overlapping relationship between the second chemical dosing prediction value and the second chemical dosing prediction level, and control the chemical dosing amount in the water treatment process based on the target chemical dosing prediction value.

[0140] In an exemplary embodiment, the determining the target chemical dosing prediction value according to the overlapping relationship between the second chemical dosing prediction value and the second chemical dosing prediction level includes: when the second chemical dosing prediction value is within the first chemical dosing range indicated by the second chemical dosing prediction level, determining the second chemical dosing prediction value as the target chemical dosing prediction value; when the second chemical dosing prediction value is not within the first chemical dosing range indicated by the second chemical dosing prediction level, obtaining the water flow rate and water turbidity detected at multiple detection points during the process from the raw water point to the influent point, and in the case where the water flow rate is greater than a first preset value and / or the water turbidity is greater than a second preset value, adjusting the first chemical dosing range indicated by the second chemical dosing prediction level according to historical experience to obtain a second chemical dosing range, and determining the target chemical dosing prediction value according to the overlapping relationship between the second chemical dosing range and the second chemical dosing prediction value.

[0141] In an exemplary embodiment, determining the target drug dosage prediction value according to the overlapping relationship between the second drug dosage range and the second drug dosage prediction value includes: when the second drug dosage prediction value is within the second drug dosage range, determining the second drug dosage prediction value as the target drug dosage prediction value; when the second drug dosage prediction value is not within the second drug dosage range, determining the target drug dosage prediction value according to the second drug dosage prediction value and the minimum value in the first drug dosage range indicated by the second drug dosage prediction level.

[0142] In an exemplary embodiment, the method for determining the preset drug dosage prediction model includes: respectively generating a first label data set corresponding to each drug dosage prediction feature according to the historical data corresponding to each drug dosage prediction feature and the historical drug dosage value corresponding to the historical data; respectively generating a second label data set corresponding to each drug dosage prediction feature according to the historical data corresponding to each drug dosage prediction feature and the drug dosage level to which the historical data belongs in different drug dosage levels corresponding to the drug dosage prediction feature; respectively training a first number of candidate regression prediction models based on the first label data set corresponding to each drug dosage prediction feature to obtain a second number of drug dosage regression prediction models, where the second number and the first number are in a multiple relationship; respectively training a third number of candidate classification prediction models based on the second label data set corresponding to each drug dosage prediction feature to obtain a fourth number of drug dosage classification prediction models, where the fourth number and the third number are in a multiple relationship; performing model testing on the second number of drug dosage regression prediction models, and determining a target drug dosage regression prediction model from the second number of drug dosage regression prediction models according to the test results; performing model testing on the fourth number of drug dosage classification prediction models, and determining a target drug dosage classification prediction model from the fourth number of drug dosage classification prediction models according to the test results; determining the preset drug dosage prediction model according to the target drug dosage regression prediction model and the target drug dosage classification prediction model.

[0143] In an exemplary embodiment, the method for determining different drug dosage levels corresponding to any drug dosage prediction feature includes: collecting the historical data of the drug dosage prediction feature, and clustering the collected historical data; determining different drug dosage prediction levels according to the drug dosage indicated by the historical data in each clustering category in the clustering result.

[0144] In an exemplary embodiment, determining different chemical dosage prediction levels according to the chemical dosages indicated by the historical data in each clustering category in the clustering result includes: determining the magnitude relationship of the chemical dosage levels corresponding to different clustering categories according to the magnitude relationship of the means of the chemical dosages indicated by the historical data in each clustering category in the clustering result; determining the chemical dosage range indicated by the chemical dosage level corresponding to the clustering category according to the minimum value and the maximum value of the chemical dosages indicated by the historical data in the clustering category.

[0145] In an exemplary embodiment, controlling the chemical dosage in the water treatment process based on the target chemical dosage prediction value includes: controlling the chemical dosing pump to perform the chemical dosing operation in the water treatment process according to the first target chemical dosage indicated by the target chemical dosage prediction value; in response to monitoring that the water turbidity in the sedimentation tank decreases and for the first time the turbidity difference between adjacent sampling times is less than the third preset value, starting the radar device and the video acquisition device acting on the sedimentation tank at the same time; measuring the first average sedimentation speed of the flocs in the sedimentation tank within the first preset period by the radar device; collecting the video within the first preset period in the sedimentation tank by the video acquisition device, performing floc target tracking and video depth estimation on the video, and determining the second average sedimentation speed of the flocs in the sedimentation tank within the first preset period according to the results of the floc target tracking and the video depth estimation; fusing the first average sedimentation speed and the second average sedimentation speed to obtain the target average sedimentation speed; adjusting the first target chemical dosage according to the difference between the target average sedimentation speed and the preset sedimentation speed to obtain the second target chemical dosage, and controlling the chemical dosing pump to perform the chemical dosing operation in the water treatment process according to the second chemical dosage; wherein, the preset sedimentation speed is determined according to the sedimentation speed of the flocs corresponding to the target effluent quality of the sedimentation tank.

[0146] The specific details of each part in the above device have been described in detail in the embodiments of the method part. The details not disclosed can be seen in the content of the embodiments of the method part, and thus will not be repeated.

[0147] It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the exemplary embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0148] In addition, although the steps of the methods in the present disclosure are described in a specific order in the accompanying drawings, this does not require or imply that these steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.

[0149] The exemplary embodiments of the present disclosure also provide a computer program product. The computer program product includes a computer program which, when executed by a processor, implements the dosing control method in the above-mentioned water treatment process.

[0150] In one embodiment, the computer program product may be a tangible product containing the computer program, such as a computer-readable storage medium storing the computer program. The readable storage medium may be a storage medium based on signals such as electricity, magnetism, light, electromagnetic, infrared, etc., including but not limited to: random access memory (RAM), read-only memory (ROM), magnetic tape, floppy disk, flash memory (Flash), hard disk drive (HDD), solid state drive (SSD), and so on. Exemplarily, the computer program product may be implemented as a non-volatile storage medium storing the computer program, such as read-only memory, NAND flash memory, etc.

[0151] In one embodiment, the computer program product may be an intangible product containing the computer program. Exemplarily, the computer program product may be implemented as a virtual digital product, such as an executable file storing the computer program, an installation package, and other digital files.

[0152] The code of the computer program can be written in one or more programming languages. Programming languages such as C, Java, C++, Python, etc. The program code can be executed entirely on the user's computing device, or partially on the user's computing device, or executed as an independent software package, or partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, such as a local area network (LAN), a wide area network (WAN), etc., or can be connected to an external computing device (for example, through an Internet connection provided by an operator).

[0153] A computer program can be carried or transmitted by signals such as electricity, magnetism, light, electromagnetic, infrared, etc. An electronic device can convert the signal carrying the computer program into a digital signal and then run the computer program. When the computer program runs on the electronic device, its code is used to cause the electronic device to execute (more specifically, to cause the processor of the electronic device to execute) the method steps of various exemplary embodiments of the present disclosure.

[0154] Exemplary embodiments of the present disclosure also provide an electronic device. The electronic device may include a processor and a memory. The memory stores executable instructions of the processor, such as a computer program. The processor executes the method steps of various exemplary embodiments of the present disclosure by executing the executable instructions. In addition, the electronic device may further include a display for displaying a graphical user interface.

[0155] Reference is made below Figure 8 , and an electronic device is exemplarily described in the form of a general computing device. It should be understood that Figure 8 the electronic device 800 shown is merely an example and should not impose limitations on the functions and usage scope of the embodiments of the present disclosure.

[0156] As Figure 8 shown, the electronic device 800 may include: a processor 810, a memory 820, a bus 830, an I / O (input / output) interface 840, a network adapter 850, and a display 860.

[0157] The memory 820 may include volatile memory, such as RAM 821 and a cache unit 822, and may also include non-volatile memory, such as ROM 823. The memory 820 may further include one or more program modules 824. Such program modules 824 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Implementations of a network environment may be included in each or some combination of these examples. For example, the program module 824 may include each module in the above device.

[0158] The processor 810 may include one or more processing units. For example: the processor 810 may include an AP (Application Processor), a modem processor, a GPU (Graphics Processing Unit), an ISP (Image Signal Processor), a controller, an encoder, a decoder, a DSP (Digital Signal Processor), a baseband processor, and / or an NPU (Neural-Network Processing Unit), etc.

[0159] The processor 810 can be used to execute executable instructions stored in the memory 820, such as the chemical dosing control method in the above-mentioned water treatment process.

[0160] The bus 830 is used to connect different components of the electronic device 800 and can include a data bus, an address bus, and a control bus.

[0161] The electronic device 800 can communicate with one or more external devices 900 (such as a keyboard, a mouse, an external controller, etc.) through the I / O interface 840.

[0162] The electronic device 800 can communicate with one or more networks through the network adapter 850. For example, the network adapter 850 can provide mobile communication solutions such as 3G / 4G / 5G, or wireless communication solutions such as wireless local area network, Bluetooth, and near field communication. The network adapter 850 can communicate with other modules of the electronic device 800 through the bus 830.

[0163] The electronic device 800 can display a graphical user interface through the display 860, such as a chemical dosing prediction value interface.

[0164] Although Figure 8 not shown in the figure, other hardware and / or software modules can also be set in the electronic device 800, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0165] In addition, the above-mentioned drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, rather than for limiting purposes. It is easy to understand that the processes shown in the above-mentioned drawings do not indicate or limit the time sequence of these processes. Additionally, it is also easy to understand that these processes can be executed synchronously or asynchronously in, for example, multiple modules.

[0166] As can be seen from the above, the technical solution of the present disclosure can be implemented as a method, a device, a system, a computer program product, a storage medium, an electronic device, etc. Those skilled in the art can understand that various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware implementation manner, a complete software implementation manner (including firmware, microcode, etc.), or an implementation manner combining hardware and software aspects, such as can be respectively referred to as "circuit", "module", or "system".

[0167] It should be understood that the present disclosure is not limited to the specific method steps or structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. Those skilled in the art will readily think of other embodiments based on the specific embodiments provided by the present disclosure. Therefore, the specific embodiments provided by the present disclosure are only exemplary, and the scope and spirit of the present disclosure are pointed out by the claims, which should cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common general knowledge or conventional technical means in the technical field not disclosed by the present disclosure.

Claims

1. A chemical dosing control method in a water treatment process, characterized in that Including: Obtaining the eigenvalue of the chemical dosage prediction feature in the water treatment process, where the chemical dosage prediction feature includes floc characteristics, influent parameter characteristics, and effluent parameter characteristics; Respectively inputting the eigenvalue of each chemical dosage prediction feature into a preset chemical dosage prediction model, a chemical dosage regression prediction model with the chemical dosage prediction feature as a single input, and respectively obtaining the first chemical dosage prediction value corresponding to each chemical dosage prediction feature according to the output of each chemical dosage regression prediction model; Respectively inputting the eigenvalue of each chemical dosage prediction feature into the preset chemical dosage prediction model, a chemical dosage classification prediction model with the chemical dosage prediction feature as a single input, and respectively obtaining the first chemical dosage prediction level corresponding to each chemical dosage prediction feature according to the output of each chemical dosage classification prediction; Determining a second chemical dosage prediction value based on the first chemical dosage prediction value corresponding to each chemical dosage prediction feature, and determining a second chemical dosage prediction level based on the first chemical dosage level corresponding to each chemical dosage prediction feature; Determining a target chemical dosage prediction value according to the overlapping relationship between the second chemical dosage prediction value and the second chemical dosage prediction level, and controlling the chemical dosage in the water treatment process based on the target chemical dosage prediction value.

2. The method according to claim 1, characterized in that, The determining the target chemical dosage prediction value according to the overlapping relationship between the second chemical dosage prediction value and the second chemical dosage prediction level includes: When the second chemical dosage prediction value is within the first chemical dosage range indicated by the second chemical dosage prediction level, determining the second chemical dosage prediction value as the target chemical dosage prediction value; When the second chemical dosage prediction value is not within the first chemical dosage range indicated by the second chemical dosage prediction level, obtaining the water flow rate and water turbidity detected at multiple detection points during the process from the raw water point to the influent point. In the case where the water flow rate is greater than a first preset value and / or the water turbidity is greater than a second preset value, adjusting the first chemical dosage range indicated by the second chemical dosage prediction level according to historical experience to obtain a second chemical dosage range, and determining the target chemical dosage prediction value according to the overlapping relationship between the second chemical dosage range and the second chemical dosage prediction value.

3. The method according to claim 2, wherein The determining the target chemical dosage prediction value according to the overlapping relationship between the second chemical dosage range and the second chemical dosage prediction value includes: When the second chemical dosage prediction value is within the second chemical dosage range, determining the second chemical dosage prediction value as the target chemical dosage prediction value; When the second chemical dosage prediction value is not within the second chemical dosage range, determining the target chemical dosage prediction value according to the minimum value in the second chemical dosage prediction value and the first chemical dosage range indicated by the second chemical dosage prediction level.

4. The method according to claim 1, characterized in that, The determination method of the preset chemical dosage prediction model includes: Respectively generating a first label data set corresponding to each chemical dosage prediction feature according to the historical data corresponding to each chemical dosage prediction feature and the historical chemical dosage value corresponding to the historical data; Generate a second label data set corresponding to each chemical dosage prediction feature respectively according to the historical data corresponding to each chemical dosage prediction feature and the chemical dosage level to which the historical chemical dosage value corresponding to the historical data belongs among different chemical dosage levels corresponding to the chemical dosage prediction feature; Train a first number of candidate regression prediction models respectively based on the first label data set corresponding to each chemical dosage prediction feature to obtain a second number of chemical dosage regression prediction models, where the second number and the first number are in a multiple relationship; Train a third number of candidate classification prediction models respectively based on the second label data set corresponding to each chemical dosage prediction feature to obtain a fourth number of chemical dosage classification prediction models, where the fourth number and the third number are in a multiple relationship; Perform model testing on the second number of chemical dosage regression prediction models, and determine a target chemical dosage regression prediction model from the second number of chemical dosage regression prediction models according to the test results; Perform model testing on the fourth number of chemical dosage classification prediction models, and determine a target chemical dosage classification prediction model from the fourth number of chemical dosage classification prediction models according to the test results; Determine the preset chemical dosage prediction model according to the target chemical dosage regression prediction model and the target chemical dosage classification prediction model; 5. The method according to claim 4, wherein The determination method of different chemical dosage levels corresponding to any chemical dosage prediction feature includes: Collect the historical data of the chemical dosage prediction feature, and cluster the collected historical data; Determine different chemical dosage prediction levels according to the chemical dosage indicated by the historical data in each clustering category in the clustering result; 6. The method according to claim 5, wherein The determination of different chemical dosage prediction levels according to the chemical dosage indicated by the historical data in each clustering category in the clustering result includes: Determine the magnitude relationship of the chemical dosage levels between different clustering categories corresponding to different chemical dosage levels according to the magnitude relationship of the means of the chemical dosages indicated by the historical data in each clustering category in the clustering result; Determine the chemical dosage range indicated by the chemical dosage level corresponding to the clustering category according to the minimum and maximum values of the chemical dosage indicated by the historical data in the clustering category; 7. The method according to any one of claims 1 to 6, characterized in that The control of the chemical dosage in the water treatment process based on the target chemical dosage prediction value includes: Control the chemical dosing pump to perform the chemical dosing operation in the water treatment process according to the first target chemical dosage indicated by the target chemical dosage prediction value; In response to detecting that the water turbidity in the sedimentation tank decreases and the turbidity difference between adjacent sampling times is less than a third preset value for the first time, simultaneously start the radar device and the video acquisition device acting on the sedimentation tank; Measure the first average sedimentation velocity of the flocs in the sedimentation tank within a first preset period through the radar device; Collect the video of the sedimentation tank within the first preset period through the video acquisition device, perform floc target tracking and video depth estimation on the video, and determine the second average sedimentation velocity of the flocs in the sedimentation tank within the first preset period according to the results of the floc target tracking and video depth estimation; Fuse the first average sedimentation velocity and the second average sedimentation velocity to obtain a target average sedimentation velocity; Adjust the first chemical dosage according to the difference between the target average sedimentation rate and the preset sedimentation rate to obtain a second chemical dosage, and control the chemical dosing pump to perform the chemical dosing operation in the water treatment process according to the second chemical dosage; Wherein, the preset sedimentation rate is determined according to the floc sedimentation rate corresponding to the target effluent quality of the sedimentation tank.

8. A chemical dosing control device in a water treatment process, characterized in that, Comprising: An eigenvalue acquisition module configured to acquire the eigenvalues of the chemical dosage prediction features in the water treatment process, where the chemical dosage prediction features include floc features, influent parameter features, and effluent parameter features; A first prediction module configured to input the eigenvalues of each chemical dosage prediction feature into a preset chemical dosage prediction model, a chemical dosage regression prediction model with the chemical dosage prediction feature as a single input, and respectively obtain the first chemical dosage prediction values corresponding to each chemical dosage prediction feature according to the outputs of each chemical dosage regression prediction model; A second prediction module configured to input the eigenvalues of each chemical dosage prediction feature into the preset chemical dosage prediction model, a chemical dosage classification prediction model with the chemical dosage prediction feature as a single input, and respectively obtain the first chemical dosage prediction levels corresponding to each chemical dosage prediction feature according to the outputs of each chemical dosage classification prediction; A target prediction module configured to determine a second chemical dosage prediction value based on the first chemical dosage prediction values corresponding to each chemical dosage prediction feature, and determine a second chemical dosage prediction level based on the first chemical dosage levels corresponding to each chemical dosage prediction feature; A chemical dosing control module configured to determine a target chemical dosage prediction value according to the overlapping relationship between the second chemical dosage prediction value and the second chemical dosage prediction level, and control the chemical dosage in the water treatment process based on the target chemical dosage prediction value.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, Comprising: One or more processors; A storage device for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the method according to any one of claims 1 to 7.

Images