Flocculation effect detection method, device and equipment

Abstract

The invention provides a flocculation effect detection method, device and equipment, and the method comprises the following steps: respectively analyzing continuously acquired liquid images of target liquid through an image analysis technology, and identifying floccules in the target liquid; obtaining the concentration parameter of the flocculating constituent in the target liquid through a particle detection technology; based on the identified flocculating constituent and the flocculating constituent concentration parameter, determining the flocculating constituent characteristic parameter of the liquid; and determining the flocculation index of the current liquid by adopting a predetermined flocculation index algorithm according to the characteristic parameters of the flocculating constituent. According to the method provided by the invention, various characteristic parameters of the flocculating constituent can be combined with the film paving effect of the filter layer of the powder covering filter, so that the film paving effect can be evaluated and predicted cooperatively by various parameters, and the accuracy is improved; and through an image processing technology, errors of a visual observation method are avoided.

Description

Technical Field

[0001] This disclosure relates to the field of water treatment, and in particular to a method, apparatus and equipment for detecting flocculation effect. Background Technology

[0002] Powder coated filters are widely used in water treatment systems. They are devices that purify liquids by uniformly coating tubular filter elements with cation and anion powder resins to form a filter membrane, utilizing the adsorption and retention effects of the membrane. In the condensate polishing system of air-cooled units, powder coated filters are mainly used to remove iron ions from the condensate. If the powder coating is ineffective, the filter efficiency will decrease, allowing iron ion products to enter the air-cooled unit's feedwater system and cause scaling on thermal equipment. If the coating is uneven or incomplete, powder may enter the unit's boiler, leading to increased boiler water conductivity and decreased pH, which in turn can cause corrosion of thermal equipment and affect the safe operation of the unit.

[0003] To ensure effective membrane deposition, power plants often use powdered resin and fiber powder in specific ratios. The size and flocculation state of the flocs formed by the cation and anion powdered resins and fiber powder are key factors affecting membrane deposition quality. Overly large flocs can lead to difficulties in membrane deposition on the upper part of the filter element, uneven deposition, and a loose filter membrane; while overly small flocs may result in an overly dense filter membrane, leading to a larger initial pressure differential in the system.

[0004] Currently, the settling performance and physical properties of powdered resins are typically evaluated based on the state of flocs in a liquid. However, existing evaluation methods are mostly conventional visual methods, which have significant limitations: on the one hand, they rely on only a few parameters such as visual turbidity and settling ratio, lacking a multi-parameter synergistic evaluation system; on the other hand, visual observation lacks sensitivity and is difficult to accurately capture the characteristic parameters of flocs. Therefore, there is an urgent need for a method to detect flocculation effects that can accurately obtain multiple characteristic parameters of flocs and perform synergistic evaluation. Summary of the Invention

[0005] This disclosure provides a method, apparatus, and equipment for detecting flocculation effect, which solves the problem that existing evaluation methods cannot accurately obtain the characteristic parameters of flocs and perform collaborative evaluation.

[0006] In view of the above problems, firstly, embodiments of this disclosure provide a method for detecting flocculation effect, including: Image analysis technology is used to analyze continuously acquired liquid images of the target liquid to identify flocs in the target liquid; and particle detection technology is used to obtain the concentration parameters of flocs in the target liquid. Based on the identified flocs and the floc concentration parameters, the floc characteristic parameters of the liquid are determined. Based on the floc characteristic parameters, the flocculation index of the current liquid is determined using a pre-determined flocculation index algorithm.

[0007] In conjunction with the first aspect, in one possible implementation, determining the flocculant characteristic parameters of the liquid based on the identified flocculants and the flocculant concentration parameters includes: Determine the particle size of each floc, and based on the particle size of each floc, determine the median particle size of the flocs in the liquid; Based on the changes of each floc in continuous liquid images, the settling velocity of each floc is determined; and based on the settling velocity of each floc, the average settling velocity of the flocs in the liquid is determined. The concentration decay rate of flocs in the liquid is determined based on the changes in floc concentration parameters within a preset time period. The median particle size, the concentration decay rate, and the average settling velocity are determined as the floc characteristic parameters.

[0008] In conjunction with the first aspect, in one possible implementation, determining the flocculation index of the current liquid using a predetermined flocculation index algorithm based on the floc characteristic parameters includes: We determined that each floc characteristic parameter should be normalized separately. The normalized floc characteristic parameters are substituted into the flocculation index calculation formula to obtain the flocculation index; wherein the flocculation index calculation formula is a predetermined function with each floc characteristic parameter as the independent variable and the flocculation index as the dependent variable.

[0009] In conjunction with the first aspect, in one possible implementation, the flocculation index calculation formula is determined using the following method: Construct a functional relationship determination model based on a neural network model; Multiple sets of sample data containing flocculant characteristic parameters are used as input, and the flocculation index corresponding to each set of sample data is used as the true value. Through the learning process of the neural network, the functional relationship is adjusted to determine the network parameters of the model. Based on the network parameters, the weight coefficient of each floc characteristic parameter in the flocculation index calculation formula is determined, thus obtaining the flocculation index calculation formula.

[0010] In conjunction with the first aspect, in one possible implementation, it further includes: comparing the flocculation index with a preset threshold to determine the flocculation state of the liquid; Adjust the feeding situation for the liquid according to the flocculation state.

[0011] In conjunction with the first aspect, in one possible implementation, comparing the flocculation index with a preset threshold to determine the flocculation state of the liquid includes: If the flocculation index is greater than a first threshold, the liquid is determined to be in a normal flocculation state. If the flocculation index is greater than the second threshold and less than the first threshold, the liquid is determined to be in a state of insufficient flocculation. If the flocculation index is less than the second threshold, the liquid is determined to be in an abnormal flocculation state.

[0012] In conjunction with the first aspect, in one possible implementation, adjusting the feed dosage for the liquid based on the flocculation state includes: For the normal flocculation state, the feed ratio for the current liquid is not adjusted; To address the insufficient flocculation, increase the proportion of filter aid for the current liquid. In response to the aforementioned abnormal flocculation state, adjust the feed rate for the current liquid.

[0013] In conjunction with the first aspect, in one possible implementation, the root analyzes continuously acquired liquid images of the target liquid using image analysis technology to identify flocculants in the target liquid, including: Continuously acquire liquid images of the target liquid, and preprocess the liquid images to obtain grayscale images for image recognition; Based on the grayscale image, the floc boundaries in the grayscale image are identified by a detection operator. Connectivity analysis was performed on the identified floc boundaries to delineate each independent floc region in continuous images.

[0014] Secondly, embodiments of this disclosure provide a flocculation effect detection device, comprising: The identification module is used to analyze continuously acquired liquid images of the target liquid using image analysis technology to identify flocs in the target liquid; and to obtain floc concentration parameters in the target liquid using particle detection technology. The first determining module is used to determine the flocculant characteristic parameters of the liquid based on the identified flocculants and the flocculant concentration parameters. The second determining module is used to determine the flocculation index of the current liquid based on the flocculant characteristic parameters and using a pre-determined flocculation index algorithm.

[0015] Thirdly, this disclosure provides a flocculation effect detection device, including: a camera device, a main control device, and a communication device; The camera device is used to continuously acquire liquid images of the target liquid in the water treatment system and output the liquid images to the main control device. The main control device is used to analyze continuously acquired liquid images of the target liquid using image analysis technology to identify flocs in the target liquid; and to obtain floc concentration parameters in the target liquid using particle detection technology; based on the identified flocs and the floc concentration parameters, to determine the floc characteristic parameters of the liquid; and based on the floc characteristic parameters, to determine the floc index of the current liquid using a pre-determined floc index algorithm. The communication device is used to transmit floc characteristic parameters and flocculation index to the host computer.

[0016] The beneficial effects of the embodiments disclosed herein include: This disclosure provides a method, apparatus, and device for detecting flocculation effect, comprising: analyzing continuously acquired liquid images of a target liquid using image analysis technology to identify flocs in the target liquid; obtaining floc concentration parameters in the target liquid using particle detection technology; determining floc characteristic parameters of the liquid based on the identified flocs and the floc concentration parameters; and determining the flocculation index of the current liquid using a pre-determined flocculation index algorithm based on the floc characteristic parameters. According to the method provided in this disclosure, multiple characteristic parameters of flocs can be combined with the membrane formation effect of the filter layer of a powder-coated filter to achieve synergistic evaluation and prediction of the membrane formation effect using multiple parameters, thus improving accuracy; and the image processing technology avoids errors associated with visual observation. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the flocculation effect detection method provided in the embodiments of this disclosure; Figure 2 This is a schematic diagram of the water treatment system structure provided in an embodiment of the present disclosure; Figure 3 A comparison diagram of the film-laying effect between the detection method provided in this embodiment and the traditional detection method; Figure 4 This is a schematic diagram of the flocculation effect detection device provided in an embodiment of the present disclosure. Detailed Implementation

[0018] This disclosure provides a method, apparatus, and device for detecting flocculation effect. Preferred embodiments of this disclosure are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustrative and explanatory purposes only and are not intended to limit this disclosure. Furthermore, the embodiments and features described in this application can be combined with each other unless otherwise specified.

[0019] This disclosure provides a method for detecting flocculation effect, such as... Figure 1 As shown, it can be implemented as follows: S101. Analyze the continuously acquired liquid images of the target liquid using image analysis technology to identify flocs in the target liquid; and obtain the floc concentration parameters in the target liquid using particle detection technology. S102. Based on the identified flocs and floc concentration parameters, determine the floc characteristic parameters of the liquid. S103. Based on the characteristic parameters of the flocs, the flocculation index of the current liquid is determined using a pre-determined flocculation index algorithm.

[0020] In this embodiment, before the powder-coated filter officially begins filtration, the exchange resin and filter aid need to be prepared into a slurry according to a certain ratio. Through circulation or sedimentation, these powders are uniformly deposited on the surface of the filter element, forming an initial filter layer with a porous structure. During this process, after the exchange resin and filter aid are added to the water treatment system, they combine and adsorb, gradually forming larger flocculent structures, i.e., flocs, which are deposited and cover the filter element to form a filter layer. In this disclosure, the exchange resin can be anionic or cationic powder resin, mainly targeting ionic impurities dissolved in water. Among them, cation resin powder can remove cationic impurities by exchanging with cations in water through its own hydrogen or ammonium ions; anionic resin powder can remove anionic impurities and organic acid radicals in water by exchanging with anions in water through its own hydroxide ions. The filter aid can be fiber powder, which can form the skeleton of the filter layer and effectively intercept insoluble impurities such as suspended solid particles, colloidal substances, corrosion products (such as iron and copper oxides), microorganisms, and broken resin fragments in the water through physical interception.

[0021] The settling and flocculation effects of flocs determine the formation effect of the filter layer in a powder-coated filter. In a water treatment system, a high-definition microscope camera can be installed at a viewing mirror to capture real-time images of the target liquid within the system. Image analysis technology can be used to analyze the liquid images to identify flocs. By analyzing the morphology of the flocs in the images, their characteristics can be obtained. Furthermore, particle detection technology can be used to obtain the concentration parameters of the flocs and analyze their concentration changes over a period of time, ultimately acquiring the characteristic parameters of the flocs. Based on these characteristic parameters, a pre-determined flocculation index algorithm is introduced to determine the flocculation index of the current liquid. This flocculation index is then used to determine the potential membrane formation effect of the powder-coated filter layer. This method combines multiple characteristic parameters of the flocs with the membrane formation effect of the powder-coated filter layer, enabling a synergistic evaluation and prediction of the membrane formation effect, improving evaluation accuracy. Moreover, image processing technology avoids the errors inherent in visual observation.

[0022] In another embodiment provided in this disclosure, the above-mentioned step S102, "determining the flocculant characteristic parameters of the liquid based on the identified flocculants and flocculant concentration parameters," can be implemented as follows: Step 1: Determine the particle size of each floc and, based on the particle size of each floc, determine the median particle size of the flocs in the liquid; Step 2: Based on the changes of each floc in the continuous liquid image, determine the settling velocity of each floc; and based on the settling velocity of each floc, determine the average settling velocity of the flocs in the liquid. Step 3: Determine the concentration decay rate of flocs in the liquid based on the changes in floc concentration parameters within a preset time period; Step 4: Determine the median particle size, concentration decay rate, and average settling velocity as the characteristic parameters of the flocs.

[0023] In this embodiment of the disclosure, image analysis can be performed on the acquired liquid image to determine the particle size of each floc in the image, thereby determining the median particle size of all flocs. Based on the positional changes of each floc over a period of time, the settling velocity of that floc can be determined, and further, the settling velocity of all flocs in the liquid can be determined. Furthermore, based on the detected changes in liquid concentration, the floc concentration decay rate over a period of time can be determined accordingly. In practical applications, this time period can be defined as 3 minutes to characterize the floc concentration decay rate.

[0024] In another embodiment provided in this disclosure, the above step S103, "determining the flocculation index of the current liquid using a predetermined flocculation index algorithm based on the floc characteristic parameters," can be implemented as follows: Step 1: Determine the normalization process for each floc characteristic parameter; Step 2: Substitute the normalized floc characteristic parameters into the flocculation index calculation formula to obtain the flocculation index; The formula for calculating the flocculation index is predetermined and is a functional relationship with various floc characteristic parameters as independent variables and the flocculation index as the dependent variable.

[0025] In this embodiment of the disclosure, normalization processing can be used to classify each floc characteristic parameter into the same dimension, eliminating dimensional differences between different characteristic parameters and preventing numerical instability in the model due to dimensional differences. The median particle size can be normalized using the maximum detected particle size; the average settling velocity can be normalized using the maximum detected settling velocity; and the concentration decay rate can be normalized using the maximum concentration decay rate measured multiple times.

[0026] The normalized characteristic parameters of each floc are substituted into the flocculation index calculation formula. The current flocculation index value is calculated by this formula, thereby judging the flocculation effect of the current liquid and further determining the final film-laying effect.

[0027] In yet another embodiment provided in this disclosure, the formula for calculating the flocculation index is determined using the following steps: Step 1: Construct a functional relationship determination model based on a neural network model; Step 2: Take multiple sets of sample data containing flocculant characteristic parameters as input, take the flocculation index corresponding to each set of sample data as the true value, and determine the network parameters of the model by adjusting the functional relationship through the learning process of the neural network. Step 3: Based on the network parameters, determine the weight coefficient of each floc characteristic parameter in the flocculation index calculation formula to obtain the flocculation index calculation formula.

[0028] In this embodiment of the disclosure, the formula for calculating the flocculation index can be determined through the learning process of a neural network model. First, a neural network model can be constructed as a functional relationship determination model to determine the functional relationship between the floc characteristic parameters and the flocculation index.

[0029] In the functional relationship determination model, the number of input layer neurons can be determined based on the number of flocculant characteristic parameters. In this disclosure, three input layer neurons are used: median particle size, average settling velocity, and concentration decay rate. The number of flocculant characteristic parameters and the number of input neurons can be adjusted according to the needs of the actual application. For cases requiring further improvement in measurement and prediction accuracy, the number of flocculant characteristic parameters can be increased; for cases requiring improved prediction speed, the number of flocculant characteristic parameters can be decreased.

[0030] In the output layer, a neuron can be set as the output neuron. By configuring corresponding network parameters for each flocculant characteristic parameter, the functional relationship determines the model's final output: the predicted flocculation index. The predicted flocculation index is compared with a pre-determined flocculation index used as the true input, and the difference between the predicted and true values ​​is quantified using a loss function (e.g., the root mean square error of the predicted flocculation index). Using the backpropagation algorithm, the network parameters in the model are adjusted in the direction of reducing the loss function until the loss function value is minimized or the number of adjustments to the network parameters reaches a preset number. Training can then be considered complete, and the network parameters in the functional relationship determiner model can reflect the relationship between the flocculant characteristic parameters and the flocculation index.

[0031] The weight coefficients corresponding to the characteristic parameters of each floc can be extracted from the network parameters to obtain the flocculation index calculation formula. The obtained flocculation index calculation formula is shown in formula (1).

[0032] (1) in, The flocculation index; The median particle size of the flocculent; The average settlement velocity; The concentration decay rate is the three-minute depletion rate (TDR). 、 、 These are the weighting coefficients for each floc characteristic parameter.

[0033] In one possible implementation, the neural network model can preferably be a Long Short-Term Memory (LSTM) neural network model, which can incorporate time series data into the training process of the network model. By adding time series data as training data to the correspondence between flocculant characteristic parameters and flocculation index, the correlation between flocculant characteristic parameters and flocculation index over time can be determined.

[0034] In yet another embodiment provided in this disclosure, the flocculation effect detection method further includes the following steps: Step 1: Compare the flocculation index with the preset threshold to determine the flocculation state of the liquid; Step 2: Adjust the feeding amount of the liquid according to the flocculation state.

[0035] In this embodiment of the disclosure, a preset threshold is set according to the predetermined correspondence between the flocculation index and the final film-laying effect.

[0036] The membrane coating effect can be characterized by indicators such as coverage, uniformity, and initial pressure difference. Coverage refers to the extent of the filter layer formed by the flocculants settling on the filter element; uniformity refers to the smoothness of the flocculants covering the filter element; initial pressure difference is the pressure difference between the inlet and outlet of the filter when the powder covers the filter at the beginning of filtration. A small pressure difference indicates stable and smooth filter operation, while a large pressure difference indicates filter layer blockage and inability to filter normally. Among the above, uniformity... ,in, The average particle size of the flocs is _____. This represents the number of flocs. Let be the particle size of the i-th floc. The standard deviation and uniformity of the flocs. The closer the value is to 1, the higher the uniformity of the membrane layer. Based on multiple indicators, the final membrane laying effect can be evaluated. According to the final membrane laying effect, multiple thresholds can be selected from the flocculation index to characterize the correspondence between the flocculation index and the membrane laying effect.

[0037] Depending on the flocculation index, the ratio of exchange resin and filter aid added to the water treatment system can be changed to adjust the flocculation index to the target value, so as to ensure that the final membrane laying effect reaches the target state.

[0038] In another embodiment provided in this disclosure, step one above, "comparing the flocculation index with a preset threshold to determine the flocculation state of the liquid," can be implemented as follows: Step 1: If the flocculation index is greater than the first threshold, determine that the liquid is in a normal flocculation state; Step 2: If the flocculation index is greater than the second threshold and less than the first threshold, determine that the liquid is in a state of insufficient flocculation. Step 3: If the flocculation index is less than the second threshold, the liquid is determined to be in an abnormal flocculation state.

[0039] In this embodiment of the disclosure, if the flocculation index is above the first threshold, it can be determined that the filter layer obtained by the final membrane is stable, has high uniformity and can cover the filter element, and the various properties of the filter layer can meet the standards for performing filtration operations, and the exchange agent and filter aid are in a normal flocculation state.

[0040] If the flocculation index is between the second and first thresholds, it can be determined that the final membrane-coated filter layer is slightly abnormal, with a loose structure (insufficient coverage) and localized agglomeration (insufficient uniformity). These phenomena are generally due to insufficient proportion of filter aid, resulting in weak connection ability of the flocs composed of exchange agent and filter aid, and the exchange agent and filter aid being in an insufficient flocculation state.

[0041] If the flocculation index is less than the second threshold, it can be determined that the mixed system of exchange agent and filter aid has a serious abnormality. In the end, the filter layer may be difficult to form and cannot perform filtration. The exchange agent and filter aid are in an abnormal flocculation state.

[0042] In one possible implementation, the first threshold can be implemented as 0.8; the second threshold can be implemented as 0.6.

[0043] In another embodiment provided in this disclosure, step two above, "adjusting the feeding situation for the liquid according to the flocculation state," can be implemented as follows: Step 1: For normal flocculation, do not adjust the feed ratio for the current liquid; Step 2: To address insufficient flocculation, increase the proportion of filter aid for the current liquid. Step 3: Adjust the feed rate for the current liquid in response to abnormal flocculation conditions.

[0044] In this embodiment of the disclosure, the ratio of exchange agent and filter aid reaches a relative balance under normal flocculation conditions, so there is no need to adjust the feeding ratio and the status quo can be maintained.

[0045] When flocculation is insufficient, the performance of the mixed system composed of exchange agent and filter aid exhibits slight abnormalities. This can be remedied by increasing the proportion of filter aid in the mixed system, enhancing its binding effect, compensating for the current system's strength or stability deficiencies, and causing the flocculation index to rise back above the first threshold.

[0046] In cases of abnormal flocculation, the concentrations of the exchange resin and filter aid in the liquid may be too low to form an effective mixture; or the concentrations may be too high, causing the liquid to become too viscous and preventing the effective formation of a filter layer. The feed rates can be adjusted accordingly to ensure proper filter layer formation on the filter elements. Simultaneously, the operating status of the water treatment system's circulation pump can be altered to agitate the liquid, ensuring uniform mixing of the flocs and preventing localized data from affecting the flocculation index measurement. In cases of abnormal flocculation, an alarm signal can also be output to alert personnel to inspect the water treatment system.

[0047] In another embodiment provided in this disclosure, the above-mentioned step S101, "analyzing the continuously acquired liquid images of the target liquid using image analysis technology to identify flocculants in the target liquid," can be implemented as follows: Step 1: Continuously acquire liquid images of the target liquid and preprocess the liquid images to obtain grayscale images for image recognition; Step 2: Based on the grayscale image, identify the floc boundaries in the grayscale image using detection operators; Step 3: Perform connectivity analysis on the identified floc boundaries to delineate each independent floc region in the continuous images.

[0048] In this embodiment, a high-precision industrial camera combined with magnification equipment can be used to acquire the microscopic morphology of flocs in real time. Generally, during the acquisition process, backlighting can be provided to the liquid using illumination equipment to improve the contrast between the flocs and the background, thereby increasing the recognition success rate. Furthermore, during preprocessing, the acquired images can be converted to grayscale images to reduce the amount of data and improve the accuracy of edge recognition. Through algorithms such as Gaussian filtering and median filtering, random noise in the image can be removed, the image smoothed, and edge information preserved. For cases where the contrast between the flocs and the background is not obvious, the image contrast can be adjusted accordingly to enhance the edges of the flocs, making them easier to identify.

[0049] For the preprocessed grayscale image, an edge detection operator is used to traverse the grayscale image, binarizing and separating the edges of the flocs from the background to obtain a binarized image. Isolated noise points in this image (generally identified by small pores in the flocs) can be eliminated or improved based on morphological operations, thereby determining all floc boundaries in the image.

[0050] For these floc boundaries, connectivity analysis can be performed to identify the complete region enclosed by each floc boundary as an independent floc region. Based on the typical size range of flocs, excessively small and low connectivity regions are considered noise regions, and floc regions that do not conform to the normal floc morphology are removed using the general morphology of flocs.

[0051] Furthermore, by labeling the detected flocs in consecutive images, the changes in each floc can be tracked, and the characteristics of the flocs can be determined. For the particle size of each floc, the area of ​​each detected floc can be represented as an equivalent circle, and the diameter of the circle can be determined as the particle size of the corresponding floc.

[0052] Based on the above, an embodiment is provided here as an example to illustrate the flocculation effect detection process.

[0053] In this embodiment, the water treatment system that needs to be tested for flocculation effect is as follows: Figure 2 As shown. Liquid images can be acquired through the viewing port of the water treatment system using a high-speed microscopic imaging system. The high-speed microscopic imaging system can employ a high-speed camera (preferably with a frame rate of [missing information]). The above-mentioned equipment, equipped with a microscope lens (preferably with a magnification of 10× to 100×), allows for real-time acquisition of the microscopic morphology of flocs. A particle counter based on the optical obscuration principle can also be installed to detect the floc concentration in the internal liquid in real time through a viewing hole. The acquired liquid images and floc concentration data can be sent to a computing device with image recognition, calculation, and communication capabilities for further processing.

[0054] This computing device can be used to analyze and obtain the median particle size, concentration decay rate, and average settling velocity of the flocs. In this embodiment, the median particle size is assumed to be... Concentration decay rate The average settlement velocity is These values ​​were normalized respectively, assuming the maximum particle size was... The maximum settling velocity is The maximum concentration decay rate is In the predetermined formula for calculating the flocculation index , and The values ​​are 0.45, 0.35, and 0.2, respectively. Substituting these numbers into the flocculation index calculation formula, we can obtain... This indicates that the current state is one of insufficient flocculation. The computing device can send a signal to the feeding device to control the feeding device to increase the proportion of filter aid added, thereby improving the flocculation index. In this embodiment, the filter aid can be implemented as fiber powder.

[0055] If the influent flow rate of the water treatment system suddenly decreases, it will lead to uneven mixing of the filter aid and exchange resin, causing a sharp increase in the median and maximum particle size of the flocs. This will further affect parameters such as settling velocity and concentration decay rate, resulting in a flocculation index below 0.6. The computing device can send control signals to the membrane-laying pump and the holding pump of the water treatment system to increase the influent flow rate, accelerate the mixing of the filter aid and exchange resin, and simultaneously send an alarm signal to the host computer of the water treatment system until the flocculation index returns to normal.

[0056] The final film-laying effect achieved by the method provided in this embodiment is as follows: Figure 3 As shown in the image on the right, the membrane laying effect of the traditional visual settlement method is as follows: Figure 3 As shown on the left, it can be seen that the membrane filter layer obtained by this method is more uniform and flat, and the filter layer is also more dense than that obtained by visual sedimentation method.

[0057] This disclosure also provides a device for detecting flocculation effect, such as Figure 4 As shown, it includes: The identification module 401 is used to analyze the continuously acquired liquid images of the target liquid using image analysis technology to identify flocs in the target liquid; and to obtain the floc concentration parameters in the target liquid using particle detection technology. The first determining module 402 is used to determine the flocculant characteristic parameters of the liquid based on the identified flocculants and flocculant concentration parameters. The second determining module 403 is used to determine the flocculation index of the current liquid based on the characteristic parameters of the flocculants and using a pre-determined flocculation index algorithm.

[0058] In another embodiment provided in this disclosure, the first determining module 402 is further configured to determine the particle size of each floc, and based on the particle size of each floc, determine the median particle size of the floc in the liquid; based on the changes of each floc in the continuous liquid image, determine the settling velocity of each floc; and based on the settling velocity of each floc, determine the average settling velocity of the floc in the liquid; determine the concentration decay rate of the floc in the liquid according to the changes of the floc concentration parameter within a preset time period; and determine the median particle size, concentration decay rate, and average settling velocity as floc characteristic parameters.

[0059] In another embodiment provided in this disclosure, the second determining module 403 is further configured to determine that each floc characteristic parameter is normalized; and to substitute the normalized floc characteristic parameter into the flocculation index calculation formula to obtain the flocculation index; wherein the flocculation index calculation formula is a predetermined function relationship with each floc characteristic parameter as the independent variable and the flocculation index as the dependent variable.

[0060] In yet another embodiment provided in this disclosure, such as Figure 4 As shown, the flocculation effect testing device also includes: The comparison module 404 is used to compare the flocculation index with a preset threshold to determine the flocculation state of the liquid. The adjustment module 405 is used to adjust the feeding situation of the liquid according to the flocculation state.

[0061] In another embodiment provided in this disclosure, the comparison module 404 is further configured to determine that the liquid is in a normal flocculation state when the flocculation index is greater than a first threshold; determine that the liquid is in a state of insufficient flocculation when the flocculation index is greater than a second threshold and less than the first threshold; and determine that the liquid is in a state of abnormal flocculation when the flocculation index is less than the second threshold.

[0062] In another embodiment provided in this disclosure, the adjustment module 405 is further configured to: not adjust the feed ratio of the current liquid in a normal flocculation state; increase the proportion of filter aid in the current liquid in a state of insufficient flocculation; and adjust the feed amount of the current liquid in a state of abnormal flocculation.

[0063] In another embodiment provided in this disclosure, the identification module 401 is further configured to continuously acquire the current liquid image and preprocess the liquid image to obtain a grayscale image for image recognition; based on the grayscale image, the floc boundaries in the grayscale image are identified by a detection operator; and the floc boundaries are analyzed for connectivity to divide each independent floc region in the continuous image.

[0064] This disclosure also provides a flocculation effect testing device, including: a camera device, a main control device, and a communication device; The camera device is used to continuously acquire liquid images of the target liquid in the water treatment system and output the liquid images to the main control device. The main control unit is used to analyze continuously acquired liquid images of the target liquid using image analysis technology to identify flocs in the target liquid; and to obtain floc concentration parameters in the target liquid using particle detection technology; based on the identified flocs and floc concentration parameters, to determine the floc characteristic parameters of the liquid; and based on the floc characteristic parameters, to determine the floc index of the current liquid using a pre-determined floc index algorithm. A communication device is used to transmit floc characteristic parameters and flocculation index to a host computer.

[0065] Through the above description of the embodiments, those skilled in the art can clearly understand that the embodiments of this disclosure can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions of the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of this disclosure.

[0066] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes in the drawings are not necessarily essential for implementing this disclosure.

[0067] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.

[0068] The sequence numbers of the embodiments disclosed above are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0069] Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from its spirit and scope. Therefore, if such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, this disclosure is also intended to include such modifications and variations.

Claims

1. A method for detecting flocculation effect, characterized in that, include: Image analysis technology is used to analyze continuously acquired liquid images of the target liquid to identify flocs in the target liquid; and particle detection technology is used to obtain the concentration parameters of flocs in the target liquid. Based on the identified flocs and the floc concentration parameters, the floc characteristic parameters of the liquid are determined. Based on the floc characteristic parameters, the flocculation index of the current liquid is determined using a pre-determined flocculation index algorithm.

2. The method as described in claim 1, characterized in that, The determination of the flocculant characteristic parameters of the liquid based on the identified flocculants and the flocculant concentration parameters includes: Determine the particle size of each floc, and based on the particle size of each floc, determine the median particle size of the flocs in the liquid; Based on the changes of each floc in continuous liquid images, the settling velocity of each floc is determined; and based on the settling velocity of each floc, the average settling velocity of the flocs in the liquid is determined. The concentration decay rate of flocs in the liquid is determined based on the changes in floc concentration parameters within a preset time period. The median particle size, the concentration decay rate, and the average settling velocity are determined as the floc characteristic parameters.

3. The method as described in claim 1, characterized in that, The step of determining the flocculation index of the current liquid based on the floc characteristic parameters and using a pre-determined flocculation index algorithm includes: We determined that each floc characteristic parameter should be normalized separately. The normalized floc characteristic parameters are substituted into the flocculation index calculation formula to obtain the flocculation index; wherein the flocculation index calculation formula is a predetermined function with each floc characteristic parameter as the independent variable and the flocculation index as the dependent variable.

4. The method as described in claim 3, characterized in that, The flocculation index is determined using the following method: Construct a functional relationship determination model based on a neural network model; Multiple sets of sample data containing flocculant characteristic parameters are used as input, and the flocculation index corresponding to each set of sample data is used as the true value. Through the learning process of the neural network, the functional relationship is adjusted to determine the network parameters of the model. Based on the network parameters, the weight coefficient of each floc characteristic parameter in the flocculation index calculation formula is determined, thus obtaining the flocculation index calculation formula.

5. The method as described in claim 1, characterized in that, Also includes: The flocculation index is compared with a preset threshold to determine the flocculation state of the liquid. Adjust the feeding situation for the liquid according to the flocculation state.

6. The method as described in claim 5, characterized in that, The step of comparing the flocculation index with a preset threshold to determine the flocculation state of the liquid includes: If the flocculation index is greater than a first threshold, the liquid is determined to be in a normal flocculation state. If the flocculation index is greater than the second threshold and less than the first threshold, the liquid is determined to be in a state of insufficient flocculation. If the flocculation index is less than the second threshold, the liquid is determined to be in an abnormal flocculation state.

7. The method as described in claim 6, characterized in that, Adjusting the feed dosage for the liquid based on the flocculation state includes: For the normal flocculation state, the feed ratio for the current liquid is not adjusted; To address the insufficient flocculation, increase the proportion of filter aid for the current liquid. In response to the aforementioned abnormal flocculation state, adjust the feed rate for the current liquid.

8. The method as described in claim 1, characterized in that, The step of analyzing continuously acquired liquid images of the target liquid using image analysis technology to identify flocculants in the target liquid includes: Continuously acquire liquid images of the target liquid, and preprocess the liquid images to obtain grayscale images for image recognition; Based on the grayscale image, the floc boundaries in the grayscale image are identified by a detection operator. Connectivity analysis was performed on the identified floc boundaries to delineate each independent floc region in continuous images.

9. A flocculation effect testing device, characterized in that, include: The identification module is used to analyze continuously acquired liquid images of the target liquid using image analysis technology to identify flocs in the target liquid; and to obtain floc concentration parameters in the target liquid using particle detection technology. The first determining module is used to determine the flocculant characteristic parameters of the liquid based on the identified flocculants and the flocculant concentration parameters. The second determining module is used to determine the flocculation index of the current liquid based on the flocculant characteristic parameters and using a pre-determined flocculation index algorithm.

10. A flocculation effect testing device, characterized in that, include: Camera equipment, main control unit, and communication equipment; The camera device is used to continuously acquire liquid images of the target liquid in the water treatment system and output the liquid images to the main control device. The main control device is used to analyze continuously acquired liquid images of the target liquid using image analysis technology to identify flocs in the target liquid; and to obtain floc concentration parameters in the target liquid using particle detection technology; based on the identified flocs and the floc concentration parameters, to determine the floc characteristic parameters of the liquid; and based on the floc characteristic parameters, to determine the floc index of the current liquid using a pre-determined floc index algorithm. The communication device is used to transmit floc characteristic parameters and flocculation index to the host computer.

Images