Intelligent dosing system and method for chlorine dioxide disinfectant in water supply plant

Abstract

The present invention discloses an intelligent dosing system and method for chlorine dioxide disinfectant in a water supply plant. By constructing a chlorine dioxide concentration prediction model for the clear water tank outlet and the terminal point of the pipe network in the water supply system, the chlorine dioxide concentrations at these two nodes can be accurately predicted based on the current working conditions of the water supply system, and the minimum dosage that can ensure that the concentration parameters of the disinfectant at the terminal of the pipe network meet the standards is solved. The system provides theoretical guidance and basis for the disinfectant dosage in the water supply plant, and adjusts the dosage of the dosing equipment in real time to achieve intelligent and precise dosage, thereby reducing the risk of microbial growth and the risk of excessive disinfection by-products in the water supply system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water treatment, and particularly relates to a small water supply plant chlorine dioxide disinfectant intelligent adding system. BACKGROUND

[0002] Chlorine dioxide is a high-efficiency and broad-spectrum disinfectant and is widely used in drinking water production. Although compared with other chlorine-based disinfectants, chlorine dioxide produces less organic by-products in the use process, but it produces inorganic by-products mainly in the form of chlorite. The traditional chlorine dioxide adding of the water supply plant is in the mode of manual control, and the adding amount cannot be adjusted in real time according to the working condition of the pipe network system, so that the disinfection effect is not ideal. When the adding amount is too low, there is a risk of breeding of pathogenic microorganisms at the end of the pipe network, which brings health and safety hazards to the users; and blindly increasing the adding amount of the disinfectant of the water supply plant or adding the disinfectant again in the pipe network brings the risk of exceeding the standard of disinfection by-products.

[0003] Therefore, it is of great significance to design an intelligent disinfection system capable of accurately adding chlorine dioxide to improve the water supply quality and ensure the water safety of the people. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a chlorine dioxide disinfectant intelligent adding system suitable for a small water supply plant, which uses the chlorine dioxide concentration at the outlet of a clear water pool and the end of a pipe network to predict the minimum adding amount to ensure real-time and accurate adding.

[0005] To achieve the above purpose, the present application provides the following technical scheme:

[0006] The chlorine dioxide disinfectant intelligent adding system for the water supply plant provided by the present application comprises a chlorine dioxide online monitoring device, an intelligent management platform and a dosing device.

[0007] The chlorine dioxide online monitoring device is arranged at the water outlet end and is used to obtain the chlorine dioxide concentration at the water outlet end and the water temperature at the end of the pipe network, and the collected data is used to construct a chlorine dioxide concentration prediction model and to control the chlorine adding amount in real time.

[0008] The intelligent management platform calculates the adding amount required by the front chlorine adding device and the rear chlorine adding device respectively through the chlorine dioxide concentration prediction model, and controls the dosing device to add the reagent.

[0009] The dosing device comprises a front chlorine adding device arranged at the front end of the clear water pool and a rear chlorine adding device arranged at the rear end of the clear water pool, the front end of the clear water pool is the inlet of the clear water pool, and the rear end of the clear water pool is a water collecting well, and the front chlorine adding and the rear chlorine adding are performed respectively.

[0010] Further, the chlorine dioxide concentration prediction model solves the required dosing amount according to the current water temperature, taking the chlorine dioxide concentration at the outlet of the clear water tank as the target concentration.

[0011] Further, the difference between the minimum chlorine dioxide concentration of the factory water output by the intelligent management platform and the chlorine dioxide concentration value at the outlet of the clear water tank after chlorination by the front chlorination equipment is output as the required dosing amount of the rear chlorination equipment.

[0012] Further, the intelligent management platform includes a clear water tank outlet chlorine dioxide concentration prediction model, a clear water tank dosing amount optimization model, a pipe network terminal chlorine dioxide concentration prediction model, a factory water chlorine dioxide concentration optimization model, and a dosing control module.

[0013] Further, the clear water tank outlet chlorine dioxide concentration prediction model is constructed according to the following steps:

[0014] Obtain the parameters of the water supply system;

[0015] Construct a clear water tank hydraulic model, which includes a first node, a second node, a third node, and a fourth node; the first node is used to set the concentration of the chlorine dioxide disinfectant input into the clear water tank; the second node is used to control the water inlet of the clear water tank; the third node is used to set the size parameters and mixing model of the clear water tank; and the fourth node is used to set the water consumption at the outlet of the clear water tank.

[0016] Determine the chlorine dioxide decay coefficient;

[0017] Construct a water quality prediction model, input the chlorine dioxide decay coefficient and the clear water tank inlet chlorine dioxide dosing amount into the clear water tank hydraulic model to calculate the chlorine dioxide concentration prediction value at the outlet of the clear water tank.

[0018] Further, the clear water tank dosing amount optimization model is calculated by the clear water tank outlet chlorine dioxide concentration prediction model to obtain the minimum dosing amount that meets a certain target concentration at the outlet of the clear water tank, and the calculation is performed according to the following steps:

[0019] First, input the chlorine dioxide decay coefficient corresponding to the current water temperature into EPANET;

[0020] Then, input the chlorine dioxide dosing concentration in the range of 0-1 mg / L into the EPANET model in ascending order for a certain length of water quality simulation;

[0021] Determine whether the chlorine dioxide concentration at the outlet of the clear water tank is higher than the target concentration;

[0022] Output the minimum dosing amount that meets the requirements;

[0023] Further, the pipe network tail end chlorine dioxide concentration prediction model is an LSTM neural network model based on a Self Attention mechanism, and the chlorine dioxide concentration at the pipe network tail end is predicted according to the chlorine dioxide concentration of the factory water and the water temperature data at the pipe network tail end.

[0024] Further, the factory water chlorine dioxide concentration optimization model is specifically performed according to the following steps:

[0025] At the beginning, the factory water chlorine dioxide concentration x is set as the outlet target concentration;

[0026] It is judged whether the factory water chlorine dioxide concentration satisfies the following relationship: x < Th1, if not, the factory water chlorine dioxide concentration is output; if yes, the next step is entered; the Th1 represents a factory water chlorine dioxide preset concentration value;

[0027] The prediction result is calculated by the pipe network tail end chlorine dioxide concentration prediction model;

[0028] The tail end chlorine dioxide concentration prediction result y is output;

[0029] It is judged whether the prediction result satisfies the following relationship: y >= Th2, if not, the factory water chlorine dioxide concentration is changed; if yes, the factory water chlorine dioxide concentration is output, and then the process is ended; the Th2 represents a preset prediction concentration value.

[0030] The water supply plant chlorine dioxide disinfectant intelligent adding method provided by the application comprises the following steps:

[0031] The chlorine dioxide concentration of the clear water pool outlet end and the water temperature of the pipe network tail end are acquired, the outlet end comprises a clear water pool outlet, a water plant outlet and a pipe network tail end, and the collected data is used to construct a chlorine dioxide concentration prediction model and to control the chlorine adding amount in real time;

[0032] The chlorine dioxide concentration prediction model is constructed, the adding amount required by the front chlorine adding equipment and the rear chlorine adding equipment under the current working condition is calculated by the chlorine dioxide concentration prediction model, and the dosing equipment is controlled to automatically add;

[0033] The dosing equipment comprises the front chlorine adding equipment arranged at the front end of the clear water pool and the rear chlorine adding equipment arranged at the rear end of the clear water pool, the front end of the clear water pool is a clear water pool inlet, and the rear end of the clear water pool is a water collecting well, that is, the front chlorine adding and the rear chlorine adding are respectively performed at the clear water pool inlet and the water collecting well.

[0034] Further, the adding amount required by the rear chlorine adding equipment is calculated in the following manner: the minimum value of the factory water chlorine dioxide concentration is acquired, the difference between the minimum value of the factory water chlorine dioxide concentration and the chlorine dioxide concentration value of the clear water pool outlet after chlorine adding by the front chlorine adding equipment is calculated, and the difference is the adding amount required by the rear chlorine adding equipment.

[0035] The present application has the advantages that:

[0036] The water supply plant chlorine dioxide disinfectant intelligent adding system and method provided by the present application can accurately predict the chlorine dioxide concentrations of the two nodes based on the current working condition of the water supply system by constructing a chlorine dioxide concentration prediction model of the clear water pool outlet and the pipe network terminal point in the water supply system, and solve the minimum dosing amount that can ensure that the pipe network terminal disinfectant concentration parameter meets the standard, thereby providing theoretical guidance and basis for water supply plant disinfectant dosing, and adjusting the dosing amount of the dosing equipment in real time to realize intelligent and accurate dosing and reduce the risk of microbial breeding and the risk of disinfection by-product exceeding the standard in the water supply system.

[0037] Other advantages, objects, and features of the present application will be in part apparent and in part pointed out hereinafter in the specification, and will be observed by variations now being suggested or hereafter devised, according to the principles of the present application. The objects and other advantages of the present application will be realized and attained by the structure particularly pointed out in the specification. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to make the purposes, technical solutions and beneficial effects of the present application clearer, the present application provides the following drawings for illustration.

[0039] Figure 1 is a clear water pool dosing amount optimization system block diagram.

[0040] Figure 2 is an EPANET clear water pool hydraulic model topological structure diagram.

[0041] Figure 3 is a Self Attention-LSTM neural network structure schematic diagram.

[0042] Figure 4 is a comparison diagram of the prediction results and the true values of the actual pipe network terminal chlorine dioxide concentration prediction model.

[0043] Figure 5 is a clear water pool dosing amount optimization model flowchart.

[0044] Figure 6 is a chlorine dioxide concentration optimization model program block diagram of finished water.

[0045] Figure 7 is a real water quality data diagram in actual engineering.

[0046] Figure 8 is a simulation result of the water quality prediction module after model optimization. DETAILED DESCRIPTION

[0047] The application will be further described in connection with the drawings and specific embodiments so that those skilled in the art can better understand the application and implement it, but the embodiments are not limiting to the application.

[0048] Embodiment 1

[0049] As Figure 1 shown, Figure 1 is a water tank dosing amount optimization system block diagram, the waterworks chlorine dioxide intelligent dosing system provided by the embodiment includes a chlorine dioxide online monitoring device, a dosing device and an intelligent management platform.

[0050] The chlorine dioxide online monitoring device is arranged at the water outlet end of the water tank and is used to obtain the chlorine dioxide concentration of the water outlet end and the water temperature of the pipe network terminal, the water outlet end includes a water tank outlet, a waterworks outlet and a pipe network terminal, the chlorine dioxide concentrations of the three points and the water temperature of the pipe network terminal are monitored respectively, and the collected data is used to construct a chlorine dioxide concentration prediction model and to control the chlorine dosage in real time.

[0051] The intelligent management platform calculates by using the constructed chlorine dioxide concentration prediction model, uses an optimization algorithm to solve the required dosing amount of the front chlorine dosing device and the rear chlorine dosing device under the current working condition, and controls the dosing device to automatically dose.

[0052] The dosing device includes a front chlorine dosing device arranged at the front end of the water tank and a rear chlorine dosing device arranged at the rear end of the water tank, the front end of the water tank is a water tank inlet, and the rear end of the water tank is a water collecting well, that is, the front chlorine dosing and the rear chlorine dosing are respectively performed at the water tank inlet and the water collecting well.

[0053] In the actual operation process, the chlorine dioxide concentration prediction model of the water tank solves the corresponding dosing amount according to the current water temperature, with the chlorine dioxide concentration of 0.02 mg / L at the water outlet of the water tank as the target concentration.

[0054] The difference between the minimum chlorine dioxide concentration of the finished water output by the intelligent management platform and the chlorine dioxide concentration value of the water tank outlet after the chlorine dosing by the front chlorine dosing device is used as the required dosing amount of the rear chlorine dosing device.

[0055] The medicament dosed by the front chlorine dosing device is to inactivate pathogenic microorganisms including bacteria and viruses in the water, so as to take a disinfection means to ensure the safety of drinking water, and the medicament dosing amount of the front chlorine dosing device satisfies the consumption of the contact reaction in the water tank and the attenuation within the residence time.

[0056] The medicament added by the post-chlorination equipment is mainly to maintain the residual disinfectant concentration of drinking water in the water supply network, so that the water can maintain a certain continuous disinfection capacity during distribution, kill microorganisms in the water, prevent water quality from being affected by external pollution, and ensure the biological safety of water quality. The medicament addition amount of the post-chlorination equipment meets the consumption required by drinking water in the pipe network. The post-chlorination equipment in the embodiment is a closed-loop control link of the entire dosing system, and realizes precise control of the medicament addition amount.

[0057] The water supply plant chlorine dioxide intelligent addition system in the embodiment is specifically performed according to the following steps:

[0058] The target concentration of the water outlet is input to the intelligent management platform;

[0059] The intelligent management platform controls the pre-chlorination equipment to add medicament before the clear water tank;

[0060] The chlorine concentration at the outlet of the clear water tank is obtained through chlorine dioxide online monitoring and input to the intelligent management platform;

[0061] The intelligent management platform controls the medicament addition amount of the post-chlorination equipment to the water collecting well, and then the medicament-containing water in the water collecting well is mixed and lifted by the secondary pump station to discharge the water plant, so as to control the chlorine concentration at the outlet of the water plant;

[0062] The intelligent management platform controls the medicament addition amount of the pre-chlorination equipment and the post-chlorination equipment according to the target concentration of the clear water tank outlet, the reference value of the water concentration, the chlorine concentration value obtained by the chlorine dioxide online monitoring at the outlet of the water plant, the chlorine concentration and water temperature obtained by the chlorine dioxide and water temperature online monitoring at the end of the pipe network;

[0063] The intelligent management platform includes a clear water tank outlet chlorine dioxide concentration prediction model, a clear water tank medicament addition amount optimization model, a pipe network end chlorine dioxide concentration prediction model, a water plant outlet chlorine dioxide concentration optimization model, and a medicament control module;

[0064] The clear water tank outlet chlorine dioxide concentration prediction model is used to predict the chlorine dioxide concentration at the outlet of the clear water tank under the condition that the water temperature is known;

[0065] The clear water tank medicament addition amount optimization model is used to solve the minimum medicament addition amount that meets a certain target concentration at the outlet of the clear water tank;

[0066] The pipe network end chlorine dioxide concentration prediction model is used to predict the pipe network end chlorine dioxide concentration according to the water plant outlet chlorine dioxide concentration and the water temperature data at the end of the pipe network;

[0067] The water plant outlet chlorine dioxide concentration optimization model is used to solve the minimum value of the water plant outlet chlorine dioxide concentration that can ensure that the pipe network end chlorine dioxide concentration meets the standard under the current pipe network working condition;

[0068] The drug control module is used for controlling the dosing amount of the two points according to the solution result of the optimization algorithm.

[0069] The dosing equipment in the embodiment includes a chlorine dioxide generator and a dosing pump.

[0070] The chlorine dioxide generator is used for preparing high-concentration chlorine dioxide as a disinfectant for disinfection; and the dosing pump is used for dosing the disinfectant into water quantitatively.

[0071] The chlorine dioxide online monitoring equipment in the embodiment is used for monitoring water quality.

[0072] The clean water tank dosing amount optimization model in the embodiment is calculated by using a clean water tank dosing amount optimization algorithm.

[0073] The chlorine dioxide concentration optimization model for finished water in the embodiment is calculated by using a chlorine dioxide concentration optimization algorithm for finished water.

[0074] The chlorine dioxide concentration prediction model for the clean water tank outlet is a mechanism model developed based on an EPANET 2.0 software platform, and the specific construction steps include:

[0075] 1. Obtain the parameters of the water supply system, that is, collect the basic data: investigate the parameters of the water supply system, including the structural size parameters of the clean water tank, the pipe diameter and length of the inlet and outlet, the drinking water treatment process control logic, and the water consumption of each hour within 24 hours in the water supply area;

[0076] 2. Hydraulic model construction: a model as shown in Figure 2 is constructed in the EPANET platform, Figure 2 which is a topological structure diagram of the EPANET clean water tank hydraulic model; the hydraulic model includes a first node, a second node, a third node, and a fourth node.

[0077] The first node 1 represents the water treatment process before the clean water tank, and the initial water quality size of the node can represent the concentration of the chlorine dioxide disinfectant input into the clean water tank;

[0078] The second node 2 is a water pump, and by changing the on-off state thereof, the opening and closing of the water treatment process can be simulated, so as to control the water inlet of the clean water tank, and the drinking water treatment process control logic in the basic data is input into the EPANET.

[0079] The third node 3 is the clean water tank, and the size parameters of the node are assigned according to the collected basic data, the mixing model of the clean water tank is set as a FIFO plug flow mixing model, and the reaction order is first order.

[0080] The fourth node 4 is the outlet of the clean water tank, and the water consumption of each hour in 24 hours is input into the node, and the water demand size is the size of the clean water tank outlet flow;

[0081] 3. Determine the chlorine dioxide decay coefficient by beaker experiment:

[0082] Take 3 portions of untreated process effluent and pour them into 500m brown glass bottles, add 1.0mg / L concentration of chlorine dioxide disinfectant and stir evenly, place in a thermostat, set the thermostat temperature to 25℃, 30℃, 35℃ respectively; and measure the residual chlorine dioxide concentration value C at 0h, 2h, 4h, 6h, 8h, 10h, 12h, 24h, 36h, 48h, 60h, 72h; each measurement is repeated 3 times; the concentration time sequence of each water temperature is made into ln(C t / C0)~t curve, and the slope K d of the curve is the chlorine dioxide decay coefficient, i.e. the decay rate constant at this water temperature; according to the Arrhenius model, the reaction rate constant K b and the Kelvin temperature T are related as follows:

[0083]

[0084] In the formula, A represents the pre-factor, also known as the Arrhenius constant; R represents the molar gas constant; E a represents the experimental activation energy, which can be regarded as a constant independent of temperature;

[0085] 4. Water quality prediction model construction: according to the mathematical expression obtained from the Arrhenius model in the previous step, the corresponding chlorine dioxide decay coefficient can be solved under the condition of known water temperature, and the decay coefficient and the chlorine dioxide dosage at the inlet of the clean water tank are input into the clean water tank hydraulic model, so that the change of chlorine dioxide concentration at the outlet of the clean water tank can be obtained.

[0086] The chlorine dioxide dosage optimization model in this embodiment is an algorithm developed based on the chlorine dioxide concentration prediction model at the outlet of the clean water tank, which can solve the minimum dosage that meets a certain target concentration at the outlet of the clean water tank, and the solving process is:

[0087] First, input the chlorine dioxide decay coefficient corresponding to the current water temperature into EPANET;

[0088] Then input the chlorine dioxide dosage concentration in the range of 0~1mg / L into EPANET model in the order from small to large for a certain length of time water quality simulation;

[0089] And determine whether the chlorine dioxide concentration at the outlet is higher than the target concentration;

[0090] Finally output the minimum dosage that meets the requirements;

[0091] The pipe network terminal chlorine dioxide concentration prediction model in this embodiment is an LSTM neural network model constructed based on a Self Attention mechanism. The chlorine dioxide concentration at the pipe network terminal is predicted according to the chlorine dioxide concentration of the factory water and the water temperature data at the pipe network terminal.

[0092] As shown in Figure 3 , Figure 3 is a Self Attention-LSTM neural network structure diagram. The Self Attention-LSTM neural network structure is constructed and optimized through the Self Attention mechanism. The specific process is as follows:

[0093] Suppose m indexes are used as features. The data at time i can be represented by the vector X i =[x1,x2,x3,...,x m ]. If the size of the moving window is n, the data point at i+1 is to be predicted.

[0094] Then the n x m matrix [X i-n+1 ,X i-n+2 ,X i-n+3 ,...,X i ] of data is transmitted to the input layer.

[0095] After that, it enters the LSTM hidden layer. The hidden layer outputs a weight vector H = [h1, h2, h3,..., h n ] after learning.

[0096] The weight vector H = [h1, h2, h3,..., h n ] output by the hidden layer is input into the Self Attention mechanism layer for weight redistribution.

[0097] Self Attention mechanism is a technique used to capture relationships between different parts of an input sequence.

[0098] It determines which parts are more important in a given task by calculating the attention weights between each element and other elements. This mechanism enables the model to focus on the most relevant information in the input sequence.

[0099] Self Attention mechanism helps better understand the dependency relationships and patterns of input data in time series data prediction.

[0100] For example, some data points can be more critical to predict future values, while other data points can be relatively less important, and the Self Attention mechanism can help the model automatically identify these important data points. In this process, the features with high correlation will be strengthened to obtain a new weight vector A = [a1, a2, a3,..., a n ]. Finally, the dimension of the vector H is transformed by the fully connected layer to obtain the predicted value X i+1 .

[0101] As shown in Figure 4 , Figure 4 is a comparison chart of the predicted results of the actual pipe network terminal chlorine dioxide concentration prediction model and the true value.

[0102] The model uses the Sigmod function as the activation function of the LSTM unit and the ReLU function as the activation function of the output layer.

[0103] During the training phase of the model, after each batch of training data is input into the model, the forward propagation outputs the predicted value, and then the loss function calculates the error between the predicted value output by the model and the true value, i.e. the loss value.

[0104] Then, according to the loss value, the model updates the parameters in the neural network model using backpropagation to reduce the loss between the true value and the predicted value, so that the predicted value output by the model continuously approaches the true value, which is the process of model learning.

[0105] The specific construction steps are as follows:

[0106] 1. Installation of equipment and data collection: online monitoring equipment is installed at the outlet of the water plant and the terminal point of the pipe network, respectively, to collect chlorine dioxide concentration data and water temperature data at the outlet of the water plant and the terminal point of the pipe network.

[0107] 2. Data preprocessing: the process includes identification and replacement of outliers, data denoising, and data normalization.

[0108] 3. Data set division: the data set is divided into training set, validation set and test set in the order of 64%, 16% and 20% proportion. The training set is used to train the neural network model, and then the validation set is used to evaluate the effectiveness of the model to select the best model.

[0109] 4. Model training: input the data set into the Self Attention-LSTM neural network model, adjust the sliding window size, the number of hidden layer neurons, the training batch size and other hyperparameters of the model, compare and select the combination with better performance parameters, and complete the determination of the model hyperparameters.

[0110] As shown in Figure 5 ,Figure 5 The figure is a flow chart of the optimization model for the dosage of clear water tank, which is specifically carried out in the following steps:

[0111] At the beginning, set the chlorine dioxide dosage x of the clear water tank: 0mg / L. The chlorine dioxide control target C at the outlet of the clear water tank and the chlorine dioxide decay rate constant K b ;

[0112] Determine whether the amount of chlorine dioxide added to the clear water tank satisfies the following relationship: x<Th 投 If no, then output the chlorine dioxide dosage; if yes, then go to the next step;

[0113] In this embodiment, Th 投 Set to 1 mg / L; (Th 投 The value can be 0.5~1mg / L according to the actual project)

[0114] The prediction results are calculated using the clear water tank chlorine dioxide concentration prediction model;

[0115] Output the predicted result y of chlorine dioxide concentration at the outlet of the clear water tank;

[0116] Determine whether the prediction result satisfies the following relationship: y ≥ C. If not, increase the chlorine dioxide dosage by a preset value; if yes, output the chlorine dioxide dosage and then end.

[0117] Among them, C is the chlorine dioxide control target at the outlet of the clear water tank;

[0118] The preset value can be set to 0.01 mg / L; (depending on the dosage accuracy of the water plant, it can be set to 0.005-0.1 mg / L)

[0119] like Figure 6 As shown, Figure 6 This is a flowchart of the chlorine dioxide concentration optimization model for factory water. The chlorine dioxide concentration optimization model for factory water is specifically carried out in the following steps:

[0120] Initially, the chlorine dioxide concentration x of the factory water is set as the target concentration at the outlet. In this example, the target concentration at the outlet is set to 0.1 mg / L. Other parameters include the current monitoring value.

[0121] Determine whether the chlorine dioxide concentration of the factory water satisfies the following relationship: x < Th1. If not, output the chlorine dioxide concentration of the factory water; if yes, proceed to the next step; Th1 represents the preset concentration value of chlorine dioxide in the factory water; in this embodiment, Th1 is set to 0.8 mg / L;

[0122] The prediction results are calculated using the network terminal chlorine dioxide concentration prediction model (Attention-LSTM prediction model);

[0123] Output the predicted result y of the terminal chlorine dioxide concentration;

[0124] Determine whether the prediction result satisfies the following relationship: y ≥ Th2. If not, adjust the chlorine dioxide concentration of the factory water. In this example, the chlorine dioxide concentration of the factory water is increased by 0.01 mg / L. If yes, output the chlorine dioxide concentration of the factory water and terminate. Th2 represents the preset predicted concentration value; in this example, Th2 is set to 0.2 mg / L.

[0125] like Figure 7 As shown, Figure 7 This is a graph of actual water quality data from the water plant outlet and the end of the pipe network on a certain day in an actual project. The upper blue curve represents the residual chlorine concentration in the water plant outlet, and the lower red curve represents the residual chlorine concentration in the water at the end of the pipe network. According to the characteristics of the change in chlorine dioxide concentration, the water quality data of the pipe network on this day can be divided into two time periods. In time period 1, the chlorine dioxide dosage was low, and the chlorine dioxide concentration of the factory water and the terminal chlorine dioxide concentration generally did not meet the standards; after the factory water concentration was increased in time period 2, the terminal chlorine dioxide concentration also increased, with the highest value approaching 0.16 mg / L, far exceeding the specified value of 0.02 mg / L.

[0126] like Figure 8 As shown, Figure 8 This is the simulation result of the water quality prediction module after model optimization. The water quality prediction module can adjust the chlorine dioxide concentration of the factory water in time according to the current working conditions of the pipe network, and stably control the terminal chlorine dioxide concentration at around 0.02 mg / L. The terminal water qualification rate has increased from 22.6% to 88.9%, greatly improving the accuracy of disinfectant addition.

[0127] The factory water chlorine dioxide concentration optimization model is an algorithm developed based on the pipeline network terminal chlorine dioxide concentration prediction model. It can solve the minimum factory water chlorine dioxide concentration that can ensure that the pipeline network terminal chlorine dioxide concentration meets the standard under the current pipeline network operating conditions. The solution process is:

[0128] Within the range of 0.1-0.8 mg / L, the chlorine dioxide concentration of the factory water is input into the SelfAttention-LSTM model from small to large for prediction until the chlorine dioxide concentration at the end of the pipeline network in the output prediction result is completely qualified. At this time, the input chlorine dioxide concentration of the factory water is the minimum value.

[0129] The dosing control module includes a front chlorination module located at the inlet of the clear water tank and a post chlorination module at the outlet of the clear water tank, and controls the dosing amount of the two points according to the solution of the optimization algorithm.

[0130] The embodiment constructs a chlorine dioxide concentration prediction model for the outlet of the clear water pool, the outlet of the water plant and the end point of the pipe network in the water supply system, can accurately predict the chlorine dioxide concentration of the two nodes of the clear water pool outlet and the end point of the pipe network based on the current working condition of the water supply system, and solve the minimum dosing amount that can ensure that the disinfectant concentration parameter of the end point of the pipe network meets the standard, thereby providing a theoretical guidance and basis for the disinfectant dosing of the water supply plant, and adjusting the dosing amount of the dosing equipment in real time, realizing intelligent and accurate dosing, and reducing the risk of microbial breeding and the risk of exceeding the standard of disinfection by-products in the water supply system.

[0131] The water supply plant in the embodiment is mainly used for small water plants, which refer to the small water plants that meet the provisions of the Standardized Upgrading Technical Regulations for Small Rural Water Supply Projects (SL / T 825-2024) issued by the Ministry of Water Resources.

[0132] Embodiment 2

[0133] The method for intelligent dosing of chlorine dioxide disinfectant in the water supply plant provided in the embodiment comprises the following steps:

[0134] The chlorine dioxide concentration of the outlet of the clear water pool and the water temperature of the end point of the pipe network are obtained, the outlet includes the outlet of the clear water pool, the outlet of the water plant and the end point of the pipe network, and the collected data is used to construct a chlorine dioxide concentration prediction model and control the chlorine dosing amount in real time;

[0135] The chlorine dioxide concentration prediction model is constructed, the dosing amount required by the front chlorine dosing equipment and the rear chlorine dosing equipment under the current working condition is calculated through the chlorine dioxide concentration prediction model, and the dosing equipment is controlled to automatically dose;

[0136] The dosing equipment includes the front chlorine dosing equipment arranged at the front end of the clear water pool and the rear chlorine dosing equipment arranged at the rear end of the clear water pool, the front end of the clear water pool is the inlet of the clear water pool, and the rear end of the clear water pool is the water collecting well, that is, the front chlorine dosing and the rear chlorine dosing are respectively performed at the inlet of the clear water pool and the water collecting well.

[0137] The dosing amount required by the rear chlorine dosing equipment is calculated as follows: the minimum chlorine dioxide concentration of the outlet water is obtained, the difference between the minimum chlorine dioxide concentration of the outlet water and the chlorine dioxide concentration value of the outlet of the clear water pool after chlorine dosing by the front chlorine dosing equipment is calculated, and the difference is the dosing amount required by the rear chlorine dosing equipment.

[0138] The above embodiments are only preferred embodiments for fully illustrating the present application, and the protection scope of the present application is not limited thereto. Equivalent substitutions or transformations made by those skilled in the art on the basis of the present application are within the protection scope of the present application. The protection scope of the present application is subject to the claims.

Claims

1. Intelligent dosing system of chlorine dioxide disinfectant in water supply plant, characterized by: Including chlorine dioxide online monitoring equipment, intelligent management platform, and dosing equipment; The chlorine dioxide online monitoring device is installed at the water outlet, which includes the outlet of the clear water tank, the outlet of the water plant, and the end of the pipe network, and is used to obtain the chlorine dioxide concentration at the water outlet and the water temperature at the end of the pipe network. The collected data is used to build a chlorine dioxide concentration prediction model and control the chlorine addition amount in real time; The intelligent management platform calculates the dosage required for the pre-chlorination equipment and the post-chlorination equipment respectively through the chlorine dioxide concentration prediction model, and controls the dosing equipment to add the dosage; The dosing equipment includes a front chlorination device arranged at the front end of the clear water tank and a rear chlorination device arranged at the rear end of the clear water tank. The front end of the clear water tank is the clear water tank entrance, and the rear end of the clear water tank is the water collection well, which respectively perform front chlorination and rear chlorination. The intelligent management platform includes a chlorine dioxide concentration prediction model at the outlet of the clear water tank, a clear water tank dosage optimization model, a chlorine dioxide concentration prediction model at the end of the pipe network, a chlorine dioxide concentration optimization model for the factory water, and a dosage control module; The chlorine dioxide concentration prediction model at the outlet of the clear water tank is used to predict the chlorine dioxide concentration at the outlet of the clear water tank under the condition of known water temperature; The clear water tank dosage optimization model is used to solve the minimum dosage that meets a certain target concentration at the outlet of the clear water tank; The chlorine dioxide concentration prediction model at the end of the pipe network is used to predict the chlorine dioxide concentration at the end of the pipe network based on the chlorine dioxide concentration of the outgoing water and the water temperature data at the end of the pipe network; The chlorine dioxide concentration optimization model for the outlet water is used to solve the minimum chlorine dioxide concentration of the outlet water that can ensure that the chlorine dioxide concentration at the end of the pipeline network meets the standard under the current pipeline network operating conditions; The dosing control module is used to control the dosage of the two points according to the solution of the optimization algorithm.

2. The intelligent chlorine dioxide disinfectant dosing system for a water supply plant according to claim 1, characterized in that: The intelligent management platform uses a chlorine dioxide concentration prediction model to calculate the dosage required for the pre-chlorination equipment based on the current water temperature and the chlorine dioxide concentration at the outlet of the clear water tank as the target concentration.

3. The intelligent chlorine dioxide disinfectant dosing system for a water supply plant according to claim 1, characterized in that: The difference between the minimum chlorine dioxide concentration of the factory water output by the intelligent management platform and the chlorine dioxide concentration value at the outlet of the clear water tank after chlorination by the front chlorination equipment is used as the dosage required for the post-chlorination equipment.

4. The intelligent chlorine dioxide disinfectant dosing system for a water supply plant according to claim 1, characterized in that: The chlorine dioxide concentration prediction model at the outlet of the clear water tank is constructed according to the following steps: Get the parameters of the water supply system; Constructing a hydraulic model of a clear water tank, the hydraulic model comprising a first node, a second node, a third node, and a fourth node; the first node is used to set the concentration of a chlorine dioxide disinfectant fed into the clear water tank; the second node is used to control the water inlet into the clear water tank; the third node is used to set the size parameters and mixing model of the clear water tank; and the fourth node is used to set the water consumption at the outlet of the clear water tank; Determine the chlorine dioxide attenuation coefficient; A water quality prediction model was constructed, and the chlorine dioxide attenuation coefficient and the chlorine dioxide dosage at the clear water tank inlet were input into the clear water tank hydraulic model to calculate the predicted value of chlorine dioxide concentration at the clear water tank outlet.

5. The intelligent chlorine dioxide disinfectant dosing system for a water supply plant according to claim 1, characterized in that: The clean water tank dosage optimization model is to calculate the minimum dosage that meets a certain target concentration at the clean water tank outlet through the clean water tank outlet chlorine dioxide concentration prediction model, and the calculation is specifically performed according to the following steps: First, the chlorine dioxide attenuation coefficient corresponding to the current water temperature is input into EPANET; Then, the chlorine dioxide concentration was input into the EPANET model in ascending order within the range of 0 to 1 mg / L to simulate water quality for a certain period of time; Determine whether the chlorine dioxide concentration at the outlet of the clean water tank is higher than the target concentration; Output the minimum dosage required.

6. The intelligent chlorine dioxide disinfectant dosing system for a water supply plant according to claim 1, characterized in that: The chlorine dioxide concentration prediction model for the pipe network terminal is an LSTM neural network model built based on the Self-Attention mechanism. It predicts the chlorine dioxide concentration at the pipe network terminal based on the chlorine dioxide concentration of the factory water and the water temperature data at the pipe network terminal.

7. The intelligent chlorine dioxide disinfectant dosing system for a water supply plant according to claim 1, characterized in that: The chlorine dioxide concentration optimization model for the factory water is specifically carried out according to the following steps: At the beginning, the chlorine dioxide concentration x of the factory water is set as the target concentration at the outlet; Determine whether the chlorine dioxide concentration in the factory water satisfies the following relationship: x< If no, then output the chlorine dioxide concentration of the factory water; if yes, then go to the next step; Indicates the preset concentration value of chlorine dioxide in factory water; The prediction results are calculated by using the chlorine dioxide concentration prediction model at the end of the pipe network; Output the predicted result y of the terminal chlorine dioxide concentration; Determine whether the prediction result satisfies the following relationship: y≥ If not, change the chlorine dioxide concentration of the factory water; if yes, output the chlorine dioxide concentration of the factory water, and then end; Indicates the preset predicted concentration value.

8. Intelligent dosing method of chlorine dioxide disinfectant in water supply plants, characterized by: The following steps are involved: Obtain the chlorine dioxide concentration at the outlet and the water temperature at the end of the pipe network. The collected data is used to build a chlorine dioxide concentration prediction model and control the chlorine addition in real time. Construct a chlorine dioxide concentration prediction model, calculate the dosage required by the pre-chlorination equipment and post-chlorination equipment under the current working conditions through the chlorine dioxide concentration prediction model, and control the dosing equipment to automatically add chlorine; The dosing equipment includes a pre-chlorination device arranged at the front end of the clear water tank and a post-chlorination device arranged at the rear end of the clear water tank. The front end of the clear water tank is the clear water tank entrance, and the rear end of the clear water tank is a water collection well, that is, pre-chlorination and post-chlorination are carried out at the clear water tank entrance and the water collection well respectively.

9. The intelligent dosing method of chlorine dioxide disinfectant for a water supply plant according to claim 8, characterized in that: The dosage required for the post-chlorination equipment is calculated in the following manner: obtain the minimum chlorine dioxide concentration of the factory water, calculate the difference between the minimum chlorine dioxide concentration of the factory water and the chlorine dioxide concentration value at the outlet of the clear water tank after chlorination by the pre-chlorination equipment, and the difference is the dosage required for the post-chlorination equipment.

Images