Method and device for predicting quality of cotton dyeing effluent and application

Abstract

The invention provides a cotton dyeing effluent quality prediction method, a cotton dyeing effluent quality prediction device and application, belongs to the technical field of printing and dyeing, and aims to predict the cotton dyeing effluent quality and obtain effluent quality parameters of a plurality of working sections of a cotton dyeing process to be operated in a future period of time, so that the cotton dyeing effluent is conveniently classified and treated, and the quality of the cotton dyeing effluent is improved. The requirements of all process links on the water quality of recycled water are met, the treatment efficiency of the effluent of the cotton dyeing process is improved, and the recycling cost is reduced. According to the method, process parameters and dye parameters of a to-be-operated cotton dyeing process are input into an effluent quality prediction model, and the effluent quality prediction model outputs effluent quality parameters of the to-be-operated cotton dyeing process.

Description

Technical Field

[0001] The invention relates to the technical field of printing and dyeing, and in particular to a method, a device and an application for predicting the water quality of cotton dyeing effluent. Background Art

[0002] The printing and dyeing industry is a typical high-pollution, high-water consumption industry, among which the cotton dyeing process is the most important production unit and pollution-discharging unit in the printing and dyeing industry.

[0003] The cotton dyeing process can be divided into two stages: dyeing and cleaning. In the dyeing stage, the cotton fiber woven grey cloth and the active dye solution are first put into the dyeing vat, and then the diffusion of the active dye is promoted by heating. After reaching the appropriate temperature, a neutral electrolyte is added to promote the adsorption of the active dye by the fiber. After a certain period of time, alkali is added to promote the covalent bonding of the active dye and the cotton fiber, and the reaction continues until the set time. In the cleaning stage, the first water washing, soap washing and second water washing are carried out in sequence to wash away the residual chemicals. The wastewater generated in the dyeing stage is dyeing wastewater, which contains some residual active dyes, neutral electrolytes, etc., which are difficult to be effectively degraded by conventional biochemical processes; the wastewater generated in the cleaning stage is finishing wastewater, which has a lower pollutant load than dyeing wastewater. Both dyeing wastewater and finishing wastewater belong to the effluent of the cotton dyeing process.

[0004] For the effluent from the cotton dyeing process, the existing treatment methods include: a mixed flow treatment method in which dyeing wastewater and finishing wastewater are mixed and then treated uniformly, and part of the wastewater after unified treatment is reused; and an empirical classification treatment and reuse method in which dyeing wastewater and finishing wastewater are divided into clear wastewater and turbid wastewater for separate treatment, and the treated clear wastewater is reused. However, in actual production, each process link has different requirements for the quality of recycled water. The existing treatment method is relatively simple for the classification of cotton dyeing effluent, and cannot meet the requirements of each process link for the quality of recycled water, so it needs further treatment, which makes the treatment efficiency of cotton dyeing process effluent low and the reuse cost high. Summary of the invention

[0005] The present invention proposes a method, device and application for predicting the water quality of cotton dyeing effluent, which can predict the water quality of cotton dyeing effluent and obtain effluent water quality parameters of multiple sections of the cotton dyeing process to be operated in the future, so as to facilitate the classification and treatment of the cotton dyeing effluent to meet the requirements of each process link for the quality of recycled water, improve the treatment efficiency of the effluent from the cotton dyeing process and reduce the reuse cost.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] In the first aspect, the present invention provides a method for predicting the water quality of cotton dyeing effluent, comprising: obtaining process parameters and dye parameters of multiple sections of a cotton dyeing process to be run, the process parameters including dye dosage, salt dosage and soda ash dosage, and the dye parameters including the topological polar surface area, octanol-water distribution ratio, number of oxidizable atoms and number of unsaturated bonds of the dye used in the cotton dyeing process to be run. The process parameters and dye parameters of multiple sections of the cotton dyeing process to be run are input into an effluent water quality prediction model. The effluent water quality prediction model predicts the input process parameters and dye parameters of the cotton dyeing process to be run, and outputs effluent water quality parameters of multiple sections of the cotton dyeing process to be run in the future; the effluent water quality parameters include chemical oxygen demand and conductivity.

[0008] In the method for predicting the water quality of cotton dyeing effluent provided by the present invention, the dye dosage, salt dosage and soda ash dosage are used as process parameters, and the topological polar surface area, octanol-water distribution ratio, number of oxidizable atoms and number of unsaturated bonds of the dye used in the cotton dyeing process to be operated are used as dye parameters. The process parameters and dye parameters are predicted by an effluent quality prediction model to obtain effluent quality parameters of multiple sections of the cotton dyeing process to be operated in the future, thereby facilitating classified treatment of the cotton dyeing effluent, so as to meet the requirements of each process link for the quality of recycled water, improve the treatment efficiency of the effluent from the cotton dyeing process, and reduce the reuse cost.

[0009] In the second aspect, the present invention provides a prediction device for cotton dyeing effluent water quality, including a parameter determination module, a parameter input module and a prediction result output module. The parameter determination module is used to obtain the process parameters and dye parameters of the cotton dyeing process to be run, the process parameters include the dye dosage, the salt dosage and the soda ash dosage, and the dye parameters include the topological polar surface area, octanol-water distribution ratio, the number of oxidizable atoms and the number of unsaturated bonds of the dye used in the cotton dyeing process to be run. The parameter input module is used to input the process parameters and dye parameters of the cotton dyeing process to be run into the effluent water quality prediction model. The prediction result output module is used for the effluent water quality prediction model to predict the input process parameters and dye parameters of the cotton dyeing process to be run, and output the effluent water quality parameters of multiple sections of the cotton dyeing process to be run in the future; the effluent water quality parameters include chemical oxygen demand and conductivity.

[0010] In an implementation of the second aspect, the prediction device further includes: an optimization module. The optimization module is used to optimize the effluent water quality prediction model.

[0011] In an implementation of the first aspect and the second aspect, the effluent water quality prediction model is constructed based on a random forest model.

[0012] In an implementation of the first aspect and the second aspect, the outlet water quality prediction model includes an input end, multiple decision trees whose root nodes are connected to the input end, a combiner connected to the leaf nodes of the multiple decision trees, and an output end connected to the combiner.

[0013] In an implementation of the first aspect and the second aspect, the training process of the effluent water quality prediction model includes: obtaining a training set of the effluent water quality prediction model, the training set includes: data samples and labels of the data samples, wherein the data samples include multiple groups of process parameters and dye parameters of cotton dyeing processes, and the labels of the data samples include multiple groups of effluent water quality parameters of multiple sections of the cotton dyeing process corresponding to the multiple groups of process parameters and dye parameters of the cotton dyeing process. Multiple sub-training sets are constructed based on the training set, and each of the multiple sub-training sets is used to train a decision tree. Multiple decision trees are constructed based on hyperparameters and training sets. Multiple decision trees are trained using multiple sub-training sets, and for each decision tree in the multiple decision trees, the decision tree is further expanded by minimizing the error between the output parameters of the decision tree and the effluent water quality parameter samples corresponding to the decision tree, until the decision tree reaches a preset stop condition, thereby obtaining an effluent water quality prediction model.

[0014] In an implementation of the first aspect and the second aspect, the hyperparameters include the number of decision trees, the maximum depth of the tree, and the minimum number of samples at leaf nodes; the preset stopping conditions include the depth of the decision tree reaching the maximum depth of the tree, the decision tree reaching the maximum number of features, the decision tree reaching the maximum number of leaf nodes, the decision tree reaching the minimum impurity of leaf nodes and / or the decision tree reaching the minimum sample weight.

[0015] In an implementation of the first aspect and the second aspect, the training process of the effluent water quality prediction model further includes: optimizing hyperparameters by a grid search method.

[0016] In an implementation of the first aspect and the second aspect, each point in the hyperparameter grid corresponds to a set of hyperparameters of the effluent water quality prediction model.

[0017] In an implementation of the first aspect and the second aspect, the process parameters further include bath ratio, cloth weight, water volume, reaction temperature and holding time.

[0018] In a third aspect, the present invention provides an electronic device, comprising a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory, so that the electronic device executes the method as described in the first aspect or any one of its implementations.

[0019] In a fourth aspect, the present invention provides a computer-readable storage medium, comprising computer program instructions, which, when executed by a computer, enable the computer to execute the method in the first aspect or any one of its implementations.

[0020] In a fifth aspect, the present invention provides a computer program product, comprising computer program instructions, which, when executed on a computer, enable the computer to execute the method in the first aspect or any one of its implementations.

[0021] In the sixth aspect, the present invention provides an application of a method for predicting the water quality of cotton dyeing effluent. Based on the above-mentioned method for predicting the water quality of cotton dyeing effluent, the method is characterized in that, based on the prediction results of the effluent quality prediction model, the effluent from multiple sections of the cotton dyeing process is classified, treated and reused.

[0022] The technical effects corresponding to the above-mentioned second to sixth aspects and their possible implementations can refer to the above-mentioned description of the technical effects of the first aspect and its possible implementations, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is one of the schematic diagrams of a method for predicting the water quality of cotton dyeing effluent provided in the embodiments of the present application;

[0024] Figure 2 It is a schematic diagram of the structure of the effluent water quality prediction model provided in the embodiment of the present application;

[0025] Figure 3 This is a second schematic diagram of a method for predicting the water quality of cotton dyeing effluent provided in an embodiment of the present application;

[0026] Figure 4 This is one of the structural schematic diagrams of a device for measuring the water quality of cotton dyeing provided in an embodiment of the present application;

[0027] Figure 5 This is the second structural schematic diagram of a device for improving the water quality of cotton dyeing provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms "first" and "second" etc. in the description and claims of the present invention are used to distinguish different objects rather than to describe a specific order of the objects.

[0029] The “and / or” in the embodiments of the present application indicates the relationship between objects. For example, A and / or B may indicate the following three situations: A exists alone, B exists alone, and A and B exist at the same time.

[0030] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0031] In the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to two or more than two. For example, multiple decision trees refers to two or more decision trees.

[0032] The method and device provided in the embodiments of the present application relate to the prediction of effluent water quality of cotton dyeing. Specifically, an effluent water quality prediction model is used to predict the effluent water quality parameters of multiple sections of the cotton dyeing process to be run in the future based on the process parameters and dye parameters of the cotton dyeing process to be run.

[0033] In order to solve the problem that the existing treatment methods in the background technology are relatively simple in classifying cotton dyeing effluent, cannot meet the requirements of each process link for the water quality of recycled water, and thus need further treatment, which leads to low treatment efficiency and high reuse cost of the effluent from the cotton dyeing process, the embodiment of the present application provides a method, device and application for predicting the water quality of cotton dyeing effluent, taking the dye dosage, salt dosage and soda ash dosage as process parameters, and taking the topological polar surface area, octanol-water distribution ratio, number of oxidizable atoms and number of unsaturated bonds of the dye used in the cotton dyeing process to be operated as dye parameters, and predicting the process parameters and dye parameters through the effluent quality prediction model, and obtaining the effluent quality parameters of multiple sections of the cotton dyeing process to be operated in a future period of time, so as to facilitate the classification and treatment of the cotton dyeing effluent, so as to meet the requirements of each process link for the water quality of recycled water, improve the treatment efficiency of the effluent from the cotton dyeing process, and reduce the reuse cost.

[0034] Exemplarily, the method for predicting the water quality of cotton dyeing effluent provided in the embodiment of the present invention can be executed by an electronic device with a processing function, for example, the electronic device can be a computer, a server, etc. Taking the electronic device as a computer as an example, the hardware part of the computer can include: a processor, a memory, a network interface, a user interface, a communication bus, etc.

[0035] The processor is used to control the electronic device to perform related processing and computing tasks, such as obtaining process parameters and dye parameters, predicting effluent water quality parameters through an effluent water quality prediction model, etc. The processor may include a central processing unit (CPU) or other processors, and the processor may be single-core or multi-core, for example, the processor may include multiple CPUs.

[0036] The memory is used to store computer instructions and related data, for example, process parameters, dye parameters, effluent water quality prediction models and effluent water quality parameters. The memory can be a random access memory (RAM), a read only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, or an optical memory, a disk storage medium or other magnetic storage device, or any other medium that can be used to store program codes or data that can be accessed by a computer. Optionally, the memory can be integrated into the processor, and the memory can also be independent of the processor.

[0037] The network interface is used for the computer to communicate with other devices or communication networks. The network interface can be a transceiver with transceiver functions. Optionally, the network interface can include a standard wired interface, a wireless interface (such as a WI-FI interface, a Bluetooth interface, a 5G interface).

[0038] The communication bus is used to realize the connection and communication between various components. For example, the processor, memory, network interface and user interface mentioned above can be interconnected through the communication bus.

[0039] The user interface may include a display screen and an input unit (such as a keyboard). Optionally, the user interface may also include a standard wired interface and a wireless interface.

[0040] Those skilled in the art will appreciate that the above-mentioned computer may also include more or fewer components, or a combination of certain components, or different arrangements of components, which is not limited in the embodiments of the present application.

[0041] like Figure 1 As shown, the method for predicting the water quality of cotton dyeing effluent provided in the embodiment of the present application includes S101-S103.

[0042] S101. Obtain process parameters and dye parameters of multiple sections of a cotton dyeing process to be run.

[0043] In the embodiment of the present application, the above-mentioned process parameters include the dye dosage, salt dosage and soda ash dosage, and the dye parameters include the topological polar surface area, octanol-water distribution ratio, number of oxidizable atoms and number of unsaturated bonds of the dye used in the cotton dyeing process to be run.

[0044] Among the process parameters, the dye dosage refers to the amount of dye added to the dyeing medium in the cotton dyeing process; the salt dosage refers to the amount of salt added to the dyeing medium in the cotton dyeing process; the soda dosage refers to the amount of soda ash (Na2CO3) added to the dyeing medium in the cotton dyeing process. The units of dye dosage, salt dosage and soda dosage are usually mass or volume.

[0045] In some cases, reactive dyes such as Reactive Yellow K-6G, CAS No. 50662-99-2, Reactive Red 195, CAS No. 93050-79-4, and Reactive Yellow 145, CAS No. 93050-80-7 may be included; and salts may include sodium chloride or sodium sulfate.

[0046] Among the dye parameters,

[0047] Topological Polar Surface Area (TPSA) is a parameter used in computational chemistry to characterize the polarity of a molecule. TPSA represents the sum of the surface areas of the polar regions within the molecule.

[0048] Octanol-Water Partition Ratio (OWR) refers to the ratio of the concentration of a substance in the octanol phase and the water phase when equilibrium is reached under specific conditions. OWR is an important chemical parameter used to describe the hydrophobicity or hydrophilicity of a substance and its distribution behavior in the environment.

[0049] The number of oxidizable atoms refers to the number of atoms in a molecule that can participate in redox reactions;

[0050] The number of unsaturated bonds refers to the total number of chemical bonds without single bonds in an organic molecule. The number of unsaturated bonds is an important parameter to measure the number of multiple bonds and rings in a molecule, which reflects the structural characteristics and potential reactivity of the molecule.

[0051] Optionally, the process parameters also include bath ratio, cloth weight, water volume, reaction temperature and holding time.

[0052] The bath ratio refers to the ratio of the mass of the dyed material to the mass of the dye solution prepared during immersion dyeing or exhaustion dyeing; the cloth weight refers to the grams of cotton cloth per square meter after dyeing (g / m 2 ).

[0053] In some application scenarios, the multiple sections of the cotton dyeing process to be run include: feeding section (putting cotton fiber grey cloth and reactive dyes into the dyeing vat), heating section (heating the dyeing vat to promote dye diffusion), salting section (adding neutral electrolytes to promote dye adsorption), alkali section (adding alkali to promote covalent bonding of dyes and cotton fibers), drainage section (discharging dyeing wastewater), and washing / soaping section (discharging washing wastewater).

[0054] S102, inputting process parameters of multiple sections of the cotton dyeing process to be run and dye parameters into an outlet water quality prediction model.

[0055] In the embodiment of the present application, the above-mentioned effluent water quality prediction model is constructed based on the random forest model.

[0056] refer to Figure 2 ,The effluent water quality prediction model includes an input end, a plurality of decision trees whose root nodes are connected to the input end, a combiner connected to the leaf nodes of the plurality of decision trees, and an output end connected to the combiner.

[0057] Continue to refer Figure 2 The structure of a decision tree includes: a root node (the starting point of the decision tree, containing all samples to be classified), internal nodes connected to the root node (also called split nodes, used to divide samples according to feature values, each internal node corresponds to a feature), branches connecting two nodes (representing the corresponding feature values), and leaf nodes (the terminal nodes of the decision tree, representing the prediction results of the decision tree).

[0058] In one implementation, the training process of the effluent water quality prediction model includes:

[0059] S1. Obtain a training set for an effluent water quality prediction model, the training set comprising: data samples and labels of the data samples, wherein the data samples comprise multiple groups of process parameters and dye parameters of cotton dyeing processes, and the labels of the data samples comprise multiple groups of effluent water quality parameters of multiple working sections of the cotton dyeing processes corresponding to the multiple groups of process parameters and dye parameters of the cotton dyeing processes.

[0060] It can be understood that the above-mentioned multiple groups of effluent water quality parameter samples of multiple sections of the cotton dyeing process, as well as the process parameters and dye parameters of the cotton dyeing process corresponding to the multiple groups of effluent water quality parameter samples of multiple sections of the cotton dyeing process, come from historical data of multiple sections of the cotton dyeing process.

[0061] S2. Construct multiple sub-training sets based on the training set, and each of the multiple sub-training sets is used to train a decision tree.

[0062] The above sub-training sets can be obtained by performing Bootstrap sampling on the training set. Bootstrap sampling refers to sampling with replacement. It can be understood that the data samples in each sub-training set correspond to a set of process parameters and dye parameters of the cotton dyeing process, and the labels of the data samples correspond to a set of effluent water quality parameters of multiple sections of the cotton dyeing process.

[0063] S3. Construct multiple decision trees based on hyperparameters, process parameters in the training set, and dye parameters.

[0064] In the above process, hyperparameters may include the number of decision trees, the maximum depth of the tree, and the minimum number of samples for leaf nodes.

[0065] Exemplarily, the splitting criterion may be information gain or Gini impurity, etc.; the termination condition may be that the decision tree reaches a set depth, or that the number of nodes of the decision tree reaches a minimum number of node samples.

[0066] Optionally, the above-mentioned end condition may also be that the decision tree reaches a maximum number of features, the decision tree reaches a maximum number of leaf nodes, the decision tree reaches a minimum impurity of leaf nodes and / or the decision tree reaches a minimum sample weight.

[0067] S4. Multiple decision trees are trained through multiple sub-training sets. For each of the multiple decision trees, the decision tree is further expanded by minimizing the error between the output parameters of the decision tree and the effluent water quality parameter samples corresponding to the decision tree, until the decision tree reaches a preset stopping condition, thereby obtaining an effluent water quality prediction model.

[0068] The process of building and training a decision tree is as follows.

[0069] 1. Initialize the tree structure

[0070] Initialize the root node of the decision tree, which includes all the training data samples and their labels in the sub-training set.

[0071] 2. Choose the best split point

[0072] Calculate the split point: For all features of the current node, calculate the split criterion value corresponding to the possible split point, such as the mean square error value (the mean square error between the actual label value of all samples in the child node after the split and the average label value of the node). Each split point is a certain threshold of the feature value. The data sample of the current node is divided into two parts. For each possible split point, calculate the mean square error of the two child nodes after the split, and then take the weighted average. Select the split point that minimizes the weighted average mean square error value as the split point of the current node.

[0073] 3. Execute split

[0074] Based on the selected optimal split point, the data of the current node is divided into two subsets, and a new child node is created for each subset.

[0075] 4. Recursive Splitting

[0076] Repeat the above steps for each newly created child node until the stopping condition is met.

[0077] 5. Generate the final decision tree

[0078] When all nodes meet the stopping condition, the training process of the decision tree ends and the final decision tree is generated. This model consists of each node and its split point. Each leaf node contains the average label value of the training samples in this node, that is, the predicted value of the leaf node.

[0079] In the process of training multiple decision trees through multiple sub-training sets, each decision tree can randomly select a different input feature subset (i.e., input features selected from process parameters and dye parameters), and when each decision tree performs node splitting, only features from the corresponding feature subset are selected as splitting points to calculate the separation criterion value, thereby enhancing the generalization ability of the model.

[0080] Optionally, the preset stopping condition in S4 may be that the depth of the decision tree reaches the maximum depth of the tree.

[0081] In one application scenario, the training process of the effluent water quality prediction model also includes: optimizing hyperparameters through a grid search method.

[0082] Specifically, multiple groups of hyperparameters are selected, and a hyperparameter grid is constructed based on the multiple groups of hyperparameters. Each point in the hyperparameter grid corresponds to a group of hyperparameters of the effluent water quality prediction model; the hyperparameters include the number of decision trees, the maximum depth of the tree, and the minimum number of samples of leaf nodes.

[0083] Furthermore, based on multiple sets of hyperparameters corresponding to multiple points in the hyperparameter grid, multiple models are constructed and trained, and the hyperparameters of the model with the best prediction effect are taken as the optimal hyperparameters.

[0084] Optionally, the above-mentioned effluent water quality prediction model can also be constructed based on the extreme gradient boosting model, or can be constructed by a neural network algorithm, such as a convolutional neural network (CNN), a recurrent neural network (RNN), a fully connected neural network, etc.

[0085] S103, the effluent water quality prediction model predicts the input process parameters and dye parameters of the cotton dyeing process to be run, and outputs effluent water quality parameters of multiple sections of the cotton dyeing process to be run in the future.

[0086] Among them, the effluent water quality parameters include chemical oxygen demand and conductivity.

[0087] In an embodiment of the present application, after the process parameters and dye parameters of the cotton dyeing process to be run are input into the effluent water quality prediction model, the input process parameters and dye parameters enter the multiple decision trees of the effluent water quality prediction model through the input end, and the output results of the leaf nodes of the multiple decision trees are input into the connector. After the connector processes the output results of the multiple decision trees, it outputs the prediction results of the effluent water quality prediction model through the output end.

[0088] In one implementation, the combiner processes the output results of multiple decision trees by weighted averaging, and the averaging method satisfies the following formula:

[0089]

[0090] In the above formula, H (x) is the output of the combiner, T is the number of decision trees, w i is the weight coefficient of the i-th decision tree, h i (x) is the output result of the i-th decision tree.

[0091] It should be noted that the chemical oxygen demand among the effluent water quality parameters of the effluent water quality prediction model is strongly correlated with the dye parameters input into the effluent water quality prediction model.

[0092] Specifically, since the COD (chemical oxygen demand) in the effluent of the cotton dyeing process mainly comes from the residual reactive dyes, and different types of reactive dyes have different topological polar surface areas, octanol-water distribution ratios, oxidizable atomic numbers, and unsaturated bond numbers (i.e., dye parameters), even if they have the same concentration, the final COD concentration is different. Therefore, the above dye parameters have a better prediction effect on the chemical oxygen demand in the effluent water quality parameters.

[0093] Furthermore, the conductivity among the above-mentioned effluent water quality parameters is strongly correlated with the process parameters input into the effluent water quality prediction model.

[0094] Specifically, the conductivity is strongly correlated with the salt dosage in multiple stages of the cotton dyeing process. The greater the dye dosage, salt dosage, and soda ash dosage (i.e., process parameters), the higher the conductivity. Therefore, the above process parameters have a better prediction effect on the conductivity of the effluent water quality parameters.

[0095] It can be understood that COD refers to the amount of oxidant consumed when a certain strong oxidant is used to treat a water sample under certain conditions. It is an indicator of the amount of reducing substances in the water. The larger the COD value in the wastewater, the more serious the degree of water pollution. The COD of the effluent from the cotton dyeing process can reflect the degree of water pollution. The conductivity of the effluent from the cotton dyeing process refers to the measure of the ion migration ability in the water, which can reflect the content of inorganic salts and other soluble substances in the effluent from the cotton dyeing process. It can be seen that the effluent water quality parameters of the effluent water quality prediction model can better reflect the pollution of the effluent from the cotton dyeing process.

[0096] In summary, when the effluent water quality prediction model can achieve a better prediction of the effluent water quality parameters through dye parameters and process parameters, the effluent water quality parameters can better reflect the pollution situation of the effluent from the cotton dyeing process, which means that the effluent water quality prediction model can have a better prediction effect on the effluent water quality of the cotton dyeing process.

[0097] In the prediction method provided in the embodiment of the present application, the dye dosage, salt dosage and soda ash dosage are used as process parameters, and the topological polar surface area, octanol-water distribution ratio, number of oxidizable atoms and number of unsaturated bonds of the dye used in the cotton dyeing process to be operated are used as dye parameters. The process parameters and dye parameters are predicted by the effluent water quality prediction model to obtain the effluent water quality parameters of multiple sections of the cotton dyeing process to be operated in the future, so as to facilitate the classification and treatment of the cotton dyeing effluent, so as to meet the requirements of each process link for the quality of recycled water, improve the treatment efficiency of the effluent from the cotton dyeing process, and reduce the reuse cost.

[0098] In some application scenarios, combined with Figure 1 ,like Figure 3 As shown, the method for predicting the water quality of cotton dyeing effluent provided in the embodiment of the present application further includes: S104, optimizing the effluent water quality prediction model.

[0099] Optionally, historical data with low prediction accuracy of the effluent water quality prediction model is obtained and composed into a new training set, and a new decision tree is constructed and trained based on the new training set.

[0100] The above process can not only improve the prediction accuracy in new process scenarios, but also maintain the understanding of old process scenarios and achieve continuous self-optimization of the model.

[0101] Correspondingly, the embodiment of the present application provides an application of the above-mentioned cotton dyeing effluent water quality prediction method, and based on the prediction results of the effluent water quality prediction model, the effluent from multiple sections of the cotton dyeing process is classified, treated and reused.

[0102] The application of the above-mentioned method for predicting the water quality of cotton dyeing effluent is described in detail below.

[0103] Based on the prediction results of the effluent quality prediction model obtained by the prediction method of cotton dyeing effluent quality and the water demand of the reuse target section, the effluents of multiple sections of the cotton dyeing process to be operated are classified into the four levels in the following table.

[0104] Table 1 Water classification table

[0105] Effluent classification destination COD(mg / L) Conductivity (μS / cm) Direct reuse ≤120 ≤6000 Simple processing and reuse ≤350 ≤12000 Deep processing and reuse ≤1000 ≤25000 Treatment of emissions that meet standards ＞1000 ＞25000

[0106] After obtaining the classification results, treatment equipment is installed at the water outlets of multiple work sections according to the classification results to achieve the treatment and reuse of cotton dyeing effluent.

[0107] Understandably, in the cotton dyeing process, the wastewater COD, conductivity, chromaticity, hardness and iron and manganese ion concentration have a serious impact on product quality. Since no new hardness and iron and manganese ions are introduced into the dyeing and wastewater treatment process, the hardness and iron and manganese ion concentration in the process all come from the initial fresh water, which will not affect the wastewater reuse; at the same time, after the wastewater is classified and treated, the molecular structure of the dye that produces the water color is basically destroyed, so it only has an impact when it is directly reused. In the actual operation process, only the light-colored effluent of the same batch of orders and the same type of process with the same color and different shades will be directly reused, which can be achieved through human judgment.

[0108] Accordingly, the present application embodiment provides a device for predicting the water quality of cotton dyeing effluent, such as Figure 4 As shown, it includes a parameter determination module 501, a parameter input module 502 and a prediction result output module 503.

[0109] The parameter determination module 501 is used to obtain the process parameters and dye parameters of the cotton dyeing process to be run, the process parameters include the dye dosage, salt dosage and soda dosage, and the dye parameters include the topological polar surface area, octanol-water distribution ratio, oxidizable atomic number and unsaturated bond number of the dye used in the cotton dyeing process to be run. For example, the parameter determination module 501 is used to implement S101 of the above prediction method.

[0110] The parameter input module 502 is used to input the process parameters and dye parameters of the cotton dyeing process to be run into the effluent water quality prediction model. For example, the parameter input module 502 is used to implement S102 of the above prediction method.

[0111] The prediction result output module 503 is used for the effluent water quality prediction model to predict the input process parameters and dye parameters of the cotton dyeing process to be run, and output the effluent water quality parameters of multiple sections of the cotton dyeing process to be run in the future; the effluent water quality parameters include chemical oxygen demand and conductivity. For example, the prediction result output module 503 is used to implement S103 of the above prediction method.

[0112] Optionally, combined Figure 4 ,like Figure 5 As shown, the prediction device further includes an optimization module 504. For example, the optimization module 504 is used to implement S104 of the prediction method.

[0113] The optimization module 504 is used to optimize the effluent water quality prediction model.

[0114] The various modules of the above-mentioned cotton dyeing effluent water quality prediction device can also be used to execute other steps in the above-mentioned method embodiment. All relevant contents involved in the above-mentioned method embodiment can be referred to the functional description of the corresponding functional module and will not be repeated here.

[0115] The embodiment of the present application also provides an electronic device, including: a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory, so that the electronic device executes the method described in the above embodiment. The processor can implement the above parameter determination module 501, parameter input module 502 and prediction result output module 503; the above memory can also be used to store process parameters, dye parameters, effluent water quality prediction model and effluent water quality parameters, etc.

[0116] An embodiment of the present application further provides a computer-readable storage medium, which includes a computer program. When the computer program runs on a computer, the method described in the above embodiment is executed.

[0117] An embodiment of the present application further provides a computer program product, which includes computer program instructions. When the computer program instructions are run on a computer, the method described in the above embodiment is executed.

[0118] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for predicting the water quality of cotton dyeing effluent, characterized in that: include: Obtaining process parameters and dye parameters of multiple sections of a cotton dyeing process to be run, wherein the process parameters include dye dosage, salt dosage and soda ash dosage, and the dye parameters include topological polar surface area, octanol-water distribution ratio, number of oxidizable atoms and number of unsaturated bonds of the dye used in the cotton dyeing process to be run; Inputting process parameters of multiple sections of the cotton dyeing process to be run and dye parameters into the effluent water quality prediction model; The effluent water quality prediction model predicts the input process parameters and dye parameters of the cotton dyeing process to be run, and outputs the effluent water quality parameters of multiple sections of the cotton dyeing process to be run in the future; the effluent water quality parameters include chemical oxygen demand and conductivity.

2. The method according to claim 1, characterized in that The effluent water quality prediction model is constructed based on a random forest model.

3. The method according to claim 2, characterized in that The effluent water quality prediction model includes an input end, multiple decision trees whose root nodes are connected to the input end, a combiner connected to the leaf nodes of the multiple decision trees, and an output end connected to the combiner.

4. The method according to claim 1 or 3, characterized in that The training process of the effluent water quality prediction model includes: Obtaining a training set of the effluent water quality prediction model, the training set comprising: data samples and labels of the data samples, wherein the data samples comprise multiple sets of process parameters and dye parameters of cotton dyeing processes, and the labels of the data samples comprise multiple sets of effluent water quality parameters of multiple sections of the cotton dyeing processes corresponding to the multiple sets of process parameters and dye parameters of the cotton dyeing processes; Randomly construct multiple sub-training sets based on the training set, each of the multiple sub-training sets is used to train a decision tree; Constructing multiple decision trees based on the hyperparameters and the training set; Multiple decision trees are trained using the multiple sub-training sets. For each of the multiple decision trees, the decision tree is further expanded by minimizing the error between the output parameter of the decision tree and the effluent water quality parameter sample corresponding to the decision tree, until the decision tree reaches a preset stopping condition, thereby obtaining the effluent water quality prediction model.

5. The method according to claim 4, characterized in that The hyperparameters include the number of decision trees, the maximum depth of the tree, and the minimum number of samples at leaf nodes; the preset stopping conditions include the depth of the decision tree reaching the maximum depth of the tree, the decision tree reaching the maximum number of features, the decision tree reaching the maximum number of leaf nodes, the decision tree reaching the minimum impurity of leaf nodes and / or the decision tree reaching the minimum sample weight.

6. The method according to claim 4, characterized in that The training process of the effluent water quality prediction model also includes: The hyperparameters are optimized by grid search method.

7. The method according to claim 6, characterized in that Each point in the hyperparameter grid corresponds to a set of hyperparameters of the effluent water quality prediction model.

8. The method according to claim 1, characterized in that The process parameters also include bath ratio, cloth weight, water volume, reaction temperature and insulation time.

9. A device for predicting the water quality of cotton dyeing effluent, characterized in that: It includes a parameter determination module, a parameter input module and a prediction result output module; The parameter determination module is used to obtain process parameters and dye parameters of the cotton dyeing process to be run, wherein the process parameters include dye dosage, salt dosage and soda ash dosage, and the dye parameters include the topological polar surface area, octanol-water distribution ratio, oxidizable atomic number and unsaturated bond number of the dye used in the cotton dyeing process to be run; The parameter input module is used to input the process parameters and dye parameters of the cotton dyeing process to be run into the effluent water quality prediction model; The prediction result output module is used for the effluent water quality prediction model to predict the input process parameters and dye parameters of the cotton dyeing process to be run, and output the effluent water quality parameters of multiple sections of the cotton dyeing process to be run in the future; the effluent water quality parameters include chemical oxygen demand and conductivity.

10. Application of a method for predicting water quality of cotton dyeing effluent, based on the method according to any one of claims 1 to 8, characterized in that: Based on the prediction results of the effluent quality prediction model, effluents from multiple sections of the cotton dyeing process are classified, treated and reused.

Images