A method for optimizing the dosage of sludge dewatering flocculant
By constructing a flocculation effect prediction operator, optimizing the type and dosage of flocculant, the problem of difficult to accurately control the dosage of flocculant during sludge dehydration is solved, and the stability and scientificity of the dehydration effect are achieved.
Patent Information
- Application Number
- CN202410942814.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2025-05-16
- Estimated Expiration
- 2044-07-15
AI Technical Summary
During the dehydration of sludge, the dosage of flocculant is difficult to accurately control, resulting in unstable dehydration effect, which depends on the experience of the operator and trial and error methods, and lacks scientificity and consistency.
By collecting historical data on sludge dehydration, a flocculation effect prediction operator is constructed, and based on the current sludge concentration and pH value, the type and dosage of flocculants are optimized to achieve accurate prediction of the dosage of flocculants.
It reduces the error caused by subjective judgment, ensures the stability of the dehydration effect, and improves the scientificity and consistency of the sludge dehydration process.
Smart Images
Figure CN119106754B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to water treatment related fields, and in particular to a method for optimizing the dosage of a sludge dewatering flocculant. Background Art
[0002] In the sludge water treatment system of the water supply plant, sludge water treatment is a crucial link, which is of great significance for maintaining the normal operation of the water supply system, reducing maintenance costs, and ensuring the safety and stability of the effluent water quality. In the process of sludge dewatering, the addition of flocculants can significantly improve the sedimentation performance and dewatering efficiency of the sludge, thereby reducing the sludge moisture content and facilitating subsequent treatment and disposal. However, the addition of flocculants is not the more the better. Excessive addition will not only increase the cost, but also have a negative impact on the treatment effect; on the contrary, insufficient addition will not achieve the expected effect. Therefore, how to accurately and efficiently determine the type and dosage of flocculants has become a key issue in the optimization of sludge dewatering process. In the existing sludge dewatering process, the type and dosage of flocculants often depend on the experience and trial and error of the operator. This method relies on the experience and judgment of the operator, lacks scientificity and consistency, and the properties of different batches of sludge may be different, resulting in fluctuations in the flocculation effect.
[0003] In the current related technologies, the addition of sludge dewatering flocculants has technical problems such as strong subjective dependence and unstable dewatering effect. Summary of the invention
[0004] The present application provides a method for optimizing the dosage of flocculant for sludge dewatering, adopts technical means such as constructing a flocculation effect prediction operator based on historical sludge dewatering data, and optimizing the type and dosage of flocculant according to the current sludge concentration and the current pH value, thereby achieving accurate prediction of the type and dosage of flocculant, reducing the error caused by subjective judgment, and ensuring the stability of the dewatering effect.
[0005] The present application provides a method for optimizing the dosage of a sludge dewatering flocculant, comprising:
[0006] A historical data set in the sludge dewatering process is collected and obtained, including a historical sludge concentration set, a historical pH value set, a historical flocculant type set, a historical flocculant dosage set and a historical flocculation effect set, wherein the historical sludge concentration set, the historical pH value set, the historical flocculant type set and the historical flocculant dosage set are used as a sludge dewatering index set; the sludge dewatering index set is used as an input parameter, and the historical flocculation effect set is used as a supervision parameter to train a flocculation effect prediction operator; the current sludge concentration and the current pH value are collected, and an optimization analysis of the flocculant type and dosage is performed based on the flocculation effect prediction operator to obtain the optimal flocculant type and the optimal dosage.
[0007] A method for optimizing the dosage of flocculant for sludge dewatering proposed in this application first collects and obtains a set of historical data in the sludge dewatering process, including a set of historical sludge concentrations, a set of historical pH values, a set of historical flocculant types, a set of historical flocculant dosages and a set of historical flocculation effects, wherein the set of historical sludge concentrations, the set of historical pH values, the set of historical flocculant types and the set of historical flocculant dosages are used as a set of sludge dewatering indexes. Then, the set of sludge dewatering indexes is used as an input parameter and the set of historical flocculation effects is used as a supervision parameter to train a flocculation effect prediction operator. Finally, the current sludge concentration and the current pH value are collected, and an optimization analysis of the type and dosage of flocculant is performed based on the flocculation effect prediction operator to obtain the optimal type of flocculant and the optimal dosage, thereby achieving the technical effect of reducing the error caused by subjective judgment and ensuring the stability of the dewatering effect. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.
[0009] Figure 1 A schematic flow chart of a method for optimizing the dosage of a sludge dewatering flocculant provided in an embodiment of the present application.
[0010] Figure 2 A schematic diagram of a process for obtaining the optimal flocculant type and optimal dosage in a sludge dewatering flocculant dosage optimization method provided in an embodiment of the present application. DETAILED DESCRIPTION
[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.
[0012] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.
[0013] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.
[0014] The present application embodiment provides a method for optimizing the dosage of sludge dewatering flocculant, such as Figure 1 As shown, the method includes:
[0015] Step S100, collect and obtain a historical data set in the sludge dewatering process, including a historical sludge concentration set, a historical pH value set, a historical flocculant type set, a historical flocculant dosage set and a historical flocculation effect set, wherein the historical sludge concentration set, the historical pH value set, the historical flocculant type set and the historical flocculant dosage set are the sludge dewatering index set. Specifically, collect relevant data of the sludge dewatering process from the automation system of the sludge treatment related department or historical records, and classify the collected data by type, including sludge concentration, pH value, type of flocculant used, dosage of flocculant and corresponding flocculation effect (such as moisture content of dehydrated sludge, compressibility of mud cake, etc.). Among them, sludge concentration refers to the content of solid matter (dry matter) in the sludge, expressed as the mass of dry solids per unit volume of sludge; pH value indicates the acidity or alkalinity of the sludge solution; flocculant is a chemical substance used for sludge dewatering, which aggregates sludge particles into larger flocs through adsorption, bridging, and electrical neutralization to facilitate dehydration; flocculant dosage refers to the amount of flocculant added per unit mass or volume of sludge; flocculation effect is an indicator to measure the sludge dewatering effect, such as the moisture content of the sludge after dehydration, the compressibility of the mud cake, the dehydration speed, etc.
[0016] Step S200, using the sludge dehydration index set as input parameters and the historical flocculation effect set as supervision parameters, a flocculation effect prediction operator is trained. Specifically, features that have a significant impact on the flocculation effect are selected from the sludge dehydration index set, and historical data (historical sludge concentration set, historical pH value set, historical flocculant type set, historical flocculant dosage set and corresponding historical flocculation effect set) are used to train a model (such as a regression model, a decision tree, a neural network, etc.), and the model parameters are adjusted so that the model can accurately predict the flocculation effect under different conditions. Among them, the flocculation effect prediction operator is a model obtained through training, which can predict the corresponding flocculation effect according to the input sludge dehydration index.
[0017] In a possible implementation, the flocculation effect prediction operator is trained using the sludge dewatering index set as an input parameter and the historical flocculation effect set as a supervision parameter, and step S200 further includes step S210, extracting a first historical flocculant type based on the historical flocculant type set. Specifically, the first historical flocculant type is any one of the historical flocculant type set, which is randomly generated and used as the object of the current analysis. Step S220, traverses the historical data set, extracts the first historical sludge concentration set, the first historical pH value set, the first historical flocculant dosage set and the first historical flocculation effect set corresponding to the first historical flocculant type, wherein the first historical flocculant type, the first historical sludge concentration set, the first historical pH value set and the first historical flocculant dosage set are the first sludge dewatering index set. Specifically, the entire historical data set is traversed, all records of using the first historical flocculant type are screened out, and the corresponding sludge concentration, pH value, flocculant dosage and flocculation effect are extracted from these records to form a new data set (first historical sludge concentration set, first historical pH value set, first historical flocculant dosage set and first historical flocculation effect set). Step S230, using the first sludge dehydration index set as an input parameter and the first historical flocculation effect set as a supervision parameter, a first flocculation effect prediction operator is trained. Specifically, using the extracted first sludge dehydration index set as an input parameter and the first historical flocculation effect set as a supervision parameter, a machine learning model, such as a regression model, a neural network, etc., is used to perform model training, and by continuously adjusting the model parameters, the model can accurately predict the flocculation effect when the first historical flocculant type is used under given sludge concentration, pH value and flocculant dosage. Step S240, according to the first flocculation effect prediction operator, train and obtain multiple first flocculation effect prediction operators corresponding to different flocculants, and integrate the multiple first flocculation effect prediction operators to obtain the flocculation effect prediction operator. Specifically, repeat steps S210 to S230, train a first flocculation effect prediction operator for each type of flocculant that has appeared in history, and use an integrated learning method to integrate these training models for different flocculants to form a unified flocculation effect prediction operator that can handle multiple flocculants. This implementation method can more accurately capture the behavior of each flocculant under specific conditions by training a model for each flocculant, thereby achieving the technical effect of improving the accuracy of model prediction.
[0018] In a possible implementation, the first sludge dehydration index set is used as an input parameter, and the first historical flocculation effect set is used as a supervision parameter to train and obtain a first flocculation effect prediction operator. Step S230 further includes step S231, and the first historical flocculation effect set is composed of a sludge moisture content set, a supernatant turbidity set, and a sludge volume reduction rate set. Specifically, the first historical flocculation effect set contains three key indicators: a sludge moisture content set, a supernatant turbidity set, and a sludge volume reduction rate set. These three indicators together reflect the effect of the flocculant in the sludge dehydration process and are important indicators for evaluating the flocculation effect. Among them, the sludge moisture content refers to the proportion of water in the sludge; the supernatant turbidity refers to the amount and size of suspended particles in the supernatant after flocculation treatment, reflecting the clarity of the flocculation effect; the sludge volume reduction rate refers to the reduction ratio of the sludge volume after flocculation treatment relative to that before treatment, which is an indicator for evaluating the sludge reduction effect. Step S232, using the first sludge dehydration index set as input parameters, and using the sludge moisture content set, the supernatant turbidity set, and the sludge volume reduction rate set as supervision parameters, respectively, train to obtain the first sludge moisture content prediction operator, the first supernatant turbidity prediction operator, and the first sludge volume reduction rate prediction operator. Specifically, using the first sludge dehydration index set (including sludge concentration, pH value, flocculant type and dosage) as input parameters, and using the sludge moisture content set, the supernatant turbidity set, and the sludge volume reduction rate set as supervision parameters, respectively, three independent prediction models are trained: the first sludge moisture content prediction operator, the first supernatant turbidity prediction operator, and the first sludge volume reduction rate prediction operator. Each prediction model is trained for a specific flocculation effect index to accurately predict the value of the index under given sludge dehydration conditions. Step S233, the first sludge moisture content prediction operator, the first supernatant turbidity prediction operator and the first sludge volume reduction rate prediction operator are connected in parallel to obtain the first flocculation effect prediction operator. Specifically, the three trained prediction operators are connected in parallel, that is, when predicting, the three prediction operators simultaneously receive the same input parameters (i.e., the first sludge dehydration index set) and respectively output corresponding prediction results (i.e., sludge moisture content, supernatant turbidity and sludge volume reduction rate). Finally, the three prediction results together constitute the output of the first flocculation effect prediction operator, which fully reflects the different aspects of the flocculation effect. This implementation method can more accurately capture the complex relationship between each indicator and the sludge dehydration condition by separately training the prediction operator for each flocculation effect index. At the same time, through parallel connection, it fully reflects the different aspects of the flocculation effect, achieving the technical effect of improving the accuracy and comprehensiveness of the prediction.
[0019] Step S300, collect the current sludge concentration and the current pH value, perform optimization analysis on the type and dosage of flocculants based on the flocculation effect prediction operator, and obtain the optimal type and dosage of flocculants. Specifically, the current sludge concentration and pH value are obtained in real time from the sludge treatment system, and the real-time data is input into the trained flocculation effect prediction operator to predict the flocculation effect under different types and dosages of flocculants. Based on the prediction results, an optimization algorithm (such as genetic algorithm, particle swarm optimization, etc.) is used to find the optimal type and dosage combination of flocculants so that the flocculation effect is optimal, and the optimal type and dosage of flocculants are output for reference and execution by operators. The embodiment of the present application adopts a flocculation effect prediction operator based on the historical data of sludge dehydration, and optimizes the type and dosage of flocculants according to the current sludge concentration and the current pH value, so as to achieve accurate prediction of the type and dosage of flocculants, reduce the error caused by subjective judgment, and ensure the stability of the dehydration effect.
[0020] like Figure 2As shown, in a possible implementation, the current sludge concentration and the current pH value are collected, and the type and dosage of flocculants are optimized and analyzed based on the flocculation effect prediction operator to obtain the optimal type of flocculant and the optimal dosage. Step S300 further includes step S310, constructing a flocculant type space according to the types of optional flocculants, and constructing a flocculant dosage space according to the minimum and maximum values in the historical flocculant dosage set. Specifically, according to the optional flocculant types, a set containing all optional types is listed as the boundary of the search space. The minimum and maximum values of all flocculant dosage values are extracted from the historical data, and these two values define the upper and lower limits of the dosage search space. Step S320, based on the flocculant type space and the flocculant dosage space, randomly generate a first flocculant and multiple first initial dosages, and combine the first flocculant and each of the multiple first initial dosages to obtain multiple first initial particles. Specifically, a flocculant is randomly selected from the flocculant type space as the first flocculant, and a plurality of dosage values are randomly generated in the flocculant dosage space as the first initial dosage. The first flocculant is combined with each first initial dosage to form a plurality of first initial particles, each of which represents a possible initial solution (i.e., a combination of flocculant type and dosage). Step S330, according to the current sludge concentration, the current pH value and the flocculation effect prediction operator, the plurality of first initial particles are iteratively optimized to obtain the first optimal solution fitness. Specifically, each first initial particle is evaluated using the current sludge concentration, the current pH value and the flocculation effect prediction operator, and its corresponding flocculation effect (such as sludge moisture content, supernatant turbidity, sludge volume reduction rate, etc.) is calculated. According to the preset optimization objectives (such as minimizing the sludge moisture content, minimizing the supernatant turbidity, etc.), the fitness of each first initial particle is calculated, and the particles are updated and optimized through an iterative optimization algorithm (such as a genetic algorithm, a particle swarm optimization, etc.) until the stopping conditions are met (such as reaching the maximum number of iterations, the fitness is no longer significantly improved, etc.). Finally, the fitness of the first optimal solution is obtained, that is, the fitness of the best solution found in the current search space (flocculant dosage space). Step S340, remove the first flocculant from the flocculant type space, randomly generate a second flocculant from the new flocculant type space after the removal, randomly generate multiple second initial dosages based on the flocculant dosage space, and combine the second flocculant and each of the multiple second initial dosages to obtain multiple second initial particles; Step S350, iteratively optimize the multiple second initial particles according to the current sludge concentration, the current pH value and the flocculation effect prediction operator to obtain the second optimal solution fitness; Step S360, use the same method to obtain the third optimal solution fitness, the fourth optimal solution fitness... until the flocculant type space is empty, and multiple optimal solution fitnesses are obtained.Specifically, the process of steps S340 to S360 is repeated, but each time a new flocculant is randomly selected from the remaining flocculant type space, and the initial particles are regenerated for iterative optimization, and all possible flocculant types are traversed to ensure that the global optimal solution is found rather than the local optimal solution. Step S370, the fitness of the multiple optimal solutions are arranged in descending order, the fitness ranked first is taken as the optimal fitness, and the flocculant and dosage corresponding to the optimal fitness are extracted as the optimal flocculant type and the optimal dosage. Specifically, the fitness of all found optimal solutions are arranged in descending order, and the fitness ranked first is selected as the optimal fitness. The optimal fitness represents the best performance among all possible solutions. The flocculant and dosage corresponding to the optimal fitness are extracted as the optimal flocculant type and the optimal dosage. This implementation method ensures the comprehensiveness of the search by constructing a flocculant type space and a flocculant dosage space, and traversing all possible combinations, reducing the risk of missing the global optimal solution, and achieving the technical effect of ensuring that the optimal flocculant type and the optimal dosage are the global optimal solution.
[0021] In a possible implementation, the multiple first initial particles are iteratively optimized according to the current sludge concentration, the current pH value and the flocculation effect prediction operator to obtain the first optimal solution fitness, and step S330 further includes step S331, constructing a fitness evaluation function. Specifically, according to the goal of sludge treatment (such as minimizing sludge moisture content, maximizing sludge settling velocity, etc.), combined with the current sludge concentration, the current pH value and the flocculation effect prediction operator, a fitness evaluation function is designed, and the fitness evaluation function receives a set of inputs (such as flocculant type, dosage, sludge concentration, pH value) and outputs a numerical value (i.e., fitness) indicating the pros and cons of the scheme. Step S332, based on the current sludge concentration, the current pH value, the flocculation effect prediction operator and the fitness evaluation function, multiple first initial fitness of the multiple first initial particles are calculated. Specifically, for each first initial particle (including a specific flocculant type and dosage), the fitness evaluation function is used to calculate its corresponding fitness to obtain the first initial fitness. Step S333, summing the multiple first initial fitnesses to obtain the first initial total fitness, taking the ratio of each first initial fitness to the first initial total fitness as the fitness probability, and obtaining multiple first initial fitness probabilities. Specifically, summing all the first initial fitnesses to obtain the first initial total fitness. Calculating the ratio of each first initial fitness to the first initial total fitness to obtain the fitness probability of each first initial particle, this probability reflects the relative importance of the first initial particle in the subsequent selection process. Step S334, selecting the multiple first initial fitness probabilities based on roulette to obtain multiple first initial selection fitnesses, wherein the multiple first initial selection fitnesses correspond to the second quantity, the multiple first initial fitnesses correspond to the first quantity, and the second quantity is less than the first quantity. Specifically, using the roulette selection algorithm (also called the proportional selection algorithm), selection is made according to the fitness probability of each first initial particle, and the roulette selection is a probability selection method, and the probability of each particle being selected is proportional to its fitness probability. The result of the selection process is a plurality of first initial selection fitnesses, and the particles corresponding to these first initial selection fitnesses are used as candidate solutions for the subsequent iterative process. Step S335, the plurality of first initial particles corresponding to the plurality of first initial selection fitnesses are used as a plurality of first initial retained particles, and a plurality of first supplementary particles are randomly generated in the flocculant dosage space. The plurality of first initial retained particles and the plurality of first supplementary particles together constitute a plurality of first updated particles, and the number of the plurality of first updated particles is the same as the first number. Specifically, the particles corresponding to the selected plurality of first initial selection fitnesses are used as a plurality of first initial retained particles, and a plurality of first supplementary particles are randomly generated in the flocculant dosage space to supplement the eliminated particles, so as to keep the total number of particles unchanged. The first initial retained particles and the first supplementary particles together constitute a plurality of first updated particles, which are used in the subsequent iterative optimization process.Step S336, iterative optimization is performed based on the multiple first updated particles, and the first optimal solution fitness is obtained after convergence. Specifically, an iterative optimization algorithm (such as genetic algorithm, particle swarm optimization, etc.) is used to optimize the multiple first updated particles. This process continuously iterates and updates the position and state of the particles to find a better solution. When the stopping condition is met (such as reaching the maximum number of iterations, the fitness is no longer significantly improved, etc.), the iterative optimization process ends and outputs the fitness of the optimal solution found by the iterative optimization process, that is, the first optimal solution fitness. This implementation method ensures that the diversity of the population is maintained during the iteration process through roulette selection and supplementary particles, avoids falling into the local optimal solution, and achieves the technical effect of maintaining the diversity of the population.
[0022] In a possible implementation, the step of constructing a fitness evaluation function, step S331 further includes step S3311, and the fitness evaluation function is expressed as:
[0023]
[0024] Among them, F i is the fitness of the ith flocculant, is the maximum value of the amount of flocculant added in the history, is the minimum value of the amount of flocculant added in the history, x i is the current particle dosage, is the maximum value of the flocculation effect of the i-th flocculant in history, is the minimum value of the flocculation effect of the i-th flocculant in history, y i is the flocculation effect of the current particle, ω1 and ω2 are the weight coefficients of the dosage and flocculation effect, respectively, and ω1+ω2=1. This implementation method ensures the consistency and accuracy of fitness calculation by clarifying the expression and parameters of the fitness evaluation function, which helps to more accurately evaluate the performance of each particle, thereby finding a better solution and achieving the technical effect of improving the accuracy of fitness evaluation.
[0025] In a possible implementation, the multiple first initial fitness of the multiple first initial particles is calculated based on the current sludge concentration, the current pH value, the flocculation effect prediction operator and the fitness evaluation function. Step S332 further includes step S3321, inputting the current sludge concentration, the current pH value and the multiple first initial particles into the flocculation effect prediction operator to obtain the predicted multiple first initial flocculation effects. Specifically, the current sludge concentration, the current pH value and the multiple first initial particles (including the type and dosage of flocculant for each particle) are used as inputs and passed to the flocculation effect prediction operator. The flocculation effect prediction operator calculates multiple first initial flocculation effects according to the input data, that is, the flocculation effect that each first initial particle can achieve under the current sludge concentration and pH value conditions. Step S3322, inputting the multiple first initial dosages and the multiple first initial flocculation effects into the fitness evaluation function to calculate multiple first initial fitness. Specifically, multiple first initial dosages and multiple first initial flocculation effects are passed to the fitness evaluation function, which calculates the fitness corresponding to each first initial particle according to the given expression and parameters, that is, the first initial fitness, which reflects the performance or effect of the first initial particle under the current conditions. This implementation method accurately predicts the flocculation effect of each particle under the current conditions by inputting the current sludge concentration, pH value and the first initial particle into the flocculation effect prediction operator, thereby achieving the technical effect of improving the accuracy and reliability of fitness calculation.
[0026] In a possible implementation, the multiple optimal solution fitnesses are arranged in descending order, the first ranked fitness is taken as the optimal fitness, and the flocculant and dosage corresponding to the optimal fitness are extracted as the optimal flocculant type and optimal dosage. Step S370 further includes step S371, extracting multiple optimal solution particles corresponding to the multiple optimal solution fitnesses, wherein each optimal solution particle corresponds to a flocculant and an dosage. Specifically, from all optimal solutions, particles corresponding to each optimal solution are extracted, and each particle contains a specific flocculant type and an dosage. Step S372, based on the type of flocculant and the dosage, the use cost of the multiple optimal solution particles is calculated, and the multiple optimal solution fitnesses are arranged in ascending order according to the use cost, and the first ranked fitness is taken as the optimal fitness, and the flocculant and dosage corresponding to the optimal fitness are extracted as the optimal flocculant type and optimal dosage. Specifically, for each optimal solution particle, the cost of use in actual application is calculated according to the type and dosage of flocculant it represents, including the purchase cost, transportation cost, storage cost, and processing cost of the flocculant. Based on the calculated cost of use, multiple optimal solution particles are sorted in ascending order, that is, the particles with the lowest cost are ranked first, and the particles with the highest cost are ranked last. The optimal solution particle ranked first after sorting (that is, the particle with the lowest cost) is taken as the optimal solution, and the type and dosage of flocculant corresponding to this optimal solution are combined as the optimal type of flocculant and the optimal dosage. This implementation method introduces the calculation of the cost of use, so that the optimization results are closer to the actual application needs, and achieves the technical effect of improving the practicality and feasibility of decision-making.
[0027] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
Claims
1. A method for optimizing the dosage of sludge dewatering flocculant, characterized in that: The method comprises: Collect and obtain a set of historical data in the sludge dewatering process, including a set of historical sludge concentrations, a set of historical pH values, a set of historical flocculant types, a set of historical flocculant dosages, and a set of historical flocculation effects, wherein the set of historical sludge concentrations, the set of historical pH values, the set of historical flocculant types, and the set of historical flocculant dosages are used as a set of sludge dewatering indicators; Taking the sludge dehydration index set as input parameters and the historical flocculation effect set as supervision parameters, training a flocculation effect prediction operator; The current sludge concentration and the current pH value are collected, and the type and dosage of the flocculant are optimized and analyzed based on the flocculation effect prediction operator to obtain the optimal type and dosage of the flocculant; The current sludge concentration and the current pH value are collected, and the type and dosage of flocculants are optimized and analyzed based on the flocculation effect prediction operator to obtain the optimal type and dosage of flocculants, including: According to the types of optional flocculants, a flocculant type space is constructed, and according to the minimum and maximum values in the historical flocculant dosage set, a flocculant dosage space is constructed; Based on the flocculant type space and the flocculant dosage space, randomly generate a first flocculant and a plurality of first initial dosages, and respectively combine the first flocculant and each of the plurality of first initial dosages to obtain a plurality of first initial particles; According to the current sludge concentration, the current pH value and the flocculation effect prediction operator, iteratively optimizing the multiple first initial particles to obtain the first optimal solution fitness; The first flocculant is removed from the flocculant type space, and after the removal, a second flocculant is randomly generated from the new flocculant type space, and a plurality of second initial dosages are randomly generated based on the flocculant dosage space, and the second flocculant is combined with each of the plurality of second initial dosages to obtain a plurality of second initial particles; Iteratively optimizing the plurality of second initial particles according to the current sludge concentration, the current pH value and the flocculation effect prediction operator to obtain the second optimal solution fitness; The same method is used to obtain the third optimal solution fitness, the fourth optimal solution fitness, and so on until the flocculant type space is empty, and multiple optimal solution fitnesses are obtained; The multiple optimal solution fitnesses are arranged in descending order, the fitness that ranks first is taken as the optimal fitness, and the flocculant and dosage corresponding to the optimal fitness are extracted as the optimal flocculant type and optimal dosage.
2. A method for optimizing the dosage of a sludge dewatering flocculant according to claim 1, characterized in that: The method of using the sludge dehydration index set as input parameters and the historical flocculation effect set as supervision parameters to train and obtain a flocculation effect prediction operator includes: Extracting a first historical flocculant type based on the historical flocculant type set; Traversing the historical data set, extracting a first historical sludge concentration set, a first historical pH value set, a first historical flocculant dosage set and a first historical flocculation effect set corresponding to the first historical flocculant type, wherein the first historical flocculant type, the first historical sludge concentration set, the first historical pH value set and the first historical flocculant dosage set are the first sludge dewatering indicator set; Taking the first sludge dehydration index set as input parameters and the first historical flocculation effect set as supervision parameters, training to obtain a first flocculation effect prediction operator; According to the first flocculation effect prediction operator, multiple first flocculation effect prediction operators corresponding to different flocculants are trained, and the multiple first flocculation effect prediction operators are integrated to obtain the flocculation effect prediction operator.
3. A method for optimizing the dosage of a sludge dewatering flocculant as claimed in claim 2, characterized in that: The first sludge dehydration index set is used as an input parameter, the first historical flocculation effect set is used as a supervision parameter, and the first flocculation effect prediction operator is trained to obtain the first flocculation effect prediction operator, including: The first historical flocculation effect set consists of a sludge moisture content set, a supernatant turbidity set, and a sludge volume reduction rate set; Taking the first sludge dehydration index set as input parameters, and taking the sludge moisture content set, the supernatant turbidity set and the sludge volume reduction rate set as supervision parameters, respectively, training to obtain a first sludge moisture content prediction operator, a first supernatant turbidity prediction operator and a first sludge volume reduction rate prediction operator; The first sludge moisture content prediction operator, the first supernatant turbidity prediction operator and the first sludge volume reduction rate prediction operator are connected in parallel to obtain the first flocculation effect prediction operator.
4. The method for optimizing the dosage of sludge dewatering flocculant according to claim 1, characterized in that: The iterative optimization of the plurality of first initial particles according to the current sludge concentration, the current pH value and the flocculation effect prediction operator to obtain the first optimal solution fitness includes: Construct a fitness evaluation function; Based on the current sludge concentration, the current pH value, the flocculation effect prediction operator and the fitness evaluation function, a plurality of first initial fitness of the plurality of first initial particles are calculated; Summing the multiple first initial fitnesses to obtain a first initial total fitness, taking the ratio of each first initial fitness to the first initial total fitness as a fitness probability, and obtaining multiple first initial fitness probabilities; Selecting the plurality of first initial fitness probabilities based on a roulette wheel to obtain a plurality of first initial selected fitnesses, wherein the plurality of first initial selected fitnesses correspond to a second quantity, the plurality of first initial fitnesses correspond to a first quantity, and the second quantity is less than the first quantity; Taking the plurality of first initial particles corresponding to the plurality of first initial selection fitness as the plurality of first initial retained particles, randomly generating a plurality of first supplementary particles in the flocculant dosage space, the plurality of first initial retained particles and the plurality of first supplementary particles together constitute a plurality of first updated particles, and the number of the plurality of first updated particles is the same as the first number; Iterative optimization is performed based on the multiple first updated particles, and a first optimal solution fitness is obtained after convergence.
5. A method for optimizing the dosage of sludge dewatering flocculant according to claim 4, characterized in that: The constructing of the fitness evaluation function comprises: The expression of the fitness evaluation function is: ； in, For the The adaptability of flocculants, For the history The maximum amount of flocculant added, For the history The minimum amount of flocculant added, is the current particle dosage, For the history The maximum value of the flocculation effect of the flocculant, For the history The minimum value of the flocculation effect of the flocculant, is the flocculation effect of the current particle, and are the weight coefficients of dosage and flocculation effect, respectively, and 1.
6. A method for optimizing the dosage of sludge dewatering flocculant according to claim 5, characterized in that: The method of calculating the multiple first initial fitness of the multiple first initial particles based on the current sludge concentration, the current pH value, the flocculation effect prediction operator and the fitness evaluation function includes: Inputting the current sludge concentration, the current pH value, and the plurality of first initial particles into the flocculation effect prediction operator to obtain a plurality of predicted first initial flocculation effects; The multiple first initial dosages and the multiple first initial flocculation effects are input into the fitness evaluation function to calculate and obtain multiple first initial fitnesses.
7. The method for optimizing the dosage of sludge dewatering flocculant according to claim 1, characterized in that: The step of arranging the multiple optimal solution fitnesses in descending order, taking the fitness that ranks first as the optimal fitness, and extracting the flocculant and dosage corresponding to the optimal fitness as the optimal flocculant type and optimal dosage includes: Extracting multiple optimal solution particles corresponding to the multiple optimal solution fitnesses, wherein each optimal solution particle corresponds to a flocculant and a dosage; The use costs of the multiple optimal solution particles are calculated based on the type and dosage of the flocculant, the multiple optimal solution fitnesses are arranged in ascending order according to the use costs, the fitness that ranks first is taken as the optimal fitness, and the flocculant and dosage corresponding to the optimal fitness are extracted as the optimal flocculant type and optimal dosage.
Citation Information
Patent Citations
PSO-LSSVM-based on-line control method and apparatus for dosing of water island
CN106371316A
Method and system for determining dosage of medicine for treating water-coal-slurry wastewater
CN117690518A