Water treatment optimization method and system based on microalgae growth prediction

Abstract

The invention relates to the technical field of biological sewage treatment, in particular to a water treatment optimization method and system based on microalgae growth prediction. Acquiring a real-time water quality pollution parameter set and real-time water environment data of the target water treatment area, constructing a real-time water environment simulation field by using the real-time water environment data, simulating and generating a microalgae future growth model through the real-time water environment simulation field, and performing spatial hash coding analysis and optimization treatment on the polluted water quality of the real-time water quality pollution parameter set; and combining the real-time water quality pollution parameter set and the real-time water environment data to construct a real-time water quality pollution scale thermal model of the target water treatment area, and querying and judging whether the water quality pollution scale of the local water treatment area is abnormal or not through the real-time water quality pollution scale thermal model to execute sewage treatment of microalgae growth regulation. According to the invention, corresponding water treatment operation can be carried out by analyzing the sewage purification performance of microalgae growth, so that water treatment rationalization is ensured, and the water quality is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of biological sewage treatment, and in particular to a water treatment optimization method and system based on microalgae growth prediction. Background Art

[0002] In the field of water treatment, microalgae has gradually become an important water treatment material due to its good water purification ability and environmental friendliness. Microalgae can absorb nutrients in water bodies through photosynthesis, especially pollutants such as nitrogen and phosphorus, thereby effectively reducing the degree of eutrophication of water bodies and improving water quality. However, due to the complexity of the water environment and microalgae growth conditions, the existing water treatment technology based on microalgae growth still has many limitations. On the one hand, the traditional methods for predicting microalgae growth mostly rely on empirical models or simple linear models, which are difficult to accurately characterize the effects of multi-factor interactions on microalgae growth rate and metabolite yield, which greatly reduces the prediction accuracy and credibility of microalgae growth. On the other hand, traditional water treatment methods usually cannot achieve optimal control of microalgae growth, making it difficult for the water treatment process to maintain the best effect under different environmental conditions, resulting in unstable treatment efficiency and high cost. Therefore, accurate prediction and optimization of water treatment processes based on microalgae growth have become a major difficulty in the current water treatment field. Therefore, it is necessary to develop a water treatment optimization method that can have extremely high prediction accuracy and can improve water treatment efficiency and reduce costs to solve the above problems. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a water treatment optimization method and system based on microalgae growth prediction.

[0004] To achieve the above object, the technical solution adopted by the present invention is: A first aspect of the present invention provides a water treatment optimization method based on microalgae growth prediction, comprising the following steps: S102: obtaining several environmental parameters of the target water treatment area and several growth parameters of the microalgae, training a microalgae growth prediction model through multiple regression calculation and residual splitting gradient optimization for each environmental parameter and each growth parameter, and performing future prediction, to obtain the future growth trend and future metabolic yield of the microalgae at the future water treatment node; S104: extracting multiple historical concentration detection parameters of one or more pollutants in the water quality, and performing logistic mapping iterations on each historical concentration detection parameter one by one based on the future growth trend and the future metabolic yield limit to obtain the purification performance of the microalgae on the water quality in the target water treatment area; S106: obtaining a real-time water quality pollution parameter set and real-time water environment data of the target water treatment area, constructing a real-time water environment simulation field using the real-time water environment data, simulating and generating a future growth model of microalgae through the real-time water environment simulation field, and performing spatial hash coding analysis of the real-time water quality pollution parameter set to optimize the treatment of polluted water quality; S108: Construct a real-time water pollution scale thermodynamic model of the target water treatment area in combination with the real-time water pollution parameter set and the real-time water environment data, and determine whether the water pollution scale of the local water treatment area is abnormal by querying the real-time water pollution scale thermodynamic model to perform wastewater treatment for microalgae growth regulation.

[0005] More specifically, the step S102 includes the following steps: Acquire a target water treatment area, and acquire several environmental parameters of the target water treatment area and several growth parameters of microalgae in the target water treatment area through online sensors; Acquiring the growth characteristics of the microalgae, constructing a multiple linear regression equation according to the growth characteristics, introducing the least square method, estimating the multiple linear regression equation in the least square method based on the several environmental parameters and the several growth parameters, and obtaining a series of regression coefficients; Determine the residual value of the multivariate linear regression equation for capturing the growth of microalgae under each environmental parameter and each growth parameter condition according to a series of regression coefficients, obtain multiple residual values, and calculate the splitting gain according to each combination of environmental parameter and each growth parameter condition, and preset different splitting points according to the splitting gain; Constructing a root node and presetting a minimum residual value, and traversing each splitting point one by one from the root node based on each residual value to split the several environmental parameters and the several growth parameters, so as to generate a gradient capture decision tree for microalgae growth; Obtaining the residual difference between the residual value and the preset minimum residual value, presetting the gradient learning rate according to the residual difference, weighting the gradient capture decision tree by the gradient learning rate to optimize the prediction result, and continuously repeating the above-mentioned iterations of splitting and constructing the gradient capture decision tree and weighted optimization steps, and finally outputting the trained microalgae growth prediction model; The treatment tasks of the target water treatment area are obtained, and the future water treatment nodes of the target water treatment area are extracted according to the treatment tasks. The actual environmental parameters and actual growth parameters of the microalgae at the future water treatment nodes are obtained, and the actual environmental parameters and actual growth parameters are predicted and calculated by the trained microalgae growth prediction model to obtain the future growth trend and future metabolic yield of the microalgae at the future water treatment nodes.

[0006] More specifically, the step S104 includes the following steps: Obtaining the treatment log and treatment tasks of the target water treatment area, and obtaining one or more pollutants in the water quality of the target water treatment area that need to be improved and treated according to the treatment tasks; Extracting the future growth trend of microalgae and the future metabolic output of microalgae when placing multiple historical concentration detection parameters of one or more polluting components in polluted water through the treatment log, and obtaining the preset detection strategy of the online sensor; Extract the detection node interval of each historical concentration detection parameter of the online sensor according to the preset detection strategy, and perform regular analysis based on the criterion of detecting each historical concentration detection parameter based on the detection node interval of one or more polluting components of the polluted water quality in the target water treatment area due to the future growth trend and future metabolic output of the microalgae, so as to obtain the temporal regularity of the microalgae growth affecting the polluted water quality; Based on the future growth trend and the future metabolic yield, a variable control parameter matrix is ​​constructed, a Logistic mapping algorithm is introduced, and based on the time sequence law, multiple historical concentration detection parameters of one or more polluting components are mapped to the state variables of each variable control parameter in the variable control parameter matrix one by one in the Logistic mapping algorithm to generate a Logistic mapping recursive equation for the effect of microalgae growth on polluted water quality; Iteratively solve the Logistic mapping recursive equation, and preset a periodic reference threshold according to the time sequence law. If the periodic changes of the future growth trend and the future metabolic output for the historical concentration detection parameters converge to the periodic reference threshold, the iterative operation is stopped, and a series of iterative trajectories are output at this time; A trajectory bifurcation diagram is drawn according to a series of iterative trajectories, and the trajectory trends in the trajectory bifurcation diagram are analyzed to determine the purification performance of the microalgae on the polluted water in the target water treatment area under the conditions of future growth trends and future metabolites.

[0007] More specifically, the step S106 includes the following steps: The water quality of the target water treatment area is detected in real time by online sensors to obtain the real-time water pollution parameter set and real-time water environment data of the target water treatment area, and to construct the spatiotemporal discretization grid of the target water treatment area; A central difference method is introduced to solve the difference equation of the real-time water environment data to obtain a solution of the real-time water environment difference equation, and the values ​​of the grid points on the iterative spatiotemporal discretization grid are cyclically updated by the solution of the real-time water environment difference equation to obtain a real-time water environment simulation field of the target water treatment area; Through the real-time water environment simulation field, the growth of the microalgae with future growth trend and future metabolic yield is simulated to obtain a future growth model of the microalgae, and a spatial hash grid unit and index table are constructed based on the purification performance; The future growth model of the microalgae is divided into a plurality of sub-model objects by using a spatial hash grid unit, and the coordinates of each sub-model object are mapped to an index table of the spatial hash grid unit to obtain a spatial hash value of each sub-model object; A spatial hash table is created, and based on the spatial hash value, each sub-model object belonging to a certain grid unit is stored under the key of the grid unit for hash coding, and finally a spatial hash coding table of the purification performance of the future growth model of microalgae is obtained; A three-dimensional geographic terrain model of the target water treatment area is constructed, and a predetermined pollution absorption area is constrained and planned within the three-dimensional geographic terrain model based on purification performance. The above steps are repeated to perform hash coding on the predetermined pollution absorption area to obtain a spatial hash decoding sequence of the predetermined pollution absorption space area, and a spatial hash coding table is used to perform water treatment optimization on the microalgae purification water quality analysis of the spatial hash decoding sequence.

[0008] More specifically, the three-dimensional geographical terrain model of the target water treatment area is constructed, and a predetermined pollution absorption area is constrained and planned in the three-dimensional geographical terrain model based on purification performance, and the above steps are repeated to perform hash coding on the predetermined pollution absorption area to obtain a spatial hash decoding sequence of the predetermined pollution absorption space area, and the water treatment optimization is performed on the microalgae purification water quality analysis of the spatial hash decoding sequence using the spatial hash coding table, which specifically includes the following steps: Preset the purification radiation boundary threshold according to the purification performance and construct a three-dimensional geographical terrain model of the target water treatment area, plan the pollution absorption range within the three-dimensional geographical terrain model with the purification radiation boundary threshold as the range constraint, and obtain the established pollution absorption range of the target water treatment area by the future growth model of microalgae; Extract multiple pollution component concentration parameters of water quality in a predetermined pollution absorption action area through a real-time water quality pollution parameter set of a target water treatment area, repeat the above-mentioned spatial hash code construction step based on the multiple pollution component concentration parameters to perform hash coding on the regional spatial pollution degree of the predetermined pollution absorption action area, and obtain a spatial hash decoding sequence of the predetermined pollution absorption spatial area; Determine whether the spatial hash decoding sequence can decode and output the spatial hash coding table. If not, obtain the missing tampering code that the spatial hash coding table cannot decode and output using the spatial hash decoding sequence, and determine the purification performance gap of the microalgae growth for the polluted water quality in the target water treatment area according to the missing tampering code; Based on the purification performance gap, the growth conditions of microalgae in the given pollution absorption space area are improved and optimized to enhance the water pollution purification efficiency of microalgae growth in the given pollution absorption space area.

[0009] More specifically, the step S108 includes the following steps: Constructing a random distribution framework of water environment flow in a three-dimensional geographic terrain model based on a real-time water environment simulation field of a target water treatment area, and extracting a random node array of the random distribution framework; Obtaining neighbor nodes of each random node in the random node array, calculating the weight between each random node and each neighbor node based on the real-time water quality pollution parameter set of the target water treatment area, generating weight values ​​of each random node corresponding to a number of random node boundaries, and determining the random probability distribution of real-time water quality pollution according to the ratio between the weight values; Preset a maximum number of steps, update the current random node to a neighbor node selected by the maximum random probability according to the random probability distribution, repeat the weight calculation between the random node and the neighbor node and the steps of random node update to continuously execute the next round of random walk until the random walk reaches the maximum number of steps, and generate a real-time water quality pollution scale model of the target water treatment area; Based on the big data network, a benchmark thermal color gamut rule for different water pollution scale ratings is obtained, and the real-time water pollution scale model is transformed by the benchmark thermal color gamut rule to obtain a real-time water pollution scale thermal model of the target water treatment area; Obtaining an online sensor layout pattern of a target water treatment area, dividing the real-time water pollution scale thermal model into M sub-model area blocks based on the online sensor layout pattern, and obtaining a real-time thermal color gamut RGB lookup table of each sub-model area block; According to the purification performance of microalgae for different water pollution scales, abnormal thermochromatic RGB values ​​of unreasonable water pollution scales are preset. If the real-time thermochromatic RGB query table can query the abnormal thermochromatic RGB values, the sub-model area block is marked as an abnormal water pollution scale area and the sewage treatment operation of microalgae growth regulation is performed.

[0010] A second aspect of the present invention provides a water treatment optimization system based on microalgae growth prediction, the water treatment optimization system comprising a memory and a processor, the memory storing a water treatment optimization method program based on microalgae growth prediction, when the water treatment optimization method program is executed by the processor, any one of the water treatment optimization method steps is implemented.

[0011] The present invention solves the technical defects existing in the background technology, and the beneficial technical effects of the present invention are: The method comprises the following steps: obtaining a plurality of environmental parameters of the target water treatment area and a plurality of growth parameters of the microalgae, training a microalgae growth prediction model through multiple regression calculation and residual splitting gradient optimization of each environmental parameter and each growth parameter, and performing future prediction, so as to obtain the future growth trend and future metabolic yield of the microalgae when the microalgae is at the future water treatment node; extracting a plurality of historical concentration detection parameters of one or more polluting components in the water quality, and performing Logistic mapping iteration on each historical concentration detection parameter one by one based on the future growth trend and the future metabolic yield limit, so as to obtain the purification performance of the microalgae on the water quality in the target water treatment area; obtaining a real-time water quality pollution parameter set and real-time water environment data of the target water treatment area, constructing a real-time water environment simulation field using the real-time water environment data, simulating and generating a future growth model of the microalgae through the real-time water environment simulation field, and performing spatial hash coding analysis and optimization of the real-time water quality pollution parameter set to treat the polluted water quality; constructing a real-time water quality pollution scale thermal model of the target water treatment area in combination with the real-time water quality pollution parameter set and the real-time water environment data, and querying and determining whether the water quality pollution scale of the local water treatment area is abnormal or not through the real-time water quality pollution scale thermal model to perform sewage treatment with microalgae growth regulation. The present invention can quickly and accurately analyze the purification performance of microalgae for polluted water in a target water treatment area by predicting the growth trend and metabolic output of microalgae at future time nodes, and further analyze and execute corresponding reasonable water treatment operations based on the purification performance, thereby utilizing the characteristics of microalgae in absorbing pollutants in water to efficiently purify sewage, ensure rational water management, and effectively improve water quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0013] Figure 1 A first method flow chart of a water treatment optimization method based on microalgae growth prediction is shown; Figure 2 A second method flow chart of a water treatment optimization method based on microalgae growth prediction is shown; Figure 3 A system framework diagram of a water treatment optimization system based on microalgae growth prediction is shown; Figure 4 A schematic diagram of analyzing the microscopic image of microalgae of the present invention is shown; Figure 5 A schematic diagram of the model for capturing the decision tree of microalgae growth gradient is shown; Figure 6 The figure shows a brief block diagram of the data processing flow of a water treatment optimization system based on microalgae growth prediction of the present invention. DETAILED DESCRIPTION

[0014] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0015] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0016] The first aspect of the present invention provides a water treatment optimization method based on microalgae growth prediction, such as Figure 1 As shown, the following steps are included: S102: obtaining several environmental parameters of the target water treatment area and several growth parameters of the microalgae, training a microalgae growth prediction model through multiple regression calculation and residual splitting gradient optimization for each environmental parameter and each growth parameter, and performing future prediction, to obtain the future growth trend and future metabolic yield of the microalgae at the future water treatment node; S104: extracting multiple historical concentration detection parameters of one or more pollutants in the water quality, and performing logistic mapping iterations on each historical concentration detection parameter one by one based on the future growth trend and the future metabolic yield limit to obtain the purification performance of the microalgae on the water quality in the target water treatment area; S106: obtaining a real-time water quality pollution parameter set and real-time water environment data of the target water treatment area, constructing a real-time water environment simulation field using the real-time water environment data, simulating and generating a future growth model of microalgae through the real-time water environment simulation field, and performing spatial hash coding analysis of the real-time water quality pollution parameter set to optimize the treatment of polluted water quality; S108: Construct a real-time water pollution scale thermodynamic model of the target water treatment area in combination with the real-time water pollution parameter set and the real-time water environment data, and determine whether the water pollution scale of the local water treatment area is abnormal by querying the real-time water pollution scale thermodynamic model to perform wastewater treatment for microalgae growth regulation.

[0017] More specifically, the step S102 includes the following steps: Acquire a target water treatment area, and acquire several environmental parameters of the target water treatment area and several growth parameters of microalgae in the target water treatment area through online sensors; Acquiring the growth characteristics of the microalgae, constructing a multiple linear regression equation according to the growth characteristics, introducing the least square method, estimating the multiple linear regression equation in the least square method based on the several environmental parameters and the several growth parameters, and obtaining a series of regression coefficients; Determine the residual value of the multivariate linear regression equation for capturing the growth of microalgae under each environmental parameter and each growth parameter condition according to a series of regression coefficients, obtain multiple residual values, and calculate the splitting gain according to each combination of environmental parameter and each growth parameter condition, and preset different splitting points according to the splitting gain; Constructing a root node and presetting a minimum residual value, and traversing each splitting point one by one from the root node based on each residual value to split the several environmental parameters and the several growth parameters, so as to generate a gradient capture decision tree for microalgae growth; Obtaining the residual difference between the residual value and the preset minimum residual value, presetting the gradient learning rate according to the residual difference, weighting the gradient capture decision tree by the gradient learning rate to optimize the prediction result, and continuously repeating the above-mentioned iterations of splitting and constructing the gradient capture decision tree and weighted optimization steps, and finally outputting the trained microalgae growth prediction model; The treatment tasks of the target water treatment area are obtained, and the future water treatment nodes of the target water treatment area are extracted according to the treatment tasks. The actual environmental parameters and actual growth parameters of the microalgae at the future water treatment nodes are obtained, and the actual environmental parameters and actual growth parameters are predicted and calculated by the trained microalgae growth prediction model to obtain the future growth trend and future metabolic yield of the microalgae at the future water treatment nodes.

[0018] It should be noted that environmental parameters include light intensity, temperature, humidity, pH value, carbon dioxide concentration and nutrient concentration, and growth parameters include cell density, chlorophyll content, biomass concentration and specific metabolite concentration. Microalgae can absorb nutrients (such as nitrogen, phosphorus), heavy metals and other pollutants in water to purify water quality, while producing oxygen and improving the ecological environment of water bodies. Reasonable microalgae growth can achieve natural plant water purification, so it is necessary to accurately predict the growth of microalgae to improve the quality of water treatment. However, traditional methods for predicting microalgae growth mostly rely on empirical models or simple linear models. It is difficult to accurately characterize the impact of multi-factor interactions on the growth rate and metabolite yield of microalgae, and it is difficult to effectively mine the complex nonlinear relationships implicit in the acquired data and predict the growth trend of microalgae in real time. In view of this, this method uses the growth characteristics of microalgae as the basis to construct a multivariate linear regression equation about the standardized growth of microalgae to solve the environmental parameters and growth parameters obtained, so that the regression coefficient of each independent variable (environmental parameter) can be used to predict the dependent variable (growth parameter) for descriptive learning, which can ensure that the prediction of microalgae growth is more accurate and reasonable compared with traditional prediction methods. In order to further improve the accuracy and robustness of prediction model training, this method splits and trains a tree by fitting the residual between the predicted value and the actual value of the multivariate linear regression equation. The residual can clearly identify the error position of the prediction model, so that the next step of the tree can make up for and avoid these errors in the prediction of microalgae growth. This method splits several environmental parameters and several growth parameters based on each residual value, traversing each split point one by one from the root node, which can ensure the causal linearity of the independent variable affecting the dependent variable of microalgae growth, thereby reducing the prediction error and improving the model performance.

[0019] It should be noted that this method uses the gradient learning rate weighted update of the residual gap to enable the new tree to correct the previous error to a certain extent; the gradient learning rate controls the contribution of each tree in the final prediction, avoiding overfitting, so that each tree tries to correct the error of the previous tree, so that the construction of each tree helps the model to gradually correct the current prediction error, so that the growth trend and metabolic yield of microalgae at the future water treatment node are more accurate, and the accuracy of microalgae growth prediction is improved. Through this method, the prediction of microalgae growth based on environmental parameters and growth parameters can be made more accurate by using the gradient optimization method of multivariate linear regression and residual fitting splitting. Compared with the traditional prediction model, it can further improve the credibility of the prediction results and reduce the water treatment optimization error caused by the prediction error.

[0020] More specifically, the step S104 includes the following steps: Obtaining the treatment log and treatment tasks of the target water treatment area, and obtaining one or more pollutants in the water quality of the target water treatment area that need to be improved and treated according to the treatment tasks; Extracting the future growth trend of microalgae and the future metabolic output of microalgae when placing multiple historical concentration detection parameters of one or more polluting components in polluted water through the treatment log, and obtaining the preset detection strategy of the online sensor; Extract the detection node interval of each historical concentration detection parameter of the online sensor according to the preset detection strategy, and perform regular analysis based on the criterion of detecting each historical concentration detection parameter based on the detection node interval of one or more polluting components of the polluted water quality in the target water treatment area due to the future growth trend and future metabolic output of the microalgae, so as to obtain the temporal regularity of the microalgae growth affecting the polluted water quality; Based on the future growth trend and the future metabolic yield, a variable control parameter matrix is ​​constructed, a Logistic mapping algorithm is introduced, and based on the time sequence law, multiple historical concentration detection parameters of one or more polluting components are mapped to the state variables of each variable control parameter in the variable control parameter matrix one by one in the Logistic mapping algorithm to generate a Logistic mapping recursive equation for the effect of microalgae growth on polluted water quality; Iteratively solve the Logistic mapping recursive equation, and preset a periodic reference threshold according to the time sequence law. If the periodic changes of the future growth trend and the future metabolic output for the historical concentration detection parameters converge to the periodic reference threshold, the iterative operation is stopped, and a series of iterative trajectories are output at this time; A trajectory bifurcation diagram is drawn according to a series of iterative trajectories, and the trajectory trends in the trajectory bifurcation diagram are analyzed to determine the purification performance of the microalgae on the polluted water in the target water treatment area under the conditions of future growth trends and future metabolites.

[0021] It should be noted that the pollutants include nitrates, nitrogen, phosphorus and heavy metals. Since microalgae can absorb nutrients (such as nitrogen, phosphorus), heavy metals and other pollutants in water, that is, microalgae have a certain ability to purify sewage, the growth trend and metabolic yield of microalgae determine the purification performance of microalgae in absorbing pollutants in sewage. This purification performance is crucial for the optimization of sewage treatment and directly affects whether the sewage treatment effect can reach the expected result. Therefore, it is necessary to analyze and calculate the purification performance of microalgae for pollutants in sewage based on the predicted future growth trend and future metabolic yield. This method first analyzes and constructs the temporal law of the effect of microalgae growth on polluted water quality. Since microalgae gradually absorb pollutants in sewage over time, their concentration changes. In this process, the growth trend and metabolites of microalgae also change synchronously over time. Therefore, the growth trend of microalgae and the change of metabolites determine the concentration change of the purified pollutants. This temporal law reveals that the concentration of pollutants in sewage is recursively purified by microalgae growth according to the time sequence. The variable mapping following this law can make the analysis and inference of purification performance more accurate and reliable. Then, the Logistic mapping algorithm is used to map the historical concentration detection parameters of the pollutants to the variable control parameter matrix constructed by the future growth trend and the future metabolic output based on the premise of following the time series law, so as to systematically describe and express the state behavior of the microalgae to the change of the concentration of the pollutants when the growth and metabolism change continuously. Since the state behavior is based on time series, the current state description depends on the former state. Therefore, this method iteratively solves the Logistic mapping recursive equation of the effect of microalgae growth on the polluted water quality and stops the iterative operation when the periodic reference threshold is reached; the Chinese name of the Logistic mapping algorithm is the logistic mapping algorithm, which is a recursive algorithm used to describe the state behavior of some dynamic variables affecting other dynamic variables in time series; this can make the purification behavior of the microalgae growth affecting the polluted water quality maintain a certain time series and stability, avoid the occurrence of time series oscillation and enter the chaotic state of the system, which affects the inference of the final purification performance, and greatly improve the calculation accuracy and credibility of the purification performance.

[0022] It should be noted that the bifurcation diagram drawn by the iterative trajectory records the steady-state or periodic orbit of the system for each historical concentration detection parameter. It can be seen that there are different branches or trajectories in the bifurcation diagram, which is manifested as the system jumping from one stable point to another, reflecting the purification ability of microalgae to absorb pollutants in sewage in a timely manner. Through this method, the future growth trend predicted by microalgae growth and the future metabolic yield state variable mapping analysis of the historical concentration detection parameters of the pollutants can be used to describe the dynamic behavior of microalgae growth absorbing and changing pollutants in sewage, so as to more accurately and efficiently infer and determine the purification performance of microalgae growth on polluted water quality in the target water treatment area, which provides a reliable analysis basis for the subsequent sewage treatment optimization based on microalgae growth, and effectively improves the accuracy and rationality of water treatment optimization.

[0023] More specifically, the step S106 includes the following steps: The water quality of the target water treatment area is detected in real time by online sensors to obtain the real-time water pollution parameter set and real-time water environment data of the target water treatment area, and to construct the spatiotemporal discretization grid of the target water treatment area; A central difference method is introduced to solve the difference equation of the real-time water environment data to obtain a solution of the real-time water environment difference equation, and the values ​​of the grid points on the iterative spatiotemporal discretization grid are cyclically updated by the solution of the real-time water environment difference equation to obtain a real-time water environment simulation field of the target water treatment area; Through the real-time water environment simulation field, the growth trend and future metabolic yield of microalgae are simulated to obtain a future growth model of microalgae, and a spatial hash grid unit and index table are constructed based on the purification performance; The future growth model of the microalgae is divided into a plurality of sub-model objects by using a spatial hash grid unit, and the coordinates of each sub-model object are mapped to an index table of the spatial hash grid unit to obtain a spatial hash value of each sub-model object; A spatial hash table is created, and based on the spatial hash value, each sub-model object belonging to a certain grid unit is stored under the key of the grid unit for hash coding, and finally a spatial hash coding table of the purification performance of the future growth model of microalgae is obtained; A three-dimensional geographic terrain model of the target water treatment area is constructed, and a predetermined pollution absorption area is constrained and planned within the three-dimensional geographic terrain model based on purification performance. The above steps are repeated to perform hash coding on the predetermined pollution absorption area to obtain a spatial hash decoding sequence of the predetermined pollution absorption space area, and a spatial hash coding table is used to perform water treatment optimization on the microalgae purification water quality analysis of the spatial hash decoding sequence.

[0024] It should be noted that microalgae can survive in water for a period of time and then continue to reproduce and grow to form microalgae clusters of different sizes and uneven distribution. Under the premise of determined purification performance, there are differences in the scale of sewage treated by microalgae clusters of different sizes. For example, the sewage pollution degree in a local area of ​​the target water treatment area is high, and the microalgae growth specifications in the local area are small. Therefore, based on the premise of purification performance, it can be known that the microalgae has a poor absorption capacity for the pollutant components in the local area, making it difficult to completely purify and treat it. Therefore, it is necessary to further optimize it to improve the water pollution condition in the local area. It can be seen from this that the water treatment optimization operation can be determined by analyzing the degree of pollution purification of microalgae in the corresponding sewage area. In this regard, this method constructs a real-time water environment simulation field of the target water treatment area by introducing the central difference method to solve the difference equation of the real-time water environment data, and uses the real-time water environment simulation field to simulate the future growth trend and future metabolic output, which can more accurately visualize the real-time specifications of the future growth of microalgae, so that the subsequent analysis and calculation of the degree of pollution purification in the polluted area is more accurate and stable.

[0025] It should be noted that the traditional method usually constructs a pollution purification performance model of microalgae growth based on the purification performance into the model formula and conducts a simple comparative analysis with the pollution degree model of the pollution area in which it is located. This not only increases the judgment error of microalgae purifying sewage, but also increases unnecessary calculation amount, making it difficult to ensure the efficiency and quality of water treatment optimization. Therefore, this method constructs a spatial hash grid unit and index table based on purification performance, and then divides the future growth model of microalgae through the spatial hash grid unit and subdivides it into the index table of the spatial hash grid unit. The grid unit and its neighborhood where the microalgae space grows can be directly located by the hash value, without traversing all model areas, so that the overall pollution purification capacity under the microalgae space growth specification based on purification performance description is more accurate, and then a spatial hash table is created to store all sub-model objects belonging to a certain grid unit under the key of the grid unit for hash coding processing, so that the overall pollution purification capacity under the microalgae space growth specification is reflected in the form of a spatial hash coding table, and only the table needs to be used for decoding and judgment in the future, removing redundant calculation steps and improving the efficiency of water treatment optimization. Finally, the spatial hash coding sequence of the pollution degree of the area is calculated by the same method. The spatial hash coding sequence is used to decode the spatial hash coding table to quickly determine whether the water pollution in the area can be completely purified under the microalgae spatial growth specifications, so as to make decisions on the corresponding water treatment optimization operations. This method can analyze and calculate whether the microalgae growth scale can completely purify the water pollution degree in the area, and make corresponding water treatment optimization decisions, thereby improving the rationality, efficiency and accuracy of water treatment optimization, and ensuring that the water quality purification treatment under the conditions of microalgae growth achieves the expected effect.

[0026] More specifically, the three-dimensional geographical terrain model of the target water treatment area is constructed, and a predetermined pollution absorption area is constrained and planned in the three-dimensional geographical terrain model based on purification performance, and the above steps are repeated to perform hash coding on the predetermined pollution absorption area to obtain a spatial hash decoding sequence of the predetermined pollution absorption space area, and the water treatment optimization is performed on the microalgae purification water quality analysis of the spatial hash decoding sequence using the spatial hash coding table, which specifically includes the following steps: Preset the purification radiation boundary threshold according to the purification performance and construct a three-dimensional geographical terrain model of the target water treatment area, plan the pollution absorption range within the three-dimensional geographical terrain model with the purification radiation boundary threshold as the range constraint, and obtain the established pollution absorption range of the target water treatment area by the future growth model of microalgae; Extract multiple pollution component concentration parameters of water quality in a predetermined pollution absorption action area through a real-time water quality pollution parameter set of a target water treatment area, repeat the above-mentioned spatial hash code construction step based on the multiple pollution component concentration parameters to perform hash coding on the regional spatial pollution degree of the predetermined pollution absorption action area, and obtain a spatial hash decoding sequence of the predetermined pollution absorption spatial area; Determine whether the spatial hash decoding sequence can decode and output the spatial hash coding table. If not, obtain the missing tampering code that the spatial hash coding table cannot decode and output using the spatial hash decoding sequence, and determine the purification performance gap of the microalgae growth for the polluted water quality in the target water treatment area according to the missing tampering code; Based on the purification performance gap, the growth conditions of microalgae in the given pollution absorption space area are improved and optimized to enhance the water pollution purification efficiency of microalgae growth in the given pollution absorption space area.

[0027] It should be noted that microalgae clusters of different sizes have a certain range of pollution component treatment. Therefore, in order to ensure the sewage treatment effect in the area where each microalgae cluster is located, this method presets the purification radiation boundary threshold constraint planning target water treatment area three-dimensional geographical terrain model through purification performance, so that each microalgae growth cluster can obtain the maximum effective range of radiation absorption, and then it can improve the water treatment optimization decision accuracy of whether the growth scale of each microalgae cluster can completely purify the water pollution degree in the area. If the spatial hash decoding sequence cannot decode the output spatial hash coding table, it means that the purification capacity of the microalgae cluster is difficult to completely remove the pollution degree in the established pollution absorption area, which means that the purification capacity of the microalgae cluster is far from enough, and there is a purification performance gap. Therefore, the growth conditions of the microalgae can be improved and optimized according to the purification performance gap, so that the microalgae can be fully grown to improve its own purification adaptability and ability for different pollution degrees, and achieve the sewage treatment optimization effect based on the coupling effect of microalgae growth, which greatly improves the water treatment efficiency and treatment quality, and reduces the cumbersome steps and optimization errors caused by the traditional method without unified global optimization under the premise of the performance distribution of microalgae growth scale, and has high economic benefits.

[0028] More specifically, the step S108 is as follows: Figure 2 As shown, the specific steps include: S202: constructing a random distribution framework of water environment flow in a three-dimensional geographical terrain model based on a real-time water environment simulation field of a target water treatment area, and extracting a random node array of the random distribution framework; S204: Obtain neighbor nodes of each random node in the random node array, calculate the weight between each random node and each neighbor node based on the real-time water quality pollution parameter set of the target water treatment area, generate weight values ​​of each random node corresponding to a number of random node boundaries, and determine the random probability distribution of real-time water quality pollution according to the ratio between the weight values; S206: Preset a maximum number of steps, update the current random node to a neighbor node selected by the maximum random probability according to the random probability distribution, repeat the weight calculation between the random node and the neighbor node and the random node update steps to continuously perform the next round of random walk until the random walk reaches the maximum number of steps, and generate a real-time water quality pollution scale model of the target water treatment area; S208: Obtaining a benchmark thermal color gamut rule for different water pollution scale ratings based on a big data network, performing a color gamut assignment transformation on the real-time water pollution scale model using the benchmark thermal color gamut rule, and obtaining a real-time water pollution scale thermal model of the target water treatment area; S210: Acquire the online sensor layout pattern of the target water treatment area, divide the real-time water pollution scale thermal model into M sub-model area blocks based on the online sensor layout pattern, and acquire the real-time thermal color gamut RGB lookup table of each sub-model area block; S212: Preset abnormal thermochromatic RGB values ​​of unreasonable water pollution scales according to the purification performance of microalgae for different water pollution scales; if the real-time thermochromatic RGB query table can find abnormal thermochromatic RGB values, mark the sub-model area block as an abnormal water pollution scale area and perform a wastewater treatment operation of microalgae growth regulation.

[0029] It should be noted that existing water treatment facilities are usually equipped with online sensors for real-time monitoring. Once the content of pollutants in the water body is found to be high, the water treatment optimization operation can be quickly responded. However, the real-time water quality monitoring method used in water treatment has poor ability to distinguish the scale of water pollution, and because the water environment is always in a dynamic state of continuous flow, the pollutants in the water body will be randomly distributed with the water flow. At this time, it is more difficult for traditional methods to detect the degree and scale of pollution in the water body area, thereby affecting the accuracy and response rate of sewage treatment. In this regard, this method first constructs a random distribution framework of water environment flow in a three-dimensional geographic terrain model, and uses a real-time water quality pollution parameter set to perform a random walk of the weight of the current random node relative to the neighboring node on the random node array in the random distribution framework, so as to describe the real-time water quality pollution scale of the pollutants in the sewage with the continuous flow of water, so that the water quality monitoring follows the randomness driven by the water flow, and can improve the accuracy of the online sensor detection and analysis of the real-time water quality pollution scale of the target water treatment area. Compared with the traditional method, it solves the problem of difficulty in real-time and accurate detection of the pollution scale in the water body.

[0030] It should be noted that this method then transforms the real-time water quality pollution scale model into the form of a thermal color gamut, so as to obtain a query table expressing the thermal color gamut of different local areas in the target water treatment area, that is, a real-time thermal color gamut RGB query table for each sub-model area block; if the real-time thermal color gamut RGB query table can query abnormal thermal chromaticity RGB values, it means that the microalgae in the water treatment area corresponding to the sub-model area block cannot effectively handle the current real-time pollution scale, that is, the pollution scale of the water treatment area is an abnormal phenomenon and a rapid response is urgently needed to perform water treatment optimization, so it is marked as an abnormal water quality pollution scale area and the sewage treatment operation of microalgae growth regulation is performed. The RGB query method of the thermal color gamut can greatly improve the decision-making and response rate of the sewage treatment operation, reduce the error amount caused by the direct calculation and comparison of the pollution scale by the traditional method, optimize the real-time monitoring quality of the online sensor, and have high reliability. This method can use the pollutant concentration data detected by random walk in the real-time water environment simulation field of the target water treatment area to ensure that the water quality pollution scale monitoring and analysis of the online sensor more closely follows the water environment movement state, and use the RGB query judgment of the thermal color domain to quickly decide and respond to water treatment optimization operations with different water pollution scales, reduce the water treatment optimization error phenomenon caused by improper traditional monitoring, ensure the rationality of water treatment, and improve the quality and efficiency of water treatment optimization.

[0031] A second aspect of the present invention provides a water treatment optimization system based on microalgae growth prediction, such as Figure 3 As shown, the water treatment optimization system includes a memory 31 and a processor 32. The memory 31 stores a water treatment optimization method program based on microalgae growth prediction. When the water treatment optimization method program is executed by the processor 32, any one of the water treatment optimization method steps is implemented.

[0032] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A water treatment optimization method based on microalgae growth prediction, characterized in that: The following steps are involved: S102: obtaining several environmental parameters of the target water treatment area and several growth parameters of the microalgae, training a microalgae growth prediction model through multiple regression calculation and residual splitting gradient optimization for each environmental parameter and each growth parameter, and performing future prediction, to obtain the future growth trend and future metabolic yield of the microalgae at the future water treatment node; S104: extracting multiple historical concentration detection parameters of one or more pollutants in the water quality, and performing logistic mapping iterations on each historical concentration detection parameter one by one based on the future growth trend and the future metabolic yield limit to obtain the purification performance of the microalgae on the water quality in the target water treatment area; S106: obtaining a real-time water quality pollution parameter set and real-time water environment data of the target water treatment area, constructing a real-time water environment simulation field using the real-time water environment data, simulating and generating a future growth model of microalgae through the real-time water environment simulation field, and performing spatial hash coding analysis of the real-time water quality pollution parameter set to optimize the treatment of polluted water quality; S108: Construct a real-time water pollution scale thermodynamic model of the target water treatment area in combination with the real-time water pollution parameter set and the real-time water environment data, and determine whether the water pollution scale of the local water treatment area is abnormal by querying the real-time water pollution scale thermodynamic model to perform wastewater treatment for microalgae growth regulation.

2. A water treatment optimization method based on microalgae growth prediction according to claim 1, characterized in that: The step S102 specifically includes the following steps: Acquire a target water treatment area, and acquire several environmental parameters of the target water treatment area and several growth parameters of microalgae in the target water treatment area through online sensors; Acquiring the growth characteristics of the microalgae, constructing a multiple linear regression equation according to the growth characteristics, introducing the least square method, estimating the multiple linear regression equation in the least square method based on the several environmental parameters and the several growth parameters, and obtaining a series of regression coefficients; Determine the residual value of the multivariate linear regression equation for capturing the growth of microalgae under each environmental parameter and each growth parameter condition according to a series of regression coefficients, obtain multiple residual values, and calculate the splitting gain according to each combination of environmental parameter and each growth parameter condition, and preset different splitting points according to the splitting gain; Constructing a root node and presetting a minimum residual value, and traversing each splitting point one by one from the root node based on each residual value to split the several environmental parameters and the several growth parameters, so as to generate a gradient capture decision tree for microalgae growth; Obtaining the residual difference between the residual value and the preset minimum residual value, presetting the gradient learning rate according to the residual difference, weighting the gradient capture decision tree by the gradient learning rate to optimize the prediction result, and continuously repeating the above-mentioned iterations of splitting and constructing the gradient capture decision tree and weighted optimization steps, and finally outputting the trained microalgae growth prediction model; The treatment tasks of the target water treatment area are obtained, and the future water treatment nodes of the target water treatment area are extracted according to the treatment tasks. The actual environmental parameters and actual growth parameters of the microalgae at the future water treatment nodes are obtained, and the actual environmental parameters and actual growth parameters are predicted and calculated by the trained microalgae growth prediction model to obtain the future growth trend and future metabolic yield of the microalgae at the future water treatment nodes.

3. The water treatment optimization method based on microalgae growth prediction according to claim 1, characterized in that: The step S104 specifically includes the following steps: Obtaining the treatment log and treatment tasks of the target water treatment area, and obtaining one or more pollutants in the water quality of the target water treatment area that need to be improved and treated according to the treatment tasks; Extracting the future growth trend of microalgae and the future metabolic output of microalgae when placing multiple historical concentration detection parameters of one or more polluting components in polluted water through the treatment log, and obtaining the preset detection strategy of the online sensor; Extract the detection node interval of each historical concentration detection parameter of the online sensor according to the preset detection strategy, and perform regular analysis based on the criterion of detecting each historical concentration detection parameter based on the detection node interval of one or more polluting components of the polluted water quality in the target water treatment area due to the future growth trend and future metabolic output of the microalgae, so as to obtain the temporal regularity of the microalgae growth affecting the polluted water quality; Based on the future growth trend and the future metabolic yield, a variable control parameter matrix is ​​constructed, a Logistic mapping algorithm is introduced, and based on the time sequence law, multiple historical concentration detection parameters of one or more polluting components are mapped to the state variables of each variable control parameter in the variable control parameter matrix one by one in the Logistic mapping algorithm to generate a Logistic mapping recursive equation for the effect of microalgae growth on polluted water quality; Iteratively solve the Logistic mapping recursive equation, and preset a periodic reference threshold according to the time sequence law. If the periodic changes of the future growth trend and the future metabolic output for the historical concentration detection parameters converge to the periodic reference threshold, the iterative operation is stopped, and a series of iterative trajectories are output at this time; A trajectory bifurcation diagram is drawn according to a series of iterative trajectories, and the trajectory trends in the trajectory bifurcation diagram are analyzed to determine the purification performance of the microalgae on the polluted water in the target water treatment area under the conditions of future growth trends and future metabolites.

4. The water treatment optimization method based on microalgae growth prediction according to claim 1, characterized in that: The step S106 specifically includes the following steps: The water quality of the target water treatment area is detected in real time by online sensors to obtain the real-time water pollution parameter set and real-time water environment data of the target water treatment area, and to construct the spatiotemporal discretization grid of the target water treatment area; A central difference method is introduced to solve the difference equation of the real-time water environment data to obtain a solution of the real-time water environment difference equation, and the values ​​of the grid points on the iterative spatiotemporal discretization grid are cyclically updated by the solution of the real-time water environment difference equation to obtain a real-time water environment simulation field of the target water treatment area; Through the real-time water environment simulation field, the growth of the microalgae with future growth trend and future metabolic yield is simulated to obtain a future growth model of the microalgae, and a spatial hash grid unit and index table are constructed based on the purification performance; The future growth model of the microalgae is divided into a plurality of sub-model objects by using a spatial hash grid unit, and the coordinates of each sub-model object are mapped to an index table of the spatial hash grid unit to obtain a spatial hash value of each sub-model object; A spatial hash table is created, and based on the spatial hash value, each sub-model object belonging to a certain grid unit is stored under the key of the grid unit for hash coding, and finally a spatial hash coding table of the purification performance of the future growth model of microalgae is obtained; A three-dimensional geographic terrain model of the target water treatment area is constructed, and a predetermined pollution absorption area is constrained and planned within the three-dimensional geographic terrain model based on purification performance. The above steps are repeated to perform hash coding on the predetermined pollution absorption area to obtain a spatial hash decoding sequence of the predetermined pollution absorption space area, and a spatial hash coding table is used to perform water treatment optimization on the microalgae purification water quality analysis of the spatial hash decoding sequence.

5. A water treatment optimization method based on microalgae growth prediction according to claim 4, characterized in that: The method comprises the following steps: constructing a three-dimensional geographical terrain model of the target water treatment area, constraining and planning a predetermined pollution absorption area in the three-dimensional geographical terrain model based on purification performance, repeating the above steps to perform hash coding on the predetermined pollution absorption area, obtaining a spatial hash decoding sequence of the predetermined pollution absorption space area, and optimizing water treatment by analyzing the microalgae purification water quality of the spatial hash decoding sequence using a spatial hash coding table: Preset the purification radiation boundary threshold according to the purification performance and construct a three-dimensional geographical terrain model of the target water treatment area, plan the pollution absorption range within the three-dimensional geographical terrain model with the purification radiation boundary threshold as the range constraint, and obtain the established pollution absorption range of the target water treatment area by the future growth model of microalgae; Extract multiple pollution component concentration parameters of water quality in a predetermined pollution absorption action area through a real-time water quality pollution parameter set of a target water treatment area, repeat the above-mentioned spatial hash code construction step based on the multiple pollution component concentration parameters to perform hash coding on the regional spatial pollution degree of the predetermined pollution absorption action area, and obtain a spatial hash decoding sequence of the predetermined pollution absorption spatial area; Determine whether the spatial hash decoding sequence can decode and output the spatial hash coding table. If not, obtain the missing tampering code that the spatial hash coding table cannot decode and output using the spatial hash decoding sequence, and determine the purification performance gap of the microalgae growth for the polluted water quality in the target water treatment area according to the missing tampering code; Based on the purification performance gap, the growth conditions of microalgae in the given pollution absorption space area are improved and optimized to enhance the water pollution purification efficiency of microalgae growth in the given pollution absorption space area.

6. The water treatment optimization method based on microalgae growth prediction according to claim 1, characterized in that: The step S108 specifically includes the following steps: Constructing a random distribution framework of water environment flow in a three-dimensional geographic terrain model based on a real-time water environment simulation field of a target water treatment area, and extracting a random node array of the random distribution framework; Obtaining neighbor nodes of each random node in the random node array, calculating the weight between each random node and each neighbor node based on the real-time water quality pollution parameter set of the target water treatment area, generating weight values ​​of each random node corresponding to a number of random node boundaries, and determining the random probability distribution of real-time water quality pollution according to the ratio between the weight values; Preset a maximum number of steps, update the current random node to a neighbor node selected by the maximum random probability according to the random probability distribution, repeat the weight calculation between the random node and the neighbor node and the steps of random node update to continuously execute the next round of random walk until the random walk reaches the maximum number of steps, and generate a real-time water quality pollution scale model of the target water treatment area; Based on the big data network, a benchmark thermal color gamut rule for different water pollution scale ratings is obtained, and the real-time water pollution scale model is transformed by the benchmark thermal color gamut rule to obtain a real-time water pollution scale thermal model of the target water treatment area; Obtaining an online sensor layout pattern of a target water treatment area, dividing the real-time water pollution scale thermal model into M sub-model area blocks based on the online sensor layout pattern, and obtaining a real-time thermal color gamut RGB lookup table of each sub-model area block; According to the purification performance of microalgae for different water pollution scales, abnormal thermochromatic RGB values ​​of unreasonable water pollution scales are preset. If the real-time thermochromatic RGB query table can query the abnormal thermochromatic RGB values, the sub-model area block is marked as an abnormal water pollution scale area and the sewage treatment operation of microalgae growth regulation is performed.

7. A water treatment optimization system based on microalgae growth prediction, characterized in that: The water treatment optimization system includes a memory and a processor. The memory stores a water treatment optimization method program based on microalgae growth prediction. When the water treatment optimization method program is executed by the processor, the water treatment optimization method steps as described in any one of claims 1 to 6 are implemented.

Images