Water quality reduction method and related equipment based on artificial intelligence

Abstract

The present application proposes a water quality emission reduction method, device, electronic device and storage medium based on artificial intelligence. The water quality emission reduction method based on artificial intelligence includes: collecting river hydrological data within a preset time period from a water quality monitoring database, the river hydrological data includes main stream hydrological data and tributary hydrological data; training a first regression model based on river hydrological data; constructing an objective function based on preset constraints and the first regression model; solving the objective function using the gradient descent method to obtain candidate emission reduction schemes; calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction schemes, and inputting the difference into the first regression model to obtain emission reduction results; recommending emission reduction schemes based on emission reduction results. The method constructs an objective function through historical hydrological data, and optimizes the objective function to obtain a calibrated emission reduction scheme, thereby improving emission reduction efficiency.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based water quality reduction method, device, electronic device and storage medium. Background Art

[0002] In recent years, environmental pollution problems have become increasingly prominent and have received widespread attention from the society. Therefore, people's requirements for the quality of water environment have become higher and higher. Scientifically managing the water environment of river basins and protecting the water ecosystem has become an important task.

[0003] Currently, river water quality is usually protected by reducing the discharge of river pollution factors. However, at this stage, the formulation of emission reduction plans usually requires a lot of manpower and material resources to repeatedly measure hydrological data to determine a more appropriate emission reduction plan, which results in low energy efficiency of the emission reduction plan. Summary of the invention

[0004] In view of the above content, it is necessary to provide an artificial intelligence-based water quality emission reduction method and related equipment to solve the technical problem of how to improve the emission reduction performance, wherein the related equipment includes an artificial intelligence-based emission reduction device, electronic equipment and storage medium.

[0005] The present application provides a water quality reduction method based on artificial intelligence, the method comprising:

[0006] Collect river hydrological data from a water quality monitoring database, wherein the river hydrological data includes mainstream hydrological data and tributary hydrological data;

[0007] Training a first regression model based on the river hydrological data, wherein the first regression model is used to characterize the corresponding relationship between the main stream hydrological data and the tributary hydrological data;

[0008] Constructing an objective function according to preset constraints and the first regression model;

[0009] Solving the objective function using a gradient descent method to obtain candidate emission reduction schemes, wherein the candidate emission reduction schemes are used to represent emission reduction schemes that meet constraint conditions;

[0010] The difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme is calculated, and the difference is input into the first regression model to obtain the emission reduction result.

[0011] The above-mentioned artificial intelligence-based water quality emission reduction method obtains the hydrological data of the river mainstream and each tributary of the river by collecting river hydrological data from the water quality monitoring database, which provides data support for subsequent model fitting and emission reduction scheme selection, and trains a first regression model based on the river hydrological data to characterize the correspondence between the mainstream hydrological data and the tributary hydrological data. The objective function is further constructed based on the preset conditions and the first regression model, and the objective function is solved using the gradient descent method to obtain candidate emission reduction schemes. Finally, the candidate emission reduction schemes are evaluated using preset environmental protection standards to obtain a calibrated emission reduction scheme, so that the final emission reduction scheme can be obtained based on the analysis of historical data, saving manpower and material resources and improving the efficiency of emission reduction scheme selection, thereby improving the energy efficiency of water quality emission reduction.

[0012] In some embodiments, the first regression model is trained based on the river hydrological data, and the first regression model is used to characterize the correspondence between the main stream hydrological data and the tributary hydrological data, including:

[0013] The tributary hydrological data is used as an independent variable, and the mainstream hydrological data is used as a dependent variable to construct a linear function;

[0014] Fitting the parameters of the linear function using the least squares method to obtain a parameter set;

[0015] Each parameter in the parameter set is matched with the tributary hydrological data to construct a first regression model.

[0016] In this way, a linear function relationship is established between the hydrological data of each tributary and the corresponding main stream hydrological data, and the parameters in the linear function are calculated according to the least squares method, thereby constructing a first regression model. The first regression model can characterize the degree of influence of the tributary hydrological data on the main stream hydrological data, thereby providing data support for planning emission reduction plans and improving emission reduction efficiency.

[0017] In some embodiments, the objective function is constructed based on the river emission reduction and the first regression model, wherein the river emission reduction refers to the difference between the river hydrological data before and after the implementation of the emission reduction plan, including:

[0018] The hydrological data of the tributaries before the implementation of the emission reduction plan is taken as the front quantity, and the hydrological data after the implementation of the emission reduction plan is taken as the back quantity, and the difference between the front quantity and the back quantity is calculated as the emission reduction amount of each tributary;

[0019] Input the front amount into the first regression model to obtain the front amount of the main stream, input the rear amount into the first regression model to obtain the rear amount of the main stream, and calculate the difference between the front amount of the main stream and the rear amount of the main stream as the main stream emission reduction amount;

[0020] An objective function is constructed based on the emission reduction of each tributary and the emission reduction of the main stream, and the objective function is used to constrain the cost-effectiveness of the emission reduction plan.

[0021] In this way, the energy efficiency of the emission reduction plan is defined by analyzing the preset constraints, and the energy efficiency formula is defined by calculating the emission reduction of tributaries and the main stream. The energy efficiency formula is further used as the objective function, which can transform the selection of emission reduction plans into an analysis of quantitative relationships, thereby improving the accuracy of the formulation of emission reduction plans.

[0022] In some embodiments, before using the gradient descent method to solve the objective function to obtain the candidate emission reduction scheme, the method further includes:

[0023] Select the maximum value of the hydrological data of each tributary of the river and the hydrological data of the main stream as the benchmark value;

[0024] Calculate the ratio of each hydrological data to the corresponding benchmark value as the standard value;

[0025] Each hydrological data was replaced with the corresponding standard value to obtain the standardized hydrological data.

[0026] In this way, by comparing all hydrological data with the corresponding benchmark values, the dimensional gap between the hydrological data is eliminated, thereby reducing the amount of calculation in the subsequent gradient descent method to solve the objective function, thereby improving the efficiency of the gradient descent method.

[0027] In some embodiments, the objective function is solved by using a gradient descent method to obtain a candidate emission reduction scheme, wherein the candidate emission reduction scheme is used to represent an emission reduction scheme that satisfies a constraint condition and includes:

[0028] a, randomly select an initial position and use it as the current position;

[0029] b, calculating the gradient of the current position, where the current position corresponds to a preset first emission reduction scheme;

[0030] c. Calculate the product of the preset step length data and the gradient as the descent distance of the current position;

[0031] d. If the descent distance does not meet the preset first judgment condition, the first emission reduction plan is updated according to the descent distance to obtain the second emission reduction plan, the position corresponding to the second emission reduction plan is used as the current position and returns to step b; otherwise, the emission reduction plan corresponding to the current position is used as a candidate emission reduction plan.

[0032] In this way, the initial emission reduction plan was obtained by randomly selecting points and calculating gradients, and the candidate emission reduction plans were obtained by iteratively calculating the objective function according to the gradient descent method, which provided data support for the selection of subsequent emission reduction plans, thereby improving the energy efficiency specified by the emission reduction plan.

[0033] In some embodiments, the calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain the real-time emission reduction result comprises:

[0034] At the same time point, the hydrological data of each tributary of the river is collected as the first real-time data, and at the same time point, the hydrological data of the main stream of the river is collected as the second real-time data;

[0035] Calculating a difference between the first real-time data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain a fitting value;

[0036] The difference between the fitting value and the second real-time data is calculated as the real-time emission reduction result.

[0037] In this way, the emission reduction results were obtained by collecting the river's hydrological data in real time and simulating the changes in the river's hydrological data after the implementation of the candidate emission reduction schemes, avoiding the waste of manpower and material resources caused by pre-testing the emission reduction schemes and improving the energy efficiency of water quality emission reduction.

[0038] In some embodiments, evaluating the candidate emission reduction schemes according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and recommending the calibrated emission reduction scheme includes:

[0039] Comparing the real-time emission reduction result with the preset environmental protection standard to obtain a comparison result, wherein the comparison result includes "qualified" and "unqualified";

[0040] If the comparison result is "qualified", the candidate emission reduction scheme is used as a calibrated emission reduction scheme and the calibrated emission reduction scheme is recommended;

[0041] If the comparison result is "unqualified", the candidate emission reduction scheme is marked as "unqualified" and the objective function is solved again using the gradient descent method to obtain the next candidate emission reduction scheme.

[0042] In this way, the emission reduction results are evaluated through preset environmental protection standards to obtain the evaluation results of the emission reduction plan, and the treatment method of the emission reduction plan is selected according to the evaluation results. It can make an accurate theoretical evaluation of feasible emission reduction plans and avoid the waste of resources caused by field experiments, thereby improving the energy efficiency of water quality emission reduction.

[0043] The present application also provides a water quality reduction device based on artificial intelligence, the device comprising:

[0044] A collection unit, used to collect river hydrological data within a preset time period from a water quality monitoring database, wherein the river hydrological data includes mainstream hydrological data and tributary hydrological data;

[0045] A training unit, used for training a first regression model based on the river hydrological data, wherein the first regression model is used to characterize the correspondence between the main stream hydrological data and the tributary hydrological data;

[0046] A construction unit, used to construct an objective function according to preset constraints and the first regression model;

[0047] An optimization unit, used for solving the objective function by using a gradient descent method to obtain a candidate emission reduction scheme, wherein the candidate emission reduction scheme is used to represent an emission reduction scheme that satisfies a constraint condition;

[0048] A calculation unit, used for calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain a real-time emission reduction result;

[0049] A recommendation unit is used to evaluate the candidate emission reduction schemes according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and recommend the calibrated emission reduction scheme.

[0050] The present application also provides an electronic device, the device comprising:

[0051] a memory storing computer readable instructions; and

[0052] A processor executes computer-readable instructions stored in the memory to implement the artificial intelligence-based water quality reduction method.

[0053] An embodiment of the present application also provides a computer-readable storage medium, in which computer-readable instructions are stored. The computer-readable instructions are executed by a processor in an electronic device to implement the artificial intelligence-based water quality reduction method.

[0054] The above-mentioned artificial intelligence-based water quality emission reduction method obtains the hydrological data of the river mainstream and each tributary of the river by collecting river hydrological data from the water quality monitoring database, which provides data support for subsequent model fitting and emission reduction scheme selection, and trains a first regression model based on the river hydrological data to characterize the correspondence between the mainstream hydrological data and the tributary hydrological data. The objective function is further constructed based on the preset conditions and the first regression model, and the objective function is solved using the gradient descent method to obtain candidate emission reduction schemes. Finally, the candidate emission reduction schemes are evaluated using preset environmental protection standards to obtain a calibrated emission reduction scheme, so that the final emission reduction scheme can be obtained based on the analysis of historical data, saving manpower and material resources and improving the efficiency of emission reduction scheme selection, thereby improving the energy efficiency of water quality emission reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1It is a flow chart of a preferred embodiment of a water quality reduction method based on artificial intelligence involved in this application.

[0056] Figure 2 It is a functional module diagram of a preferred embodiment of the water quality reduction device based on artificial intelligence involved in this application.

[0057] Figure 3 It is a structural schematic diagram of an electronic device of a preferred embodiment of the water quality reduction method based on artificial intelligence involved in this application.

[0058] Figure 4 It is a structural diagram of the hydrological data involved in this application.

[0059] Figure 5 This is a schematic diagram of the river hydrological observation involved in this application. DETAILED DESCRIPTION

[0060] In order to more clearly understand the purpose, features and advantages of the present application, the present application is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other. In the following description, many specific details are set forth to facilitate a full understanding of the present application, and the embodiments described are only a part of the embodiments of the present application, rather than all of the embodiments.

[0061] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used herein in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0063] The embodiment of the present application provides a water quality reduction method based on artificial intelligence, which can be applied to one or more electronic devices. The electronic device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0064] The electronic device may be any electronic product that can perform human-computer interaction with a user, such as a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), a game console, an interactive network television (IPTV), a smart wearable device, etc.

[0065] The electronic device may also include a network device and / or a user device, wherein the network device includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud consisting of a large number of hosts or network servers based on cloud computing.

[0066] The network where the electronic device is located includes but is not limited to the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.

[0067] like Figure 1 As shown, it is a flow chart of a preferred embodiment of the water quality reduction method based on artificial intelligence of the present application. According to different requirements, the order of the steps in the flow chart can be changed, and some steps can be omitted.

[0068] S10, collecting river hydrological data from a water quality monitoring database, wherein the river hydrological data includes mainstream hydrological data and tributary hydrological data.

[0069] In this optional embodiment, the water quality monitoring database can be a data set designed, stored and managed according to the data structure. It is an important part of the entire hydrological information processing system and the application of modern database management technology in the field of hydrology. Exemplarily, the water quality monitoring database described in this scheme can be a mariaDB database. The mariaDB database is an open source database whose function is to store historical and real-time hydrological data collected at hydrological stations, and provide services to a wide range of users quickly and accurately through a variety of output methods and processing technologies.

[0070] In this optional embodiment, the hydrological data mainly include surface water, groundwater, water quality data and river and lake topography and other related attribute data, including original monitoring data, integrated compilation results data and statistical analysis results and application support data.

[0071] Exemplarily, the hydrological data described in this scheme include COD concentration, ammonia nitrogen concentration and total phosphorus concentration. The full name of the COD is Chemical Oxygen Demand, which means chemical oxygen demand, an important comprehensive indicator for organic matter and reducing substances such as nitrite, ferrous salt and sulfide in water, and its unit is mg / L. The higher the COD concentration, the more serious the water pollution; the ammonia nitrogen refers to combined nitrogen in the form of free ammonia and ammonium ions. The ammonia nitrogen is a nutrient in the water and can cause eutrophication of water. The unit of the ammonia nitrogen concentration is mg / L. The higher the ammonia nitrogen concentration, the more serious the water pollution; the total phosphorus refers to the phosphorus element in the water in the form of elemental phosphorus, orthophosphate, condensed phosphate, pyrophosphate, metaphosphate and phosphate combined with organic groups. Phosphorus in water is a key element required for algae growth, and excessive phosphorus is the main cause of water pollution and odor. The unit of the total phosphorus concentration is mg / L. The higher the total phosphorus concentration, the more serious the water pollution.

[0072] In this optional embodiment, the main stream hydrological data refers to various data related to water quality in the main channel of the river, and the tributary hydrological data refers to various data related to water quality in the tributaries of the river.

[0073] For example, Figure 4 Shown is a schematic diagram of the structure of the hydrological data described in this scheme.

[0074] For example, Figure 5 Shown is a schematic diagram of river hydrological observation involved in this plan.

[0075] In this way, by collecting data in the water quality monitoring database, the hydrological data of the main river and the river tributaries are obtained to characterize the changes in river water quality over the past period of time, which can provide data support for subsequent river water quality analysis, thereby improving the energy efficiency of emission reduction plans.

[0076] S11, training a first regression model based on the river hydrological data, wherein the first regression model is used to characterize the correspondence between the main stream hydrological data and the tributary hydrological data.

[0077] In an optional embodiment, the first regression model is trained based on the river hydrological data, and the first regression model is used to characterize the corresponding relationship between the main stream hydrological data and the tributary hydrological data, including:

[0078] The tributary hydrological data is used as an independent variable, and the mainstream hydrological data is used as a dependent variable to construct a linear function;

[0079] Fitting the parameters of the linear function using the least squares method to obtain a parameter set;

[0080] Each parameter in the parameter set is matched with the tributary hydrological data to construct a first regression model.

[0081] In this optional embodiment, the river may include n tributaries and n is a positive integer. For example, in this solution, n may be equal to 3. The hydrological data of each tributary may be used as an independent variable. For example, the hydrological data of the first tributary may be recorded as Where x1 is a vector, the first dimension of the vector x1 is the COD concentration value in the tributary, the second dimension is the ammonia nitrogen concentration value in the tributary, and the third dimension is the total phosphorus concentration value in the tributary. Similarly, the hydrological data of the second tributary can be recorded as The hydrological data of the third tributary can be recorded as The hydrological data of the main stream can be recorded as X = [X COD , X AD , X ZL ].

[0082] In this optional embodiment, the hydrological data of the first tributary represented by x1, the hydrological data of the second tributary represented by x2, and the hydrological data of the third tributary represented by x3 can be used as independent variables, and the hydrological data of the main stream represented by X can be used as a dependent variable. A linear function is constructed based on the independent variables and the dependent variables, and the linear function satisfies the following relationship:

[0083]

[0084] Among them, a, b, c and d represent the parameters of the linear function, a is used to characterize the degree of influence of the hydrological data of the first tributary on the hydrological data of the mainstream, b is used to characterize the degree of influence of the hydrological data of the second tributary on the hydrological data of the mainstream, c is used to characterize the degree of influence of the hydrological data of the third tributary on the hydrological data of the mainstream, and d is used to characterize the possible deviation of the linear function.

[0085] In this optional embodiment, since the hydrological data includes three different data, the parameters a, b, c and d can be recorded as:

[0086]

[0087] For ease of understanding, the linear function relationship can be expressed as:

[0088] X=a·x1+b·x2+c·x3+d

[0089] Among them, X represents the hydrological data of the main stream of the river simulated and calculated by the linear function, x1 represents the hydrological data of the first tributary, x2 represents the hydrological data of the second tributary, x3 represents the hydrological data of the third tributary, and a, b, c and d represent the parameters of the linear function.

[0090] In this optional embodiment, the least squares method can be used to fit the parameters of the linear function. The least squares method is implemented by inputting each set of independent variables and the corresponding dependent variables into the linear function, and calculating the square of the difference between the contents on both sides of the linear function equal sign. Exemplarily, the square of the difference D is calculated as follows:

[0091] D=(a·x1+b·x2+c·x3+dX) 2

[0092] In this optional embodiment, a corresponding difference value can be obtained for each set of independent variables and the corresponding dependent variable, and the square of each difference value is calculated and summed to obtain a second linear function, which satisfies the following relationship:

[0093] S(a, b, c, d)=D1+D2+…+D n

[0094] Among them, S(a, b, c, d) represents the second linear function, which is used to characterize that the variables in the second linear function are a, b, c, d; D1 represents the square of the difference obtained after inputting the first set of independent variables and the corresponding dependent variables into the linear function, and its subscript corresponds to the index of the number of independent variable and dependent variable groups.

[0095] In this optional embodiment, partial derivatives of the four variables in the second linear function may be calculated respectively to obtain four partial derivative formulas, which are respectively recorded as:

[0096] In this optional embodiment, the four partial derivative formulas can be made equal to 0 respectively and each partial derivative formula can be solved to obtain the values ​​of the four parameters a, b, c, and d, where each parameter corresponds to a value.

[0097] In this optional embodiment, a, b, c, and d may be used as a parameter set.

[0098] In this optional embodiment, each parameter in the parameter set may be matched with the hydrological data of its corresponding tributary according to the format in the linear function to construct a first regression model, and the first regression model satisfies the relationship:

[0099] f(x1,x2,x3)=Y=a·x1+b·x2+c·x3+d

[0100] Among them, f(x1, x2, x3) represents that the first regression model is a function with the tributary hydrological data as the unknown variable; Y represents the predicted value of the main stream hydrological data obtained by inputting the known tributary hydrological data into the first regression model; a, b, c and d represent the parameters of the first regression model calculated by the least squares method; x1 represents the hydrological data of the first tributary; x2 represents the hydrological data of the second tributary; x3 represents the hydrological data of the third tributary.

[0101] In this way, a linear function relationship is established between the hydrological data of each tributary and the corresponding main stream hydrological data, and the parameters in the linear function are calculated according to the least squares method, thereby constructing a first regression model. The first regression model can characterize the degree of influence of the tributary hydrological data on the main stream hydrological data, thereby providing data support for planning emission reduction plans and improving emission reduction efficiency.

[0102] S12, constructing an objective function based on the river emission reduction and the first regression model, wherein the river emission reduction refers to the difference in river hydrological data before and after the implementation of the emission reduction plan.

[0103] In an optional embodiment, the objective function is constructed based on the river emission reduction and the first regression model, and the river emission reduction refers to the difference between the river hydrological data before and after the implementation of the emission reduction plan, including:

[0104] The hydrological data of the tributaries before the implementation of the emission reduction plan is taken as the front quantity, and the hydrological data after the implementation of the emission reduction plan is taken as the back quantity, and the difference between the front quantity and the back quantity is calculated as the emission reduction amount of each tributary;

[0105] Input the front amount into the first regression model to obtain the front amount of the main stream, input the rear amount into the first regression model to obtain the rear amount of the main stream, and calculate the difference between the front amount of the main stream and the rear amount of the main stream as the main stream emission reduction amount;

[0106] An objective function is constructed based on the emission reduction of each tributary and the emission reduction of the main stream, and the objective function is used to constrain the cost-effectiveness of the emission reduction plan.

[0107] In this optional embodiment, the emission reduction scheme refers to reducing the concentration of a certain number of water pollution factors in each tributary of the river, with the aim of reducing water pollution in the main stream of the river as much as possible. Exemplarily, the water pollution factors in this scheme are the COD concentration, ammonia nitrogen concentration and total phosphorus concentration in the hydrological data.

[0108] In this optional embodiment, the tributary hydrological data before the emission reduction plan is implemented can be used as the pre-quantity and recorded as and The tributary hydrological data after the emission reduction plan is implemented can be used as the post-quantity and recorded as and

[0109] In this optional embodiment, the difference in hydrological data of each tributary before and after the implementation of the emission reduction scheme can be calculated as the emission reduction amount of each tributary. For example, in this scheme, the emission reduction amount of each tributary can be recorded as Δx1, Δx2 and Δx3, where the subscript represents the index of each tributary.

[0110] In this optional embodiment, the calculation method of the tributary emission reduction is:

[0111]

[0112] Among them, the superscript of variable x represents before or after the implementation of the emission reduction plan, and the subscript of variable x represents the index of the river tributary.

[0113] In this optional embodiment, the pre-quantity can be input into the first regression model to obtain the hydrological data of the mainstream before the emission reduction plan is implemented as the pre-quantity of the mainstream and recorded as X 前 , the X 前 The calculation method is:

[0114]

[0115] in, Represents the hydrological data of the first tributary before the implementation of the emission reduction plan, Represents the hydrological data of the second tributary before the emission reduction plan was implemented. represents the hydrological data of the third tributary before the implementation of the emission reduction plan, and a, b, c and d represent the parameters in the first regression model.

[0116] In this optional embodiment, the post-volume can be input into the first regression model to obtain the hydrological data of the main stream after the emission reduction plan is implemented as the post-volume of the main stream and recorded as X 后 , the X 后 The calculation method is:

[0117]

[0118] in, Represents the hydrological data of the first tributary after the emission reduction plan is implemented, Represents the hydrological data of the second tributary after the emission reduction plan was implemented. represents the hydrological data of the third tributary after the implementation of the emission reduction plan, and a, b, c and d represent the parameters in the first regression model.

[0119] In this optional embodiment, the difference between the amount before the main flow and the amount after the main flow can be calculated as the main flow emission reduction and recorded as ΔX. The calculation method of ΔX is:

[0120] ΔX=X 前 -X 后

[0121] Among them, X 前 represents the hydrological data of the mainstream river before the implementation of the emission reduction plan, X 后 Represents the hydrological data of the mainstream river after the emission reduction plan is implemented.

[0122] In summary, in this optional embodiment, the main stream emission reduction ΔX can be characterized as:

[0123]

[0124] In this optional embodiment, an objective function may be constructed based on the emission reduction of each tributary and the emission reduction of the main stream, and the objective function is used to constrain the cost-effectiveness of the emission reduction scheme, and the objective function satisfies the following relationship:

[0125]

[0126] Among them, G(Δx1, Δx2, Δx3) represents that the objective function is a function related to the emission reduction of tributaries, Δx1 represents the emission reduction of the first tributary, Δx2 represents the emission reduction of the second tributary, Δx3 represents the emission reduction of the third tributary, and a, b, c and d represent the parameters of the first regression model.

[0127] Exemplarily, when a=1, b=1, c=1, Δx1=10, Δx2=9, Δx3=8, the value of the objective function is:

[0128]

[0129] Then the value of the objective function is 1.

[0130] In this optional embodiment, when the objective function takes the minimum value, it indicates that the emission reduction of the tributary is smaller while the emission reduction of the main stream is larger, that is, a better emission reduction effect is obtained at a smaller cost, and the emission reduction scheme corresponding to the minimum value of the objective function has a higher cost-effectiveness.

[0131] In this way, the energy efficiency of the emission reduction plan is defined by analyzing the preset constraints, and the energy efficiency formula is defined by calculating the emission reduction of tributaries and the main stream. The energy efficiency formula is further used as the objective function, which can transform the selection of emission reduction plans into an analysis of quantitative relationships, thereby improving the accuracy of the formulation of emission reduction plans.

[0132] S13, using a gradient descent method to solve the objective function to obtain candidate emission reduction schemes, where the candidate emission reduction schemes are used to represent emission reduction schemes that meet constraint conditions.

[0133] In an optional embodiment, before using the gradient descent method to solve the objective function to obtain the candidate emission reduction scheme, the method further includes:

[0134] Select the maximum value of the hydrological data of each tributary of the river and the hydrological data of the main stream as the benchmark value;

[0135] Calculate the ratio of each hydrological data to the corresponding benchmark value as the standard value;

[0136] Each hydrological data was replaced with the corresponding standard value to obtain the standardized hydrological data.

[0137] In this optional embodiment, the maximum value of the hydrological data of each tributary of the river and the hydrological data of the main stream can be selected as the reference value. For example, the reference value of the first tributary can be recorded as The reference value of the second tributary can be The reference value of the third tributary is The reference value of the main flow is

[0138] In this optional embodiment, the ratio of each hydrological data to the corresponding reference value can be calculated as the standard value. For example, the hydrological data in the first tributary can be taken as an example. When a hydrological data in the first tributary is And the base value of the first tributary is When , the calculation method of the standard value corresponding to the hydrological data is:

[0139]

[0140] The standard value corresponding to the hydrological data is

[0141] In this optional embodiment, each hydrological data can be replaced by a corresponding standard value to obtain standardized hydrological data. The purpose is to uniformly convert all hydrological data to between (0, 1) to eliminate the dimensional difference between the hydrological data, thereby reducing the amount of calculation in the subsequent gradient descent method to solve the objective function, thereby improving the efficiency of the gradient descent method.

[0142] In an optional embodiment, the objective function is solved by using a gradient descent method to obtain a candidate emission reduction scheme, wherein the candidate emission reduction scheme is used to represent an emission reduction scheme that satisfies preset constraints and has a good emission reduction effect, and includes:

[0143] a, randomly select an initial position and use it as the current position;

[0144] b, calculating a gradient of a current position, wherein the current position corresponds to a preset first emission reduction scheme;

[0145] c. Calculate the product of the preset step length data and the gradient as the descent distance of the current position;

[0146] d. If the descent distance does not meet the preset first judgment condition, the first emission reduction plan is updated according to the descent distance to obtain a second emission reduction plan, and the position corresponding to the second emission reduction plan is used as the current position and returns to step b.

[0147] In this optional embodiment, the specific implementation steps of using the gradient descent method to solve the objective function to obtain the candidate emission reduction scheme are:

[0148] A1: Randomly select an initial position, the initial position corresponds to a preset emission reduction scheme, which can be used as the first emission reduction scheme. For example, the initial position can be recorded as and The initial position may be used as the current position;

[0149] A2: Calculate the gradient of the current position. The calculation method of the gradient of the current position is:

[0150]

[0151]

[0152]

[0153] in, and represents the gradient at the current location, the superscript of the variable w represents the index of the emission reduction scheme, and the subscript of the variable w represents the index of the river tributary; represents the symbol of partial derivative, for example, Represents the objective function with respect to the variable The partial derivative of and When unchanged, The degree of influence of the change of on the change of the objective function value;

[0154] A3: In this optional embodiment, the product of the preset step length data and the gradient can be calculated as the descent distance D of the current position. In this solution, the preset step length data can be recorded as τ. Exemplarily, the step length data τ can be 0.01. The calculation method of the descent distance D is:

[0155]

[0156] in, represents the emission reduction of the jth tributary, Dj represents the distance that the emission reduction of the jth tributary decreases, j∈{1,2,3};

[0157] A4: If the descent distance does not meet the preset first discrimination condition, the first emission reduction scheme is updated according to the descent distance to obtain the second emission reduction scheme, the position corresponding to the second emission reduction scheme is used as the current position and the process returns to step A2; otherwise, the emission reduction scheme corresponding to the current position is used as a candidate emission reduction scheme, wherein the preset first discrimination condition is that the descent distance is less than a preset first threshold. Exemplarily, the value of the first threshold may be 0.1;

[0158] Exemplarily, the calculation method of the second emission reduction scheme is:

[0159]

[0160] in, represents the emission reduction of the jth tributary in the second emission reduction scheme, The superscript 1 in represents that the emission reduction scheme is the second emission reduction scheme. represents the emission reduction of the jth tributary in the first emission reduction scheme, The superscript 0 indicates that the emission reduction scheme is the first emission reduction scheme. j Represents the decrease distance of the emission reduction of the jth tributary, j∈{1,2,3}.

[0161] In this optional embodiment, a candidate emission reduction scheme can be obtained based on the above steps. The candidate emission reduction scheme is a set of data containing multiple numerical values, wherein each numerical value is used to characterize the emission reduction amount of each tributary hydrological data. Exemplarily, the candidate emission reduction schemes in this scheme can be recorded as Δx1, Δx2 and Δx3.

[0162] In this way, the initial emission reduction plan was obtained by randomly selecting points and calculating gradients, and the candidate emission reduction plans were obtained by iteratively calculating the objective function according to the gradient descent method, which provided data support for the selection of subsequent emission reduction plans, thereby improving the energy efficiency specified by the emission reduction plan.

[0163] S14, calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain a real-time emission reduction result.

[0164] In an optional embodiment, the calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain the real-time emission reduction result comprises:

[0165] At the same time point, the hydrological data of each tributary of the river is collected as the first real-time data, and at the same time point, the hydrological data of the main stream of the river is collected as the second real-time data;

[0166] Calculating a difference between the first real-time data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain a fitting value;

[0167] The difference between the fitting value and the second real-time data is calculated as the real-time emission reduction result.

[0168] In this optional embodiment, the hydrological data of each tributary of the river can be collected at the same time point as the first real-time data. For example, the time point can be recorded as t, and the first real-time data can be recorded as and They are used to characterize the hydrological data of each tributary collected in real time.

[0169] In this optional embodiment, the hydrological data of the main stream of the river can be collected in real time at time t as the second real-time data. For example, the second real-time data can be recorded as Used to characterize the hydrological data of the river mainstream collected at time point t.

[0170] In this optional embodiment, the difference between the first real-time data and the candidate emission reduction scheme may be calculated. For example, the first real-time data may be and The candidate emission reduction scheme may be and The difference between the first real-time data and the candidate emission reduction scheme can be recorded as T1, T2 and T3 respectively, and the calculation method of the difference between the first real-time data and the candidate emission reduction scheme is:

[0171]

[0172]

[0173]

[0174] In this optional embodiment, the difference is used to characterize the value of the hydrological data of each tributary of the river after the candidate emission reduction scheme is implemented. The difference can be input into the first regression model to obtain a fitting value, and the fitting value is used to characterize the value of the hydrological data of the main stream of the river after the candidate emission reduction scheme is implemented. For example, the value of the hydrological data of the main stream of the river can be recorded as T0, and the calculation method of T0 is:

[0175] T0=a·T1+b·T2+c·T3+d

[0176] Among them, T0 represents the fitting value obtained by inputting the difference into the first regression model, which is used to characterize the value of the hydrological data of the main stream of the river after the implementation of the candidate emission reduction plan; T1, T2 and T3 respectively represent the hydrological data of each tributary of the river after the candidate emission reduction plan is implemented; a, b, c and d represent the parameters of the first regression model.

[0177] For example, when the and When T0 is calculated as:

[0178]

[0179] Then the value of the mainstream hydrological data after the implementation of the candidate emission reduction scheme represented by T0 is This value is used to represent the COD concentration of 11 mg / L, the ammonia nitrogen concentration of 7 mg / L, and the total phosphorus concentration of 1 mg / L in the mainstream of the river after the implementation of the candidate emission reduction plan.

[0180] In this optional embodiment, the difference between the value of T0 and the second real-time data can be calculated as the emission reduction result, and the emission reduction result is used to characterize the amount of decrease in the hydrological data of the mainstream of the river after the implementation of the candidate emission reduction scheme. The real-time emission reduction result can be recorded as R. For example, when the value of T0 is And the second real-time data is When , the real-time emission reduction result R is calculated as follows:

[0181]

[0182] Then the value of the real-time emission reduction result R is This value is used to represent that after the implementation of the candidate emission reduction plan, the COD concentration in the mainstream of the river decreased by 4 mg / L, the ammonia nitrogen concentration decreased by 3 mg / L, and the total phosphorus concentration decreased by 2 mg / L.

[0183] In this way, the emission reduction results of the mainstream hydrological data after the implementation of the emission reduction plan are calculated through the real-time collected hydrological data, which can simulate the emission reduction effect and provide data support for the evaluation of the emission reduction plan, thereby improving the performance of the emission reduction plan.

[0184] S15, evaluating the candidate emission reduction schemes according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and recommending the calibrated emission reduction scheme.

[0185] In an optional embodiment, evaluating the candidate emission reduction schemes according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and recommending the calibrated emission reduction scheme includes:

[0186] Comparing the real-time emission reduction result with the preset environmental protection standard to obtain a comparison result, wherein the comparison result includes "qualified" and "unqualified";

[0187] If the comparison result is "qualified", the candidate emission reduction scheme is used as a calibrated emission reduction scheme and the calibrated emission reduction scheme is recommended;

[0188] If the comparison result is "unqualified", the candidate emission reduction scheme is marked as "unqualified" and the objective function is solved again using the gradient descent method to obtain the next candidate emission reduction scheme.

[0189] In this optional embodiment, the real-time emission reduction result can be compared with the preset environmental protection standard to obtain a comparison result, and the real-time emission reduction result includes multiple components, which respectively represent the numerical value of the COD concentration reduction in the main stream of the river, the numerical value of the ammonia nitrogen concentration reduction, and the numerical value of the total phosphorus concentration reduction. The preset environmental protection standard includes multiple components, which respectively represent the numerical value of the COD concentration reduction in the main stream of the river expected by the environmental protection department, the numerical value of the ammonia nitrogen concentration reduction, and the numerical value of the total phosphorus concentration reduction. If the absolute value of each component in the real-time emission reduction result is greater than the absolute value of the corresponding component in the preset environmental protection standard, the comparison result is "qualified", the candidate emission reduction plan is used as the calibrated emission reduction plan, and the calibrated emission reduction plan is recommended to the environmental protection responsible unit to protect the river water quality.

[0190] Exemplarily, when the real-time emission reduction result is And the preset environmental protection standards are When the absolute value of the real-time emission reduction result is And the absolute value of the preset environmental protection standard is If the absolute value of each component in the real-time emission reduction result is greater than the absolute value of each component in the environmental protection standard, the candidate emission reduction plan corresponding to the real-time emission reduction result can be used as a calibrated emission reduction plan and recommended to the environmental protection unit to protect the river water quality.

[0191] In this optional embodiment, if the absolute value of the emission reduction result is not greater than the absolute value of the preset environmental protection standard, the comparison result is "unqualified", and the candidate emission reduction scheme can be marked as "unqualified", and the objective function is solved again using the gradient descent method to obtain the next candidate emission reduction scheme. The real-time emission reduction result is recalculated based on the next candidate emission reduction scheme to re-evaluate the next candidate emission reduction scheme.

[0192] The above-mentioned artificial intelligence-based water quality emission reduction method obtains the hydrological data of the river mainstream and each tributary of the river by collecting river hydrological data from the water quality monitoring database, which provides data support for subsequent model fitting and emission reduction scheme selection, and trains a first regression model based on the river hydrological data to characterize the correspondence between the mainstream hydrological data and the tributary hydrological data. The objective function is further constructed based on the preset conditions and the first regression model, and the objective function is solved using the gradient descent method to obtain candidate emission reduction schemes. Finally, the candidate emission reduction schemes are evaluated using preset environmental protection standards to obtain a calibrated emission reduction scheme, so that the final emission reduction scheme can be obtained based on the analysis of historical data, saving manpower and material resources and improving the efficiency of emission reduction scheme selection, thereby improving the energy efficiency of water quality emission reduction.

[0193] like Figure 2 , which is a functional module diagram of a preferred embodiment of an emission reduction device based on artificial intelligence provided in an embodiment of the present application. The emission reduction device based on artificial intelligence 11 includes a collection unit 110, a training unit 111, a construction unit 112, an optimization unit 113, a calculation unit 114, and a recommendation unit 115. The module / unit referred to in this application refers to a series of computer program segments that can be executed by a processor 13 and can perform fixed functions, which are stored in a memory 12. In this embodiment, the functions of each module / unit will be described in detail in subsequent embodiments.

[0194] In an optional embodiment, the collection unit 110 is used to collect river hydrological data from a water quality monitoring database, where the river hydrological data includes mainstream hydrological data and tributary hydrological data.

[0195] In this optional embodiment, the water quality monitoring database can be a data set designed, stored and managed according to the data structure. It is an important part of the entire hydrological information processing system and the application of modern database management technology in the field of hydrology. Exemplarily, the water quality monitoring database described in this scheme can be a mariaDB database. The mariaDB database is an open source database whose function is to store historical and real-time hydrological data collected at hydrological stations, and provide services to a wide range of users quickly and accurately through a variety of output methods and processing technologies.

[0196] In this optional embodiment, the hydrological data mainly include surface water, groundwater, water quality data and river and lake topography and other related attribute data, including original monitoring data, integrated compilation results data and statistical analysis results and application support data.

[0197] Exemplarily, the hydrological data described in this scheme include COD concentration, ammonia nitrogen concentration and total phosphorus concentration. The full name of the COD is Chemical Oxygen Demand, which means chemical oxygen demand, an important comprehensive indicator for organic matter and reducing substances such as nitrite, ferrous salt and sulfide in water, and its unit is mg / L. The higher the COD concentration, the more serious the water pollution; the ammonia nitrogen refers to combined nitrogen in the form of free ammonia and ammonium ions. The ammonia nitrogen is a nutrient in the water and can cause eutrophication of water. The unit of the ammonia nitrogen concentration is mg / L. The higher the ammonia nitrogen concentration, the more serious the water pollution; the total phosphorus refers to the phosphorus element in the water in the form of elemental phosphorus, orthophosphate, condensed phosphate, pyrophosphate, metaphosphate and phosphate combined with organic groups. Phosphorus in water is a key element required for algae growth, and excessive phosphorus is the main cause of water pollution and odor. The unit of the total phosphorus concentration is mg / L. The higher the total phosphorus concentration, the more serious the water pollution.

[0198] In this optional embodiment, the main stream hydrological data refers to various data related to water quality in the main channel of the river, and the tributary hydrological data refers to various data related to water quality in the tributaries of the river.

[0199] For example, Figure 4 Shown is a schematic diagram of the structure of the hydrological data described in this scheme.

[0200] For example, Figure 5 Shown is a schematic diagram of river hydrological observation involved in this plan.

[0201] In an optional embodiment, the training unit 111 is used to train a first regression model based on the river hydrological data, where the first regression model is used to characterize the correspondence between the main stream hydrological data and the tributary hydrological data.

[0202] In an optional embodiment, the first regression model is trained based on the river hydrological data, and the first regression model is used to characterize the corresponding relationship between the main stream hydrological data and the tributary hydrological data, including:

[0203] The tributary hydrological data is used as an independent variable, and the mainstream hydrological data is used as a dependent variable to construct a linear function;

[0204] Fitting the parameters of the linear function using the least squares method to obtain a parameter set;

[0205] Each parameter in the parameter set is matched with the tributary hydrological data to construct a first regression model.

[0206] In this optional embodiment, the river may include n tributaries and n is a positive integer. For example, in this solution, n may be equal to 3. The hydrological data of each tributary may be used as an independent variable. For example, the hydrological data of the first tributary may be recorded as Where x1 is a vector, the first dimension of the vector x1 is the COD concentration value in the tributary, the second dimension is the ammonia nitrogen concentration value in the tributary, and the third dimension is the total phosphorus concentration value in the tributary. Similarly, the hydrological data of the second tributary can be recorded as The hydrological data of the third tributary can be recorded as The hydrological data of the main stream can be recorded as X = [X COD , X AD , X ZL ].

[0207] In this optional embodiment, the hydrological data of the first tributary represented by x1, the hydrological data of the second tributary represented by x2, and the hydrological data of the third tributary represented by x3 can be used as independent variables, and the hydrological data of the main stream represented by X can be used as a dependent variable. A linear function is constructed based on the independent variables and the dependent variables, and the linear function satisfies the following relationship:

[0208]

[0209] Among them, a, b, c and d represent the parameters of the linear function, a is used to characterize the degree of influence of the hydrological data of the first tributary on the hydrological data of the mainstream, b is used to characterize the degree of influence of the hydrological data of the second tributary on the hydrological data of the mainstream, c is used to characterize the degree of influence of the hydrological data of the third tributary on the hydrological data of the mainstream, and d is used to characterize the possible deviation of the linear function.

[0210] In this optional embodiment, since the hydrological data includes three different data, the parameters a, b, c and d can be recorded as:

[0211]

[0212] For ease of understanding, the linear function relationship can be expressed as:

[0213] X=a·x1+b·x2+c·x3+d

[0214] Among them, X represents the hydrological data of the main stream of the river simulated and calculated by the linear function, x1 represents the hydrological data of the first tributary, x2 represents the hydrological data of the second tributary, x3 represents the hydrological data of the third tributary, and a, b, c and d represent the parameters of the linear function.

[0215] In this optional embodiment, the least squares method can be used to fit the parameters of the linear function. The least squares method is implemented by inputting each set of independent variables and the corresponding dependent variables into the linear function, and calculating the square of the difference between the contents on both sides of the linear function equal sign. Exemplarily, the square of the difference D is calculated as follows:

[0216] D=(a·x1+b·x2+c·x3+dX) 2

[0217] In this optional embodiment, a corresponding difference value can be obtained for each set of independent variables and the corresponding dependent variable, and the square of each difference value is calculated and summed to obtain a second linear function, which satisfies the following relationship:

[0218] S(a, b, c, d)=D1+D2+…+D n

[0219] Among them, S(a, b, c, d) represents the second linear function, which is used to characterize that the variables in the second linear function are a, b, c, d; D1 represents the square of the difference obtained after inputting the first set of independent variables and the corresponding dependent variables into the linear function, and its subscript corresponds to the index of the number of independent variable and dependent variable groups.

[0220] In this optional embodiment, partial derivatives of the four variables in the second linear function may be calculated respectively to obtain four partial derivative formulas, which are respectively recorded as:

[0221] In this optional embodiment, the four partial derivative formulas can be made equal to 0 respectively and each partial derivative formula can be solved to obtain the values ​​of the four parameters a, b, c, and d, where each parameter corresponds to a value.

[0222] In this optional embodiment, a, b, c, and d may be used as a parameter set.

[0223] In this optional embodiment, each parameter in the parameter set may be matched with the hydrological data of its corresponding tributary according to the format in the linear function to construct a first regression model, and the first regression model satisfies the relationship:

[0224] f(x1,x2,x3)=Y=a·x1+b·x2+c·x3+d

[0225] Among them, f(x1, x2, x3) represents that the first regression model is a function with the tributary hydrological data as the unknown variable; Y represents the predicted value of the main stream hydrological data obtained by inputting the known tributary hydrological data into the first regression model; a, b, c and d represent the parameters of the first regression model calculated by the least squares method; x1 represents the hydrological data of the first tributary; x2 represents the hydrological data of the second tributary; x3 represents the hydrological data of the third tributary.

[0226] In an optional embodiment, the construction unit 112 is used to construct an objective function according to the river emission reduction amount and the first regression model, wherein the river emission reduction amount refers to the difference between the river hydrological data before and after the implementation of the emission reduction plan.

[0227] In an optional embodiment, the objective function is constructed based on the river emission reduction and the first regression model, wherein the river emission reduction refers to the difference between the river hydrological data before and after the implementation of the emission reduction plan, including:

[0228] The hydrological data of the tributaries before the implementation of the emission reduction plan is taken as the front quantity, and the hydrological data after the implementation of the emission reduction plan is taken as the back quantity, and the difference between the front quantity and the back quantity is calculated as the emission reduction amount of each tributary;

[0229] Input the front amount into the first regression model to obtain the front amount of the main stream, input the rear amount into the first regression model to obtain the rear amount of the main stream, and calculate the difference between the front amount of the main stream and the rear amount of the main stream as the main stream emission reduction amount;

[0230] An objective function is constructed based on the emission reduction of each tributary and the emission reduction of the main stream, and the objective function is used to constrain the cost-effectiveness of the emission reduction plan.

[0231] In this optional embodiment, the emission reduction scheme refers to reducing the concentration of a certain number of water pollution factors in each tributary of the river, with the aim of reducing water pollution in the main stream of the river as much as possible. Exemplarily, the water pollution factors in this scheme are the COD concentration, ammonia nitrogen concentration and total phosphorus concentration in the hydrological data.

[0232] In this optional embodiment, the tributary hydrological data before the emission reduction plan is implemented can be used as the pre-quantity and recorded as and The tributary hydrological data after the emission reduction plan is implemented can be used as the post-quantity and recorded as and

[0233] In this optional embodiment, the difference in hydrological data of each tributary before and after the implementation of the emission reduction scheme can be calculated as the emission reduction amount of each tributary. For example, in this scheme, the emission reduction amount of each tributary can be recorded as Δx1, Δx2 and Δx3, where the subscript represents the index of each tributary.

[0234] In this optional embodiment, the calculation method of the tributary emission reduction is:

[0235]

[0236] Among them, the superscript of variable x represents before or after the implementation of the emission reduction plan, and the subscript of variable x represents the index of the river tributary.

[0237] In this optional embodiment, the pre-quantity can be input into the first regression model to obtain the hydrological data of the mainstream before the emission reduction plan is implemented as the pre-quantity of the mainstream and recorded as X 前 , the X 前 The calculation method is:

[0238]

[0239] in, Represents the hydrological data of the first tributary before the implementation of the emission reduction plan, Represents the hydrological data of the second tributary before the emission reduction plan was implemented. represents the hydrological data of the third tributary before the implementation of the emission reduction plan, and a, b, c and d represent the parameters in the first regression model.

[0240] In this optional embodiment, the post-volume can be input into the first regression model to obtain the hydrological data of the main stream after the emission reduction plan is implemented as the post-volume of the main stream and recorded as X 后 , the X 后 The calculation method is:

[0241]

[0242] in, Represents the hydrological data of the first tributary after the emission reduction plan is implemented, Represents the hydrological data of the second tributary after the emission reduction plan was implemented. represents the hydrological data of the third tributary after the implementation of the emission reduction plan, and a, b, c and d represent the parameters in the first regression model.

[0243] In this optional embodiment, the difference between the amount before the main flow and the amount after the main flow can be calculated as the main flow emission reduction and recorded as ΔX. The calculation method of ΔX is:

[0244] ΔX=X 前 -X 后

[0245] Among them, X 前 represents the hydrological data of the mainstream river before the implementation of the emission reduction plan, X 后 Represents the hydrological data of the mainstream river after the emission reduction plan is implemented.

[0246] In summary, in this optional embodiment, the main stream emission reduction ΔX can be characterized as:

[0247]

[0248] In this optional embodiment, an objective function may be constructed based on the emission reduction of each tributary and the emission reduction of the main stream, and the objective function is used to constrain the cost-effectiveness of the emission reduction scheme, and the objective function satisfies the following relationship:

[0249]

[0250] Among them, G(Δx1, Δx2, Δx3) represents that the objective function is a function related to the emission reduction of tributaries, Δx1 represents the emission reduction of the first tributary, Δx2 represents the emission reduction of the second tributary, Δx3 represents the emission reduction of the third tributary, and a, b, c and d represent the parameters of the first regression model.

[0251] Exemplarily, when a=1, b=1, c=1, Δx1=10, Δx2=9, Δx3=8, the value of the objective function is:

[0252]

[0253] Then the value of the objective function is 1.

[0254] In this optional embodiment, when the objective function takes the minimum value, it indicates that the emission reduction of the tributary is smaller while the emission reduction of the main stream is larger, that is, a better emission reduction effect is obtained at a smaller cost, and the emission reduction scheme corresponding to the minimum value of the objective function has a higher cost-effectiveness.

[0255] In an optional embodiment, the optimization unit 113 is used to solve the objective function using a gradient descent method to obtain candidate emission reduction schemes, where the candidate emission reduction schemes are used to represent emission reduction schemes that meet constraint conditions.

[0256] In an optional embodiment, before using the gradient descent method to solve the objective function to obtain the candidate emission reduction scheme, the method further includes:

[0257] Select the maximum value of the hydrological data of each tributary of the river and the hydrological data of the main stream as the benchmark value;

[0258] Calculate the ratio of each hydrological data to the corresponding benchmark value as the standard value;

[0259] Each hydrological data was replaced with the corresponding standard value to obtain the standardized hydrological data.

[0260] In this optional embodiment, the maximum value of the hydrological data of each tributary of the river and the hydrological data of the main stream can be selected as the reference value. For example, the reference value of the first tributary can be recorded as The reference value of the second tributary can be The reference value of the third tributary is The reference value of the main flow is

[0261] In this optional embodiment, the ratio of each hydrological data to the corresponding reference value can be calculated as the standard value. For example, the hydrological data in the first tributary can be taken as an example. When a hydrological data in the first tributary is And the base value of the first tributary is When , the calculation method of the standard value corresponding to the hydrological data is:

[0262]

[0263] The standard value corresponding to the hydrological data is

[0264] In this optional embodiment, each hydrological data can be replaced by a corresponding standard value to obtain standardized hydrological data. The purpose is to uniformly convert all hydrological data to between (0,1) to eliminate the dimensional difference between the hydrological data, thereby reducing the amount of calculation in the subsequent gradient descent method to solve the objective function, thereby improving the efficiency of the gradient descent method.

[0265] In an optional embodiment, the objective function is solved by using a gradient descent method to obtain a candidate emission reduction scheme, and the candidate emission reduction scheme is used to represent an emission reduction scheme that satisfies preset constraints and has a good emission reduction amount, including:

[0266] a, randomly select an initial position and use it as the current position;

[0267] b, calculating a gradient of a current position, wherein the current position corresponds to a preset first emission reduction scheme;

[0268] c. Calculate the product of the preset step length data and the gradient as the descent distance of the current position;

[0269] d. If the descent distance does not meet the preset first judgment condition, the first emission reduction plan is updated according to the descent distance to obtain a second emission reduction plan, and the position corresponding to the second emission reduction plan is used as the current position and returns to step b.

[0270] In this optional embodiment, the specific implementation steps of using the gradient descent method to solve the objective function to obtain the candidate emission reduction scheme are:

[0271] A1: Randomly select an initial position, the initial position corresponds to a preset emission reduction scheme, which can be used as the first emission reduction scheme. For example, the initial position can be recorded as and The initial position may be used as the current position;

[0272] A2: Calculate the gradient of the current position. The calculation method of the gradient of the current position is:

[0273]

[0274]

[0275]

[0276] in, and represents the gradient at the current location, the superscript of the variable w represents the index of the emission reduction scheme, and the subscript of the variable w represents the index of the river tributary; represents the symbol of partial derivative, for example, Represents the objective function with respect to the variable The partial derivative of and When unchanged, The degree of influence of the change of on the change of the objective function value;

[0277] A3: In this optional embodiment, the product of the preset step length data and the gradient can be calculated as the descent distance D of the current position. In this solution, the preset step length data can be recorded as τ. Exemplarily, the step length data τ can be 0.01. The calculation method of the descent distance D is:

[0278]

[0279] in, represents the emission reduction of the jth tributary, Dj represents the distance that the emission reduction of the jth tributary decreases, j∈{1,2,3};

[0280] A4: If the descent distance does not meet the preset first discrimination condition, the first emission reduction scheme is updated according to the descent distance to obtain the second emission reduction scheme, the position corresponding to the second emission reduction scheme is used as the current position and the process returns to step A2; otherwise, the emission reduction scheme corresponding to the current position is used as a candidate emission reduction scheme, wherein the preset first discrimination condition is that the descent distance is less than a preset first threshold. Exemplarily, the value of the first threshold may be 0.1;

[0281] Exemplarily, the calculation method of the second emission reduction scheme is:

[0282]

[0283] in, represents the emission reduction of the jth tributary in the second emission reduction scheme, The superscript 1 in represents that the emission reduction scheme is the second emission reduction scheme. represents the emission reduction of the jth tributary in the first emission reduction scheme, The superscript 0 indicates that the emission reduction scheme is the first emission reduction scheme. j Represents the decrease distance of the emission reduction of the jth tributary, j∈{1,2,3}.

[0284] In this optional embodiment, a candidate emission reduction scheme can be obtained based on the above steps. The candidate emission reduction scheme is a set of data containing multiple numerical values, wherein each numerical value is used to characterize the emission reduction amount of each tributary hydrological data. Exemplarily, the candidate emission reduction schemes in this scheme can be recorded as Δx1, Δx2 and Δx3.

[0285] In an optional embodiment, the calculation unit 114 is used to calculate the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and input the difference into the first regression model to obtain the emission reduction result.

[0286] In an optional embodiment, the calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain the emission reduction result comprises:

[0287] At the same time point, the hydrological data of each tributary of the river is collected as the first real-time data, and at the same time point, the hydrological data of the main stream of the river is collected as the second real-time data;

[0288] Calculating a difference between the first real-time data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain a fitting value;

[0289] The difference between the fitting value and the second real-time data is calculated as the emission reduction result.

[0290] In this optional embodiment, the hydrological data of each tributary of the river can be collected at the same time point as the first real-time data. For example, the time point can be recorded as t, and the first real-time data can be recorded as and They are used to characterize the hydrological data of each tributary collected in real time.

[0291] In this optional embodiment, the hydrological data of the main stream of the river can be collected in real time at time t as the second real-time data. For example, the second real-time data can be recorded as Used to characterize the hydrological data of the river mainstream collected at time point t.

[0292] In this optional embodiment, the difference between the first real-time data and the candidate emission reduction scheme may be calculated. For example, the first real-time data may be and The candidate emission reduction scheme may be and The difference between the first real-time data and the candidate emission reduction scheme can be recorded as T1, T2 and T3 respectively, and the calculation method of the difference between the first real-time data and the candidate emission reduction scheme is:

[0293]

[0294]

[0295]

[0296] In this optional embodiment, the difference is used to characterize the value of the hydrological data of each tributary of the river after the candidate emission reduction scheme is implemented. The difference can be input into the first regression model to obtain a fitting value, and the fitting value is used to characterize the value of the hydrological data of the main stream of the river after the candidate emission reduction scheme is implemented. For example, the value of the hydrological data of the main stream of the river can be recorded as T0, and the calculation method of T0 is:

[0297] T0=a·T1+b·T2+c·T3+d

[0298] Among them, T0 represents the fitting value obtained by inputting the difference into the first regression model, which is used to characterize the value of the hydrological data of the main stream of the river after the implementation of the candidate emission reduction plan; T1, T2 and T3 respectively represent the hydrological data of each tributary of the river after the candidate emission reduction plan is implemented; a, b, c and d represent the parameters of the first regression model.

[0299] For example, when the and When T0 is calculated as:

[0300]

[0301] Then the value of the mainstream hydrological data after the implementation of the candidate emission reduction scheme represented by T0 is This value is used to represent the COD concentration of 11 mg / L, the ammonia nitrogen concentration of 7 mg / L, and the total phosphorus concentration of 1 mg / L in the mainstream of the river after the implementation of the candidate emission reduction plan.

[0302] In this optional embodiment, the difference between the value of T0 and the second real-time data can be calculated as the emission reduction result, and the emission reduction result is used to characterize the amount of decrease in the hydrological data of the mainstream of the river after the implementation of the candidate emission reduction scheme. The emission reduction result can be recorded as R. For example, when the value of T0 is And the second real-time data is When , the calculation method of the emission reduction result R is:

[0303]

[0304] Then the emission reduction result R is This value is used to represent that after the implementation of the candidate emission reduction plan, the COD concentration in the mainstream of the river decreased by 4 mg / L, the ammonia nitrogen concentration decreased by 3 mg / L, and the total phosphorus concentration decreased by 2 mg / L.

[0305] In this way, the emission reduction results of the mainstream hydrological data after the implementation of the emission reduction plan are calculated through the real-time collected hydrological data, which can simulate the emission reduction effect and provide data support for the evaluation of the emission reduction plan, thereby improving the performance of the emission reduction plan.

[0306] In an optional embodiment, the recommendation unit 115 is used to evaluate the candidate emission reduction schemes according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and recommend the calibrated emission reduction scheme.

[0307] In an optional embodiment, evaluating the candidate emission reduction schemes according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and recommending the calibrated emission reduction scheme includes:

[0308] Comparing the real-time emission reduction result with the preset environmental protection standard to obtain a comparison result, wherein the comparison result includes "qualified" and "unqualified";

[0309] If the comparison result is "qualified", the candidate emission reduction scheme is used as a calibrated emission reduction scheme and the calibrated emission reduction scheme is recommended;

[0310] If the comparison result is "unqualified", the candidate emission reduction scheme is marked as "unqualified" and the objective function is solved again using the gradient descent method to obtain the next candidate emission reduction scheme.

[0311] In this optional embodiment, the real-time emission reduction result can be compared with the preset environmental protection standard to obtain a comparison result, and the real-time emission reduction result includes multiple components, which respectively represent the numerical value of the COD concentration reduction in the main stream of the river, the numerical value of the ammonia nitrogen concentration reduction, and the numerical value of the total phosphorus concentration reduction. The preset environmental protection standard includes multiple components, which respectively represent the numerical value of the COD concentration reduction in the main stream of the river expected by the environmental protection department, the numerical value of the ammonia nitrogen concentration reduction, and the numerical value of the total phosphorus concentration reduction. If the absolute value of each component in the real-time emission reduction result is greater than the absolute value of the corresponding component in the preset environmental protection standard, the comparison result is "qualified", the candidate emission reduction plan is used as the calibrated emission reduction plan, and the calibrated emission reduction plan is recommended to the environmental protection responsible unit to protect the river water quality.

[0312] Exemplarily, when the real-time emission reduction result is And the preset environmental protection standards are When the absolute value of the real-time emission reduction result is And the absolute value of the preset environmental protection standard is If the absolute value of each component in the real-time emission reduction result is greater than the absolute value of each component in the environmental protection standard, the candidate emission reduction plan corresponding to the real-time emission reduction result can be used as a calibrated emission reduction plan and recommended to the environmental protection unit to protect the river water quality.

[0313] In this optional embodiment, if the absolute value of the emission reduction result is not greater than the absolute value of the preset environmental protection standard, the comparison result is "unqualified", and the candidate emission reduction scheme can be marked as "unqualified", and the objective function is solved again using the gradient descent method to obtain the next candidate emission reduction scheme. The real-time emission reduction result is recalculated based on the next candidate emission reduction scheme to re-evaluate the next candidate emission reduction scheme.

[0314] like Figure 3 , which is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device 1 includes a memory 12 and a processor 13. The memory 12 is used to store computer-readable instructions, and the processor 13 executes the computer-readable instructions stored in the memory to implement the water quality reduction method based on artificial intelligence in any of the above embodiments.

[0315] In an optional embodiment, the electronic device 1 further includes a bus, a computer program stored in the memory 12 and executable on the processor 13, such as an emission reduction program based on artificial intelligence.

[0316] Figure 3 Only the electronic device 1 having components 12-13 is shown, and those skilled in the art will understand that Figure 3The structure shown does not constitute a limitation on the electronic device 1 , and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0317] Combination Figure 1 The memory 12 in the electronic device 1 stores a plurality of computer-readable instructions to implement a water quality reduction method based on artificial intelligence, and the processor 13 can execute the plurality of instructions to implement:

[0318] Collect river hydrological data from a water quality monitoring database, wherein the river hydrological data includes mainstream hydrological data and tributary hydrological data;

[0319] Training a first regression model based on the river hydrological data, wherein the first regression model is used to characterize the corresponding relationship between the main stream hydrological data and the tributary hydrological data;

[0320] Constructing an objective function according to preset constraints and the first regression model;

[0321] Solving the objective function using a gradient descent method to obtain candidate emission reduction schemes, wherein the candidate emission reduction schemes are used to represent emission reduction schemes that meet constraint conditions;

[0322] Calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain a real-time emission reduction result;

[0323] The candidate emission reduction schemes are evaluated according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and the calibrated emission reduction scheme is recommended.

[0324] Specifically, the specific implementation method of the processor 13 for the above instructions can refer to Figure 1 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0325] Those skilled in the art will appreciate that the schematic diagram is merely an example of the electronic device 1 and does not constitute a limitation on the electronic device 1. The electronic device 1 may have either a bus structure or a star structure. The electronic device 1 may also include more or less other hardware or software than shown in the diagram, or a different arrangement of components. For example, the electronic device 1 may also include input and output devices, network access devices, etc.

[0326] It should be noted that the electronic device 1 is only an example, and other existing or future electronic products that are suitable for the present application should also be included in the protection scope of the present application and included here by reference.

[0327] Among them, the memory 12 includes at least one type of readable storage medium, and the readable storage medium may be non-volatile or volatile. The readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 12 may be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 12 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 12 may also include both an internal storage unit of the electronic device 1 and an external storage device. The memory 12 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the emission reduction program based on artificial intelligence, but also can be used to temporarily store data that has been output or is to be output.

[0328] In some embodiments, the processor 13 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips. The processor 13 is the control core (Control Unit) of the electronic device 1, and uses various interfaces and lines to connect the various components of the entire electronic device 1, and executes or executes programs or modules stored in the memory 12 (for example, executing an emission reduction program based on artificial intelligence, etc.), and calls the data stored in the memory 12 to execute various functions of the electronic device 1 and process data.

[0329] The processor 13 executes the operating system of the electronic device 1 and various installed applications. The processor 13 executes the applications to implement the steps in the above-mentioned embodiments of the water quality reduction method based on artificial intelligence, for example Figure 1 Steps shown.

[0330] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory 12 and executed by the processor 13 to complete the present application. The one or more modules / units may be a series of computer-readable instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the electronic device 1. For example, the computer program may be divided into a collection unit 110, a training unit 111, a construction unit 112, an optimization unit 113, a calculation unit 114, and a recommendation unit 115.

[0331] The above-mentioned integrated unit implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a computer device, or a network device, etc.) or a processor to execute a part of the water quality reduction method based on artificial intelligence described in each embodiment of the present application.

[0332] If the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also instruct the relevant hardware devices to complete through a computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented.

[0333] The computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory and other memory, etc.

[0334] Furthermore, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc.

[0335] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0336] The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one arrow is used in the figure, but it does not mean that there is only one bus or one type of bus. The bus is configured to realize the connection and communication between the memory 12 and at least one processor 13, etc.

[0337] Although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to each component. Preferably, the power source may be logically connected to the at least one processor 13 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, and power status indicators. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0338] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0339] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0340] An embodiment of the present application also provides a computer-readable storage medium (not shown), in which computer-readable storage medium is stored computer-readable instructions, and the computer-readable instructions are executed by a processor in an electronic device to implement the artificial intelligence-based water quality reduction method described in any of the above embodiments.

[0341] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0342] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0343] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0344] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0345] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the specification can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any specific order.

[0346] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical solution of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present application.

Claims

1. A water quality reduction method based on artificial intelligence, characterized in that: The method comprises: Collect river hydrological data from a water quality monitoring database, wherein the river hydrological data includes mainstream hydrological data and tributary hydrological data; Training a first regression model based on the river hydrological data, wherein the first regression model is used to characterize the corresponding relationship between the main stream hydrological data and the tributary hydrological data; constructing an objective function based on the river emission reduction and the first regression model, wherein the river emission reduction refers to the difference in river hydrological data before and after the implementation of the emission reduction plan; The objective function is solved by the gradient descent method to obtain a candidate emission reduction scheme, and the candidate emission reduction scheme is used to characterize the emission reduction scheme that meets the constraint conditions, including: a, randomly selecting an initial position and taking it as the current position; b, calculating the gradient of the current position, and the current position corresponds to a preset first emission reduction scheme; c, calculating the product of the preset step size data and the gradient as the descent distance of the current position; d, if the descent distance does not meet the preset first discrimination condition, then updating the first emission reduction scheme according to the descent distance to obtain a second emission reduction scheme, taking the position corresponding to the second emission reduction scheme as the current position and returning to step b, otherwise, taking the emission reduction scheme corresponding to the current position as the candidate emission reduction scheme; Calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain the real-time emission reduction result, including: collecting the hydrological data of each tributary of the river at the same time point as the first real-time data, and collecting the hydrological data of the main stream of the river at the same time point as the second real-time data; calculating the difference between the first real-time data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain a fitting value; calculating the difference between the fitting value and the second real-time data as the real-time emission reduction result; The candidate emission reduction schemes are evaluated according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and the calibrated emission reduction scheme is recommended.

2. The water quality reduction method based on artificial intelligence according to claim 1, characterized in that: The first regression model is trained based on the river hydrological data, and the first regression model is used to characterize the corresponding relationship between the main stream hydrological data and the tributary hydrological data, including: The tributary hydrological data is used as an independent variable, and the mainstream hydrological data is used as a dependent variable to construct a linear function; Fitting the parameters of the linear function using the least squares method to obtain a parameter set; Each parameter in the parameter set is matched with the tributary hydrological data to construct a first regression model.

3. The water quality reduction method based on artificial intelligence according to claim 1, characterized in that: The objective function is constructed based on the river emission reduction and the first regression model, wherein the river emission reduction refers to the difference between the river hydrological data before and after the implementation of the emission reduction plan, including: The tributary hydrological data before the implementation of the emission reduction plan is taken as the front quantity, and the tributary hydrological data after the implementation of the emission reduction plan is taken as the back quantity, and the difference between the front quantity of the tributary hydrological data and the back quantity of the tributary hydrological data is calculated as the emission reduction amount of each tributary; Input the front amount into the first regression model to obtain the front amount of the main stream, input the rear amount into the first regression model to obtain the rear amount of the main stream, and calculate the difference between the front amount of the main stream and the rear amount of the main stream as the main stream emission reduction amount; An objective function is constructed based on the emission reduction of each tributary and the emission reduction of the main stream, and the objective function is used to constrain the cost-effectiveness of the emission reduction plan.

4. The water quality reduction method based on artificial intelligence according to claim 1, characterized in that: Before using the gradient descent method to solve the objective function to obtain the candidate emission reduction scheme, the hydrological data needs to be standardized. The standardized processing of the hydrological data includes: Select the maximum value of the hydrological data of each tributary of the river and the hydrological data of the main stream as the benchmark value; Calculate the ratio of each hydrological data to the corresponding benchmark value as the standard value; Each hydrological data was replaced with the corresponding standard value to obtain the standardized hydrological data.

5. The water quality reduction method based on artificial intelligence according to claim 1, characterized in that: The step of evaluating the candidate emission reduction schemes according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and recommending the calibrated emission reduction scheme comprises: Comparing the real-time emission reduction result with the preset environmental protection standard to obtain a comparison result, wherein the comparison result includes "qualified" and "unqualified"; If the comparison result is "qualified", the candidate emission reduction scheme is used as the calibrated emission reduction scheme and the calibrated emission reduction scheme is recommended; If the comparison result is "unqualified", the candidate emission reduction scheme is marked as "unqualified" and the objective function is solved again using the gradient descent method to obtain the next candidate emission reduction scheme.

6. A water quality reduction device based on artificial intelligence, characterized in that: The device is used to implement the method according to any one of claims 1 to 5, and the device comprises: A collection unit, used to collect river hydrological data within a preset time period from a water quality monitoring database, wherein the river hydrological data includes mainstream hydrological data and tributary hydrological data; A training unit, used for training a first regression model based on the river hydrological data, wherein the first regression model is used for characterizing the correspondence between the main stream hydrological data and the tributary hydrological data; A construction unit, used to construct an objective function according to preset constraints and the first regression model; An optimization unit, used for solving the objective function by using a gradient descent method to obtain a candidate emission reduction scheme, wherein the candidate emission reduction scheme is used to represent an emission reduction scheme that satisfies a constraint condition; A calculation unit, used for calculating the difference between the real-time collected tributary hydrological data and the candidate emission reduction scheme, and inputting the difference into the first regression model to obtain a real-time emission reduction result; A recommendation unit is used to evaluate the candidate emission reduction schemes according to the real-time emission reduction results to obtain a calibrated emission reduction scheme, and recommend the calibrated emission reduction scheme.

7. An electronic device, characterized in that: The electronic device comprises: a memory storing computer readable instructions; and A processor executes computer-readable instructions stored in the memory to implement the artificial intelligence-based water quality reduction method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions are executed by a processor in an electronic device to implement the artificial intelligence-based water quality reduction method as described in any one of claims 1 to 5.

Images