CN115034434A - Method, device and medium for analyzing sensitivity of main components in sewage treatment - Google Patents

Abstract

The invention discloses a method, a device and a medium for analyzing sensitivity of main components in sewage treatment, wherein the method comprises the following steps: and establishing a principal component model by utilizing the offline data, sampling and preprocessing the data of the input elements, sequentially calculating the output result of each real-time observation sample, carrying out fluctuation in a range on the sampled data, calculating the mean variance change caused by the fluctuation, and judging the sensitivity degree of the elements to the system. If the mean variance is larger, the control point is more sensitive to the system; on the contrary, the sensitivity is small, and the control effect is not obvious. The invention fully considers the characteristics of a large amount of control quantity and mutual influence among the control quantities in the sewage control process, and can effectively improve the control efficiency and save the control cost by judging the sensitivity of each element to each control quantity in advance; the sensitivity of the control quantity can be used as a reference criterion for selecting the control point. The invention can be widely applied to the technical field of sewage denitrification of sewage treatment plants.

Description

Method, device and medium for analyzing sensitivity of main components in sewage treatment

Technical Field

The invention relates to the technical field of sewage denitrification of sewage treatment plants, in particular to a method, a device and a medium for analyzing sensitivity of main components of sewage treatment.

Background

At present, the industrialization speed is increasingly accelerated, so that the discharge amount of sewage is obviously increased. In order to cope with the increasing industrial wastewater, the construction of a sewage treatment plant with a high degree of automation has become one of the needs for social development. Under such circumstances, the activated sludge process is widely used in sewage treatment plants as a sewage treatment process with low cost and good treatment effect. However, in the denitrification process by the activated sludge process, a plurality of control variables exist in the system, so that the main control variables cannot be accurately grasped, a good control effect is achieved at a low control cost, and the control cost is increased. The actual sewage plant condition is very complicated, the sludge denitrification process can suddenly occur under the combined action of various factors, the influence effect reaction of some elements on the system denitrification is insensitive, but the slight change of some elements can cause the obvious and violent fluctuation of the denitrification effect. Therefore, sensitivity analysis of control variables is an important prerequisite in wastewater treatment control processes.

The sensitivity analysis is a preparation work for the sewage control system in the early stage, so that the system can reach the ideal control effect at the lowest cost. In the early stage of sewage treatment, denitrification by an activated sludge process is mainly achieved by controlling the SNH content and the oxygen content in the system, and for the system, the variables are controlled well. However, the effect on the system is generally different for ammonia and dissolved oxygen at different control points. Therefore, the early stage control is not obvious, the control variables of the main control points cannot be accurately judged, the early stage sewage treatment effect is often not obvious and accurate, the control cost is increased, even the control elements are contradictory, the control effect is inhibited, the control cost is wasted, and the control effect cannot be achieved.

Disclosure of Invention

In order to solve at least one technical problem existing in the prior art to a certain extent, the invention aims to provide a method, a device and a medium for analyzing sensitivity of main components in sewage treatment.

The technical scheme adopted by the invention is as follows:

a method for analyzing sensitivity of main components in sewage treatment comprises the following steps:

s1, establishing a model for the sewage treatment system by using the off-line data;

s2, determining the number M of uncertain parameters for sensitivity analysis, and determining a sampling space;

s3, determining a sampling strategy, and acquiring sample data, wherein the number r of the samples, and the final number W of model evaluation is r (M + 1);

s4, normalizing the obtained sample data and then importing the normalized sample data into the established model to obtain an estimated output Y;

s5, calculating a basic effect index according to the sample data and the estimation output Y, and acquiring a basic effect mean value mu and a basic effect standard deviation sigma according to the basic effect index;

s6, repeatedly sampling, and calculating a basic effect mean value mu and a basic effect standard deviation sigma to obtain a basic effect matrix EE;

s7, subdividing the number r of the samples so as to evaluate whether the data set is converged; repeating convergence analysis by using a bootstrapping algorithm;

s8, judging whether the data are converged, if not, replacing the sample, and repeating the steps S3-S7;

s9, increasing the basic sampling number r2, and repeating the steps S3-S8;

s10, comparing the sensitivity of different basic effect mean values mu and basic effect standard deviations sigma to the sewage treatment system, and obtaining visual control variables and control points.

Further, the modeling of the sewage treatment system includes:

and establishing a BP neural network model for the sewage treatment system.

Further, the BP neural network model is constructed and trained in the following manner:

acquiring input data and output data, wherein the input data X is original observation data, and the output data Y is observation data obtained after the input data X is led into a sewage treatment system;

randomly ordering input data X, and dividing training data and prediction data;

carrying out normalization processing on the training data;

training the BP neural network according to the training data after normalization processing to obtain a BP neural network model; setting training times epochs, learning rate lr and training target minimum error coarse in the training process;

normalizing the predicted data;

and predicting the BP neural network model obtained by training by adopting the prediction data after the normalization processing. The predicted output is compared to the expected output to derive a prediction error. And dividing the error by the sample numerical value to obtain error percentage for judging the fitting degree of the model. And determining the BP neural network model net.

Further, the normalizing the training data includes:

and (3) preprocessing the training data matrix according to the following formula:

wherein X ∈ R ^m×n Is a matrix of raw observed data, R ^m×n A real matrix representing m rows and n columns, each row of X representing an observation sample, each column representing an observation variable, μ _X And σ _X And (3) representing the sample mean and the sample standard deviation of each observed variable in X, and X' representing the training set after pretreatment.

Further, the sampling policy is:

and (3) defining the sampling strategy as Latin hypercube sampling: the total sampling result is subjected to uniform distribution, as the number r of samples and the number M of uncertain elements are determined, the function lhdesign of Latin hypercube sampling returns an r multiplied by M matrix, and the elements of each column are randomly arranged; wherein r is the number of samples to be extracted, and r is the number of layers of the sample space, and the space r is equally divided to obtain (0,1/r), (1/r,2/r),., (1-1/r, 1).

Further, the calculation formula of the basic effect index is as follows:

in the formula (d) _i (j) The basic effect of the jth group of samples of the ith parameter, wherein j is 1, 2.. the R and R are repeated sampling times; n is the number of parameters; x is the number of _i Is the ith parameter; delta is the small variation of a single parameter; f (x) ₁ ,...,x _n ) Is output for a response corresponding to the set of parameters.

Further, the calculation formula of the fundamental effect mean value μ is as follows:

the calculation formula of the standard deviation σ of the basic effect is as follows:

mu reflects the strength of the influence of the variable on the output variable, and the larger the value of mu is, the stronger the sensitivity of the input variable is; sigma represents the strong and weak relation of interaction between input variables, and the larger the sigma value is, the stronger the interaction between parameters is.

Further, the step of determining the sampling space further comprises the step of dividing the sampling interval:

setting the number of sampling element variables as M and the number of samples in a sampling space as N; x belongs to R ^n×m Is a sampled spatial data matrix, R ^n×m A real matrix representing n rows and m columns, each row of X representing an observation sample and each column representing an observation variable, X (j) _min 、X(j) _max Respectively represent the maximum and minimum values in the j-th column of X, each observed variable being in the interval [ X (j) _min ,X(j) _max ]Following a uniform distribution.

The other technical scheme adopted by the invention is as follows:

an apparatus for analyzing sensitivity to a main component in sewage treatment, comprising:

at least one processor;

at least one memory for storing at least one program;

when executed by the at least one processor, cause the at least one processor to implement the method described above.

The other technical scheme adopted by the invention is as follows:

a computer readable storage medium, in which a program executable by a processor is stored, the program executable by the processor being for performing the method as described above when executed by the processor.

The invention has the beneficial effects that: the invention effectively refines and amplifies the sensitivity information contained in the mu statistic and the sigma statistic by a feature extraction method, effectively improves the control efficiency and saves the control cost.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following description is made on the drawings of the embodiments of the present invention or the related technical solutions in the prior art, it should be understood that the drawings in the following description are only for convenience and clarity of describing some embodiments in the technical solutions of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.

FIG. 1 is a detection flow chart of a method for analyzing sensitivity of main components in sewage treatment according to an embodiment of the present invention;

FIG. 2 is the BSM1-BOD of the first embodiment of the present invention ₅ A sensitivity analysis chart;

FIG. 3 is a BSM1-COD sensitivity analysis chart of example two of the present invention;

FIG. 4 is a BSM1-TN sensitivity analysis chart of example three of the present invention;

FIG. 5 is a BSM1-SNH sensitivity analysis chart of example four of the present invention.

Detailed Description

Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention and are not to be construed as limiting the present invention. For the step numbers in the following embodiments, they are set for convenience of illustration only, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.

In the description of the present invention, it should be understood that the orientation or positional relationship referred to in the description of the orientation, such as the upper, lower, front, rear, left, right, etc., is based on the orientation or positional relationship shown in the drawings, and is only for convenience of description and simplification of description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention.

In the description of the present invention, the meaning of a plurality of means is one or more, the meaning of a plurality of means is two or more, and larger, smaller, larger, etc. are understood as excluding the number, and larger, smaller, inner, etc. are understood as including the number. If the first and second are described for the purpose of distinguishing technical features, they are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.

In the description of the present invention, unless otherwise specifically limited, terms such as set, installation, connection and the like should be understood in a broad sense, and those skilled in the art can reasonably determine the specific meanings of the above terms in the present invention by combining the specific contents of the technical solutions.

Interpretation of terms:

the Bootstrapping algorithm refers to the process of re-creating new samples sufficient to represent the distribution of the parent samples through repeated sampling with limited sample data. The usage of bootstrapping is based on many statistical assumptions, and thus the accuracy of sampling can affect the establishment of assumptions.

Example one

As shown in fig. 1, this embodiment provides a method for analyzing sensitivity of principal components in sewage treatment, which analyzes sensitivity of main control elements in a sewage treatment process, and includes first establishing a principal component model (the principal component model is a model for analyzing original variables by using some linear combinations (i.e., principal components) of the original variables instead of the original variables, and is mainly established as a BP neural network) using data during normal working conditions of a sewage plant, i.e., offline data, and then preprocessing acquired real-time field data to calculate a mean value and a variance of a single variable. Under the condition of controlling other variables to be unchanged, the original data of a single variable fluctuate within a certain range, and the mean value and the variance are recalculated. And sequentially calculating the mean value and the variance of each variable, repeatedly sampling, and finally determining the sensitivity degree of the element to the system according to the mean value and the variance of different elements. The method specifically comprises the following steps:

s101, determining an observation variable, setting the same sampling interval, collecting data every fifteen minutes for 7 days, and collecting 672 data in total. Taking a data matrix collected from a sewage treatment plant under a normal operation condition as a training set, and performing data preprocessing (normalization processing) according to the following formula:

wherein X ∈ R ^m×n Is the original training set, R ^m×n A real matrix representing m rows and n columns, where each row represents an observed sample and each column represents an observed variable including temperature, PH, etc. Mu.s _X And σ _X The sample mean and sample standard deviation of each observed variable in X are shown, and X' represents the pre-processed data matrix.

S102, randomly sequencing 672 data in the X', and confirming training data and prediction data for building the BP neural network model. In this case, 600 randomly ordered data are taken as training data, and the last 72 data are taken as prediction data.

S103, carrying out BP network training next, initializing the network, setting the training times epochs as 100, the learning rate lr as 0.1, and the training target minimum error goal as 0.00004, so as to obtain a neural network model. And importing the normalized prediction data into the trained model, and performing network prediction output. And performing inverse normalization on the output of the model to obtain the prediction output of the actual model. And finally, comparing the predicted output with the expected output to obtain a model prediction error, and estimating the percentage of the network prediction error to obtain the fitting degree of the established model. If the error percentage is too large, the data is replaced and the model is rebuilt. And if the fitting degree reaches 85%, the model is applicable, and the BP neural network model net is determined. This completes the modeling.

S104, determining the number M of elements needing to consider sensitivity, and considering 5 variables in the experiment, wherein the variables are DO3, DO4, DO5, SNO2 and Q. The number of samples in the sampling space is N. X belongs to R ^n×m Is a sampled spatial data matrix, R ^n×m A real matrix representing n rows and m columns, each row of X representing one observation sample and each column representing one observation variable. X (j) _max 、X(j) _min Respectively representing the maximum and minimum values in column j of X. Wherein the maximum value and the minimum value are determined according to actual conditions. Each variable being in the interval [ X (j) _min ,X(j) _max ]Following a uniform distribution.

And S105, determining the sampling number, the sampling method and the sampling path in the sampling space. The number of samples r is set to 100. The sampling strategy is determined as Latin Hypercube Sampling (LHS), radial collection is carried out, and the sampling result of the sample is uniformly distributed.

S106, firstly carrying out data preprocessing on the collected sample, wherein the preprocessing method is the same as that of the training set. And introducing the sample X obtained after normalization into the established BP neural network model. The expected output Y is obtained.

And S107, calculating the mean value mu and the variance sigma of the basic effect coefficient according to the obtained sample X and the output Y. The Nboot (set to 100) times was sampled repeatedly to obtain multiple sets of data. And subdividing the sample into five groups, repeating the calculation, reducing the calculation amount and determining the convergence of the observation data. If the convergence is reached, a new group of samples is replaced, and the processes of S104-S107 are repeated, and finally, the element sensitivity comparison result is obtained.

The data of the BSM1 model was verified in all examples. The observation data comprises 672 samples which are sampled at intervals of 15 minutes, and 9 observation variables comprise DO3, DO4, DO5, SNO2, Q, COD and BOD ₅ Table 1 shows the meanings of the observed variables, TN and SNH.

TABLE 1 Experimental data Observation variables

As shown in fig. 1, the general process of the sensitivity analysis of sewage from sewage plants includes: the observation data is divided into two parts, wherein the first 600 data are used as a training set to construct a model, and the last 72 samples are used as a test set to verify the effectiveness of the method. Firstly, establishing a BP neural network model by using a training set, then preprocessing a test set, sequentially calculating the mean value mu and the variance sigma of each sample, and judging the sensitivity of each element to a system according to the fluctuation condition of the mean value and the variance.

In case one, DO3, DO4, DO5, SNO2 and Q are used as input, BOD ₅ A neural network model is established for the output. Study of five elements DO3, DO4, DO5, SNO2 and Q on BOD ₅ The sensitivity of the output controls BOD ₅ On the other hand, which of the five elements is selected can achieve better control.

After the scheme is adopted, variables which have large influence on the system can be analyzed in advance, the variables which are convenient to control are selected, the control cost can be greatly reduced, and corresponding countermeasures can be taken in advance even if the sensitive variables are not easy to control.

From the experimental results of fig. 2, DO3, DO4 are relative to the remaining three variables DO5, SNO2, Q for the output BOD ₅ The sensitivity is higher, and the dissolved oxygen of the third pool and the fourth pool is preferentially controlled when the selection of the control variable is considered. DO4 has a small mean value relative to DO3, in other words DO3 outputs BOD to the system ₅ The method is more sensitive, but the variance of DO3 is larger, the fluctuation is larger for controlling DO3, the dissolved oxygen of No. four ponds is not stable, and DO4 is preferentially controlled to be used for judging the BOD of the sludge ₅ A reference standard for the degree.

Example two

The input data used for modeling the BP neural network in example two was identical to that in example one, taking into account that the output variable was COD. The parameters set in the modeling process are the same as those in the first embodiment, and the sensitivity degree of each element is still judged by calculating the mean value and the variance. The results of the experiment are shown in FIG. 3, in relation to the variable BOD ₅ In contrast, the most sensitive to reaction fluctuations for COD was the dissolved oxygen in cell three, DO 3. The effects of DO4, DO5, SNO2 and Q on the system are gradually reduced. The average value of DO3 is about twice that of DO4, and the sensitivity degree of the DO is far beyond the influence of the dissolved oxygen in No. four pond DO4 on the system. DO3 is a good choice among the five variables chosen for controlling COD.

EXAMPLE III

In the third embodiment, the input data is the same as the experimental data in the first embodiment when the model is established, but the output data is changed into TN of the total nitrogen content in the sewage. The example of the scheme researches the sensitivity of five control variables including DO3, DO4, DO5, SNO2 and Q to total nitrogen TN in a BSM1 sewage treatment platform. Total nitrogen TN is an important control index in sewage treatment and is necessary for analyzing sensitive elements of TN in advance. The results of the experiment are shown in FIG. 4. The dissolved oxygen DO4 of the fourth pond and the dissolved oxygen DO3 of the third pond have higher sensitivity to the BSM1 sewage treatment system, although the variances of the two are close, the fluctuation of the DO3 variance is larger, the upper and lower bound span is too large, and the system control is unstable. When the TN control is adopted, the third pool dissolved oxygen amount DO4 can be preferentially controlled.

Example four

Still using the experimental data from example one for modeling the inputs, the output data was changed to the ammonia content SNH. The parameters set in the modeling process are the same as in the first embodiment.

From the experimental results shown in fig. 5, the standard deviation of DO3 and DO4 is slightly higher than that of the other three variables, but the mean value of DO4 is higher than that of DO3, the reaction of the dissolved oxygen in pond four DO4 is more sensitive to the SNH content of the ammonium ions, the fluctuation of the controlled variable is far less than that of DO3, and the dissolved oxygen in pond four can be used as the main variable for controlling the ammonia content in the sludge.

It can be seen from the above embodiments that the feature extraction principal component analysis developed by the present invention has more satisfactory performance for the control analysis of the system. Specific results of the sensitivity analysis for the exemplary cases are shown in tables 2 and 3. If a dependent variable is considered alone, the observation depends mainly on the mean value, the magnitude of which reflects the degree of reaction of the control variable to the system as it changes. When the mean values are close, the variance is referred to, and the variance can show whether the controlled variable can be stably controlled during the variation. The method can well select the most sensitive control quantity, but the sewage control system usually considers a plurality of elements, and the dependent variables mentioned in the cases need to be considered, so that the control elements and the control points can not be determined by only looking at the mean value and the variance of a single group of data, and can be determined according to the actual condition of a sewage plant.

TABLE 2 results of sensitivity analysis

TABLE 3 results of sensitivity analysis

In summary, compared with the prior art, the method of the embodiment has the advantages and beneficial effects that:

(1) the invention can effectively refine and amplify the sensitivity information contained in the mu statistic and the sigma statistic through the feature extraction method.

(2) The invention can effectively find out the sensitive easily-controlled elements by simply processing the traditional mu statistic and sigma statistic, thereby improving the control effect, saving the control cost and not increasing the complexity of the algorithm.

(3) The invention can effectively and accurately provide the control quantity and the control point in the early stage of sewage control in a sewage treatment plant. Therefore, the workload of workers in the sewage treatment plant can be reduced, the control effect can be improved, enough time is reserved for the maintenance of equipment in the sewage treatment plant, and the maintenance cost is saved.

The embodiment also provides a to sewage treatment principal component sensitivity analytical equipment, includes:

at least one processor;

at least one memory for storing at least one program;

when executed by the at least one processor, cause the at least one processor to implement the method of fig. 1.

The device for analyzing sensitivity of principal components in sewage treatment according to the embodiment of the present invention can perform the method for analyzing sensitivity of principal components in sewage treatment according to the embodiment of the present invention, and can perform any combination of the method embodiments, and has corresponding functions and advantages of the method.

Embodiments of the present application also disclose a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The computer instructions may be read by a processor of a computer device from a computer-readable storage medium, and executed by the processor to cause the computer device to perform the method illustrated in fig. 1.

The embodiment also provides a storage medium, which stores instructions or programs capable of executing the method for analyzing the sensitivity of the main components of sewage treatment provided by the embodiment of the method, and when the instructions or the programs are executed, the steps can be executed in any combination of the embodiment of the method, and the corresponding functions and beneficial effects of the method are achieved.

In alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flow charts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations described as part of larger operations are performed independently.

Furthermore, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the described functions and/or features may be integrated in a single physical device and/or software module, or one or more functions and/or features may be implemented in a separate physical device or software module. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary for an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be understood within the ordinary skill of an engineer, given the nature, function, and internal relationship of the modules. Accordingly, those skilled in the art can, using ordinary skill, practice the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative of and not intended to limit the scope of the invention, which is defined by the appended claims and their full scope of equivalents.

The functions may be stored in a computer-readable storage medium if they are implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the technical solution of the present invention or a part thereof which substantially contributes to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.

It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.

In the foregoing description of the specification, reference to the description of "one embodiment/example," "another embodiment/example," or "certain embodiments/examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (10)

1. A method for analyzing sensitivity of main components in sewage treatment is characterized by comprising the following steps:

s1, establishing a model for the sewage treatment system;

s2, determining the number M of uncertain parameters for sensitivity analysis, and determining a sampling space;

s3, determining a sampling strategy and acquiring sample data, wherein the number of the samples is r;

s4, normalizing the obtained sample data and then importing the normalized sample data into the established model to obtain an estimated output Y;

s5, calculating a basic effect index according to the sample data and the estimation output Y, and acquiring a basic effect mean value mu and a basic effect standard deviation sigma according to the basic effect index;

s6, repeatedly sampling, and calculating a basic effect mean value mu and a basic effect standard deviation sigma;

s7, judging whether the data are converged, if not, replacing the sample, and repeating the steps S3-S6;

s8, increasing the basic sampling number r2, and repeating the steps S3-S7;

s9, comparing the sensitivity of different basic effect mean values mu and basic effect standard deviations sigma to the sewage treatment system, and obtaining visual control variables and control points.

2. The method for analyzing sensitivity to the principal components of sewage treatment according to claim 1, wherein the modeling of the sewage treatment system comprises:

and establishing a BP neural network model for the sewage treatment system.

3. The method for analyzing sensitivity of principal components in sewage treatment according to claim 2, wherein the BP neural network model is constructed and trained in the following way:

acquiring input data X and output data Y, wherein the input data X is original observation data, and the output data Y is observation data obtained after the input data X is led into a sewage treatment system;

randomly ordering input data X, and dividing training data and prediction data;

carrying out normalization processing on the training data;

training the BP neural network according to the training data after the normalization processing to obtain a BP neural network model;

normalizing the predicted data;

and predicting the BP neural network model obtained by training by adopting the prediction data after the normalization processing.

4. The method for analyzing the sensitivity of the principal components in wastewater treatment according to claim 3, wherein the normalizing the training data comprises:

and (3) preprocessing the training data matrix according to the following formula:

wherein X ∈ R ^m×n Is a matrix of raw observed data, R ^m×n A real matrix representing m rows and n columns, each row of X representing an observation sample, each column representing an observation variable, μ _X And σ _X And (3) representing the sample mean and the sample standard deviation of each observed variable in X, and X' representing the training set after pretreatment.

5. The method for analyzing the sensitivity of the principal components in wastewater treatment according to claim 1, wherein the sampling strategy is:

and (3) defining the sampling strategy as Latin hypercube sampling: the total sampling result is subjected to uniform distribution, as the number r of samples and the number M of uncertain elements are determined, the function lhdesign of Latin hypercube sampling returns an r multiplied by M matrix, and the elements of each column are randomly arranged; wherein r is the number of samples to be extracted, and r is the number of layers of the sample space, and the space r is equally divided to obtain (0,1/r), (1/r,2/r),., (1-1/r, 1).

6. The method for analyzing the sensitivity of the principal components in wastewater treatment according to claim 1, wherein the basic effect index is calculated by the following formula:

in the formula (d) _i (j) The basic effect of the jth group of samples of the ith parameter, j is 1, 2.. and R is the repeated sampling times; n is the number of parameters; x is the number of _i Is the ith parameter; delta is the small variation of a single parameter; f (x) ₁ ,...,x _n ) Is output for a response corresponding to the parameter set.

7. The method for analyzing sensitivity to principal components of sewage treatment according to claim 6, wherein the calculation formula of the mean value μ of the fundamental effect is as follows:

the calculation formula of the standard deviation sigma of the basic effect is as follows:

mu reflects the strength of the influence of the variable on the output variable, and the larger the value of mu is, the stronger the sensitivity of the input variable is; sigma shows the strong and weak relation of interaction between input variables, and the larger the value of sigma is, the stronger the interaction between parameters is.

8. The method for analyzing the sensitivity to the principal components of wastewater treatment according to claim 1, wherein the step of determining the sampling space further comprises the step of dividing the sampling interval:

setting the number of sampling element variables as M and the number of samples in a sampling space as N; x is formed by R ^n×m Is sampling the spatial data momentArray, R ^{n ×m} A real matrix representing n rows and m columns, each row of X representing an observation sample and each column representing an observation variable, X (j) _min 、X(j) _max Respectively represent the maximum and minimum values in the j-th column of X, each observed variable being in the interval [ X (j) _min ,X(j) _max ]Following a uniform distribution.

9. A device for analyzing sensitivity to main components in sewage treatment is characterized by comprising:

at least one processor;

at least one memory for storing at least one program;

when executed by the at least one processor, cause the at least one processor to implement the method of any one of claims 1-8.

10. A computer-readable storage medium, in which a program executable by a processor is stored, wherein the program executable by the processor is adapted to perform the method according to any one of claims 1 to 8 when executed by the processor.

Images