Method and system for measuring sewage index

Abstract

The application provides a sewage index determination method, a sewage index determination system and a storage medium. The sewage index determination method comprises the following steps: obtaining first historical data of a sewage treatment process and a query sample to be predicted; determining a first target function according to the first historical data; adding an active factor to the first target function to determine a second target function; determining a prediction function corresponding to the query sample according to the second target function; and performing regression analysis on the query sample according to the prediction function and the first historical data to determine the value of the sewage index of the query sample.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sewage treatment, and particularly relates to a sewage index determination method, a sewage index determination system and a corresponding computer readable storage medium. BACKGROUND

[0002] In the process of sewage treatment, whether the water quality index can be fed back in time is extremely crucial. Artificial detection of the water quality index is a feasible method, but since many water quality indexes are involved in the process of sewage treatment, if these water quality indexes cannot be obtained in time, the efficiency of the sewage treatment process will be greatly reduced. The use of online monitoring instruments can effectively solve the timeliness problem, but the price of these instruments is relatively high. If a fault occurs, the cost of maintenance is also relatively high.

[0003] In the field of sewage treatment, Bayesian neural networks are often used to measure and predict uncertainty, but it needs to manually set some parameters and has poor interpretability, and the prediction performance is poor when the noise is large, and the water quality index value cannot be efficiently and accurately predicted.

[0004] In order to overcome the above-mentioned defects existing in the prior art, the technical field urgently needs a sewage index determination method for accurately predicting the sewage index in the case of large noise to efficiently and accurately predict the sewage index value. SUMMARY

[0005] The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects, and is intended to neither identify key or critical elements of all aspects nor delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that is presented later.

[0006] In order to overcome the above-mentioned defects existing in the prior art, the present application provides a sewage index determination method, a sewage index determination device and its corresponding computer readable storage medium, which can accurately predict the sewage index in the case of large noise to efficiently and accurately predict the sewage index value.

[0007] Specifically, the method for determining the pollutant according to the first aspect of the present application comprises the following steps: obtaining first historical data of a sewage treatment process and a query sample to be predicted. A first objective function is determined according to the first historical data. An active factor is added to the first objective function to determine a second objective function. A prediction function corresponding to the query sample is determined according to the second objective function. Regression analysis is performed on the query sample according to the prediction function and the first historical data to determine the value of the sewage index of the query sample.

[0008] Further, in some embodiments of the present application, the step of determining the first objective function according to the first historical data comprises determining a latent function and an edge distribution of the corresponding output of the latent function according to the first historical data, wherein the latent function is assumed to be Gaussian distribution under a Gaussian process regression framework. A latent conditional probability relative to the first historical data is determined to obtain the first objective function.

[0009] Further, in some embodiments of the present application, the kernel used by the latent function under the Gaussian process regression framework includes but is not limited to a radial basis function kernel.

[0010] Further, in some embodiments of the present application, the step of determining the prediction function corresponding to the query sample according to the second objective function comprises determining the gradient of the second objective function with respect to a hyperparameter. The hyperparameter is optimized by a conjugate gradient method to determine reasonable parameter weights for the output of the first historical data, so as to determine the corrected prediction function. The step of performing regression analysis on the query sample according to the prediction function and the first historical data to determine the value of the sewage index of the query sample comprises inputting the first historical data into the prediction function to determine a first test output corresponding to the first historical data. White noise is added to the first test output to determine a first box plot of the prediction function. A corresponding second box plot is determined according to another regression method. The prediction function is optimized according to the first box plot and the second box plot. The query sample is input into the optimized prediction function to obtain the value of the sewage index of the query sample.

[0011] Further, in some embodiments of the present application, the other regression method is selected from at least one of a traditional Gaussian process regression method, a full Gaussian process regression method, and a partial Gaussian process regression method.

[0012] Further, in some embodiments of the present application, the step of optimizing the prediction function according to the first box plot and the second box plot comprises optimizing the parameter weights of the prediction function in response to the result distribution of the second box plot being more compact than the result distribution of the first box plot.

[0013] Further, in some embodiments of the present application, the step of optimizing the parameter weights of the prediction function comprises: obtaining second historical data of the wastewater treatment process. changing the parameter weights to determine a candidate prediction function; inputting the second historical data into the candidate prediction function to determine a second test output corresponding to the second historical data. adding white noise to the second test output to determine a third boxplot of the candidate prediction function. and repeating the steps of obtaining the second historical data, changing the parameter weights, determining the second test output, and determining the third boxplot until the result distribution of the third boxplot is more compact than the result distribution of the first boxplot.

[0014] In addition, the above-mentioned wastewater index measuring system according to the second aspect of the present application comprises a memory and a processor. The processor is connected to the memory and is configured to implement the above-mentioned wastewater index measuring method according to the first aspect of the present application.

[0015] In addition, the above-mentioned computer readable storage medium according to the third aspect of the present application has computer instructions stored thereon. When the computer instructions are executed by a processor, the above-mentioned wastewater index measuring method according to the first aspect of the present application is implemented. BRIEF DESCRIPTION OF DRAWINGS

[0016] The above features and advantages of the present application can be better understood by reading the following detailed description of embodiments of the present application in conjunction with the drawings, in which: in the drawings, components are not necessarily drawn to scale, and components having similar related functions or features can have the same or similar reference numbers.

[0017] Figure 1 A schematic diagram of a wastewater index measuring device according to some embodiments of the present application is shown;

[0018] Figure 2 A flowchart of a wastewater index measuring method according to some embodiments of the present application is shown;

[0019] Figure 3 Boxplots of Gaussian regression method, partial Gaussian regression method, and maximum composite conditional likelihood Gaussian regression method according to some embodiments of the present application are shown. DETAILED DESCRIPTION

[0020] The following detailed description together with the accompanying drawings will provide a fuller understanding of the application. Although the application is described in connection with the preferred embodiments, the application is not intended to be limited to the preferred embodiments. Indeed, instead of being restricted to the preferred embodiments, the application can be used with any number of variations and modifications. While the application is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that there is no intent to limit the application to the particular methods disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the application as defined by the appended claims.

[0021] In the description of the present application, it is necessary to explain that, unless explicitly defined and limited, the terms "mounting", "connection", "linking" should be understood in a broad sense, for example, it can be fixed connection, or detachable connection, or integral connection; it can be mechanical connection, or electrical connection; it can be direct connection, or indirect connection through intermediate medium, or internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0022] In addition, "up", "down", "left", "right", "top", "bottom", "horizontal", "vertical" used in the following description should be understood as the orientation shown in the section and the related drawings. The relative terms are only used for the convenience of description, and they do not mean that the device described should be manufactured or operated in a particular orientation, so they should not be understood as a limitation of the application.

[0023] It can be understood that although the terms "first", "second", "third" and the like can be used herein to describe various components, regions, layers and / or parts, these components, regions, layers and / or parts should not be limited by these terms, and these terms are only used to distinguish different components, regions, layers and / or parts. Therefore, the first component, region, layer and / or part discussed below can be called the second component, region, layer and / or part without departing from some embodiments of the application.

[0024] In the field of sewage treatment, Bayesian neural network is often used to measure the uncertainty of prediction, but it needs to manually set some parameters, and the interpretability is poor, and the prediction performance is poor when the noise is large, and it cannot efficiently and accurately predict the numerical value of the index of water quality.

[0025] In order to overcome the above-mentioned defects in the prior art, the present application provides a sewage index determination method, a sewage index determination device and a corresponding computer readable storage medium, which are used for accurately predicting the sewage index under the condition of abnormal value and large input / output noise, and efficiently and accurately predicting the value of the sewage index.

[0026] In some non-limiting embodiments, the above-mentioned sewage index determination method provided by the first aspect of the present application can be implemented via the above-mentioned sewage index determination device provided by the second aspect of the present application. Specifically, the sewage index determination device is configured with a memory and a processor. The memory includes but is not limited to the above-mentioned computer readable storage medium provided by the third aspect of the present application, and computer instructions are stored on the memory. The processor is connected to the memory and is configured to execute the computer instructions stored on the memory to implement the sewage index determination method provided by the first aspect of the present application.

[0027] First refer to Figure 1 , Figure 1 The figure shows the architecture of the pollutant determination device according to some embodiments of the present application.

[0028] Figure 1 The figure shows the architecture of the pollutant determination device according to some embodiments of the present application. The sewage index determination device includes an internal communication bus 301, a processor 302, a read-only memory (ROM) 303, a random access memory (RAM) 304, a communication port 305, and a hard disk 307. The internal communication bus 301 can realize data communication between the components of the sewage index determination device. The processor 302 can make judgments and issue prompts. In some embodiments, the processor 302 can be composed of one or more processors. The communication port 305 can realize data transmission and communication between the sewage index determination device and external input / output devices. In some embodiments, the sewage index determination device can send and receive information and data from the network through the communication port 305. In some embodiments, the sewage index determination device can perform data transmission and communication with external input / output devices in a wired form through the input / output port 306. The sewage index determination device also includes different forms of program storage units and data storage units, such as the hard disk 307, the read-only memory (ROM) 303 and the random access memory (RAM) 304, which can store various data files used by the computer processing and / or communication, and possible program instructions executed by the processor 302. The processor 302 executes these instructions to implement the main part of the method. The results processed by the processor 302 are transmitted to the external output device through the communication port 305 and displayed on the user interface of the output device.

[0029] The working principle of the above-mentioned wastewater index measuring device will be described below with reference to some embodiments of wastewater index measuring methods. Those skilled in the art will understand that these embodiments of the methods are merely non-limiting implementations provided by the present invention, intended to clearly demonstrate the main concepts of the invention and provide specific solutions convenient for public implementation, rather than limiting all functions or operating methods of the wastewater index measuring device. Similarly, the wastewater index measuring device is also only one non-limiting implementation provided by the present invention and does not limit the entities performing the steps in these wastewater index measuring methods.

[0030] Please refer to Figure 2 , Figure 2 A flowchart of a method for measuring wastewater indicators according to some embodiments of the present invention is shown.

[0031] like Figure 2 As shown in step S1, in the process of measuring wastewater treatment indicators, the present invention can first obtain the first historical data of the wastewater treatment process and the query sample to be predicted. Then, as... Figure 2 As shown in step S2, the first objective function is determined based on the aforementioned first historical data. Then, as... Figure 2 As shown in step S3, an active factor is added to the first objective function to determine the modified second objective function. Then, as... Figure 2 As shown in step S4, the prediction function corresponding to the above query sample is determined based on the second objective function. Then, as... Figure 2 As shown in step S5, the present invention can determine the prediction function corresponding to the above query sample based on the second objective function.

[0032] Optionally, in some embodiments of the present invention, those skilled in the art can obtain wastewater treatment data by installing sensors for measuring the above-mentioned wastewater indicators, and establish a wastewater indicator database containing such historical data.

[0033] Optionally, in some embodiments of the present invention, n process quantity samples can be selected from the above-mentioned wastewater index database as training samples, and the remaining samples can be used as test samples. Here, n is a preset quantity value, which can be one of 1 / 2, 2 / 3, 3 / 4, or 4 / 5 of the number of samples in the above-mentioned wastewater index database.

[0034] Optionally, in some embodiments of the present invention, the water quality indicators in the wastewater treatment process include, but are not limited to, at least one of dissolved oxygen and total biochemical oxygen demand.

[0035] Those skilled in the art can understand that the sensor of the above sewage index is only a non-limiting embodiment provided by the present application, which is intended to measure the historical index of the above sewage treatment to obtain the historical data of the sewage index, and is not used to limit the protection scope of the present application.

[0036] Further, the present application can determine a latent function and its corresponding marginal distribution of output according to the above first historical data, wherein the latent function is assumed to be Gaussian distribution under the Gaussian process regression framework. And determine the latent conditional probability with respect to the above first historical data to obtain the above first objective function.

[0037] Further, the kernel used by the latent function under the Gaussian process regression framework includes but is not limited to the radial basis function kernel.

[0038] Specifically, the relationship between the multivariate input and its corresponding output is:

[0039] y i =f i (X)+ε i

[0040] Where y i is the i-th sample of the output, f i (X) represents the key indicator (KPI), and X is n process measurement data samples from D sensors.

[0041] Suppose n samples of y are collected, y = [y1, y2, y3,..., y n ] is an n KPI sample vector, and the n KPI sample vector follows:

[0042] y = f(X) + ε

[0043] Where ε is noise, each element of which follows Gaussian distribution σ is the standard deviation to determine the amplitude of the distribution, σ 2 is the variance; f(X) is called a latent function, which is assumed to be Gaussian distribution under the GPR framework:

[0044] f(X) ~ N(0, K(X, X))

[0045] The expression of the marginal distribution of the KPI sample vector y is:

[0046] p(y|X) = ∫p(y|f(X))p(f(X)|X)df(X) K(X, X) is an n x n kernel matrix. The commonly used kernel is the radial basis function (RBF) kernel:

[0047]

[0048] If B is a matrix, diag{B} denotes a diagonal matrix containing only the diagonal elements of matrix B, D is the dimension of the input variable, represents the parameter that changes the importance of the variable.

[0049] The joint normal distribution of y and f * can be obtained by the assumed Gaussian distribution of f(X) under the GPR framework and the marginal distribution of y. The expression of the joint normal distribution of output y and latent output f * is:

[0050]

[0051] where the latent output f * is the latent output with respect to the test sample x * in the wastewater treatment process dataset, σ 2 is the variance, X is the n process measurement data sample from D sensors, and the conditional probability of f * can be obtained by the marginalization property of the joint normal distribution of y and f * , and the expression of the conditional probability of the latent output f * is:

[0052]

[0053] The objective function L is:

[0054]

[0055] The obtained objective function L is modified by adding an active factor:

[0056]

[0057] X1∩X2=Φ

[0058] X1∪X2=X

[0059] where n process measurement data samples are divided into X1 and X2, and y1, y2 are the outputs corresponding to X1 and X2.

[0060] For ease of understanding, logp2 is as follows:

[0061]

[0062] It is the probability density of the actual output y2 in the predictive distribution .

[0063] where:

[0064]

[0065] The modified target function is obtained as follows:

[0066] L' = log p1(y1|X1, θ) + log p2(y2|X2, X1, y1, θ)

[0067] = log p1(y1|X1, θ) * log p2(y2|X2, X1, y1, θ)

[0068] = log p(y2, y1|X2, X1, θ)

[0069] = log p(y|X, θ)

[0070] The above results show that the modified target function is still the conditional probability given X and θ, which is the same as the traditional target function. The difference is that the modified target function further sets the relationship between X and y, and the obvious change is that y2 and y1 are no longer conditionally independent.

[0071] The role of log p1 is the same as that of the above target function L, that is, to fit the training data as much as possible while meeting the requirement of relatively low model complexity, so as to maximize the log-likelihood value. The role of log p2 is to judge whether the model can make accurate predictions on the test samples in the wastewater treatment process data set. That is, when the training samples are contaminated by noise, due to the fitting requirement, the model constructed by (X1, y1) may be trained to contain too much noise information, thereby underestimating the noise term, resulting in poor prediction effect of total biochemical oxygen demand. On the contrary, influenced by (X2, y2), the trained model will be forced to fit (X2, y2), which becomes a constraint for hyperparameter selection, requiring the model to maintain the correct prediction trend. The modified target function not only focuses on the fitting of training data and the reduction of model complexity, but also emphasizes the weight generating function.

[0072] Further, the present application can determine the gradient of the above-mentioned second target function with respect to the hyperparameter, and optimize the above-mentioned hyperparameter by the conjugate gradient method, to determine reasonable parameter weights for the output of the above-mentioned first historical data, so as to determine the modified above-mentioned prediction function.

[0073] Specifically, let the modified target function L be log σ n Take the partial derivative, and obtain the optimal solution of the hyperparameter by the conjugate gradient method. After obtaining the hyperparameter, the function f * is obtained as follows:

[0074]

[0075] By changing X and y to X1 and y1 respectively, the form of the prediction function after optimization of the hyperparameter is obtained, wherein can be used for estimation, to represent the estimated variance information.

[0076] Further, the present application can input the first historical data into the prediction function to determine a first test output corresponding to the first historical data. Then, white noise is added to the first test output to determine a first boxplot of the prediction function. Then, a second boxplot is determined according to another regression method. Then, the prediction function is optimized according to the first boxplot and the second boxplot. Then, the query sample is input into the prediction function to obtain the value of the sewage index of the query sample.

[0077] Preferably, in some embodiments of the present application, the another regression method is selected from at least one of a traditional Gaussian process regression method, a full Gaussian process regression method, and a partial Gaussian process regression method.

[0078] Specifically, the present application can apply R^2 (R square) to quantify the prediction result of the model by adding white noise to the output.

[0079]

[0080] where y pre is the predicted output, y mean is the mean of the output sample, y * is the output of the test sample, [y * ] i is the i-th item of the output, and n represents a total of n output items.

[0081] Further, the present application can optimize the prediction function according to the first boxplot and the second boxplot. In response to a result that the result distribution of the second boxplot is more compact than the result distribution of the first boxplot, the parameter weight of the prediction function is optimized.

[0082] Please refer to Figure 3 , Figure 3 Fig. 1 shows the boxplot of the Gaussian regression method, the partial Gaussian regression method, and the maximum composite conditional likelihood Gaussian regression method according to some embodiments of the present application.

[0083] Specifically, the present application can compare the R2 values of the traditional Gaussian process regression method, the full Gaussian process regression method, the partial Gaussian process regression method, and the maximum composite conditional likelihood Gaussian regression method under white noise to evaluate the performance of different Gaussian process regression methods and obtain the maximum composite conditional likelihood Gaussian regression method model with the best prediction effect. For example, Figure 3As shown, the R^2 value of the maximum composite conditional likelihood Gaussian regression method model 31 is larger than that of the full Gaussian process regression method 32 and the partial Gaussian process regression method 33, which means that the predicted value of the sewage index is closer to the actual value of the sewage index, thereby illustrating that the present application can accurately predict the sewage index in the case of large noise (such as abnormal value, input and output noise, etc.), and efficiently and accurately predict the value of the sewage index.

[0084] Further, the present application can optimize the parameter weight of the above prediction function. Specifically, the present application can obtain the second historical data of the above sewage treatment process. Then, the parameter weight is changed to determine a candidate prediction function. After that, the second historical data is input into the candidate prediction function to determine the second test output corresponding to the second historical data. After that, white noise is added to the second test output to determine the third box plot of the candidate prediction function. After that, the steps of obtaining the second historical data, changing the parameter weight, determining the second test output, and determining the third box plot are repeated until the result distribution of the third box plot is more compact than that of the first box plot.

[0085] Specifically, the present application can compare the box plots corresponding to the maximum composite conditional likelihood Gaussian regression under different parameter weights to obtain the maximum composite conditional likelihood Gaussian regression method model with the best prediction effect, and perform multiple parameter tuning on the model. The parameters corresponding to the box plot with a more compact result distribution are taken as the optimal parameter weight, so as to realize the optimal prediction of the total biochemical oxygen demand of the sewage treatment process under the output noise interference.

[0086] Although the above methods are illustrated and described as a series of actions for the sake of simplicity of explanation, it should be understood and appreciated that the methods are not limited by the order of the actions, because according to one or more embodiments, some actions can occur in different orders and / or concurrently with other actions from those illustrated and described herein or not illustrated and described herein but can be appreciated by those skilled in the art.

[0087] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans can implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

[0088] Although the above-described embodiments of the wastewater indicator measurement system are implemented by a combination of software and hardware, it is understood that the wastewater indicator measurement system can be implemented in software, hardware, or a combination thereof. For a hardware implementation, the wastewater indicator measurement system can be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof. For a software implementation, the wastewater indicator measurement system can be implemented with separate software modules, such as procedures and functions, created and maintained by a general purpose program module created and maintained by a general purpose program module, each of which performs one or more of the functions described herein.

[0089] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor. The processor and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor and the storage medium can reside as discrete components in a user terminal.

[0090] In one or more exemplary embodiments, the functions described can be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functions can be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media can be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. Disk and disc, as used herein, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0091] The previous description of the disclosure is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the disclosure is not intended to be limited to the examples described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for determining wastewater indicators, characterized in that, Includes the following steps: Obtain the first historical data of the wastewater treatment process and the query sample to be predicted; Based on the first historical data, determine the first objective function; An active factor is added to the first objective function to determine the modified second objective function; Based on the second objective function, determine the prediction function corresponding to the query sample; The first historical data is input into the prediction function to determine the first test output corresponding to the first historical data; White noise is added to the first test output to determine a first box plot of the prediction function; The corresponding second box plot is determined using another regression method; The prediction function is optimized based on the first box plot and the second box plot; as well as The query sample is input into the optimized prediction function to obtain the value of the wastewater index of the query sample.

2. The determination method according to claim 1, characterized in that, The step of determining the first objective function based on the first historical data includes: Based on the first historical data, determine the latent function and its corresponding marginal distribution of the output, wherein the latent function is assumed to be Gaussian distributed within the Gaussian process regression framework; and Determine the potential conditional probabilities relative to the first historical data to obtain the first objective function.

3. The determination method according to claim 2, characterized in that, The kernel used for the latent function in the Gaussian process regression framework includes the radial basis function kernel.

4. The determination method according to claim 2, characterized in that, The step of determining the prediction function corresponding to the query sample based on the second objective function includes: Determine the gradient of the second objective function with respect to the hyperparameters; and The hyperparameters are optimized using the conjugate gradient method to determine reasonable parameter weights for the output of the first historical data, thereby determining the corrected prediction function.

5. The determination method according to claim 1, characterized in that, The other regression method is selected from at least one of the following: traditional Gaussian process regression method, full Gaussian process regression method, and partial Gaussian process regression method.

6. The measurement method according to claim 1, wherein the step of optimizing the prediction function based on the first box plot and the second box plot includes: In response to the result distribution of the second box plot being more compact than that of the first box plot, the parameter weights of the prediction function are optimized.

7. The determination method according to claim 6, wherein the step of optimizing the parameter weights of the prediction function includes: Obtain the second historical data of the wastewater treatment process; The parameter weights are changed to determine the candidate prediction function; The second historical data is input into the candidate prediction function to determine the second test output corresponding to the second historical data; White noise is added to the second test output to determine a third box plot of the candidate prediction function; and Repeat the steps of acquiring the second historical data, changing the parameter weights, determining the second test output, and determining the third box plot until the result distribution of the third box plot is more compact than the result distribution of the first box plot.

8. A wastewater index measurement system, comprising: Memory; as well as A processor, connected to the memory, and configured to implement the method for measuring wastewater indicators as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing computer instructions thereon, characterized in that, When the computer instructions are executed by the processor, the method for measuring wastewater indicators as described in any one of claims 1 to 6 is implemented.

Images