A control method for a sewage treatment system for kitchen waste VFAs

Abstract

The present application relates to the field of intelligent sewage treatment, and specifically to a control method for a sewage treatment system for kitchen waste VFAs. It collects the time series of sewage inflow flow by a first flow meter, and collects the time series of VFAs organic acid addition flow by a second flow meter, and uses data analysis and processing technology based on deep learning to extract and associate local time series features of the sewage inflow flow and the VFAs organic acid addition flow, so as to adaptively adjust the VFAs organic acid addition flow at the current time point according to the aggregate response representation characteristics of the sewage inflow flow and the VFAs organic acid addition flow in the full time domain. In this way, by real-time monitoring of the sewage inflow flow and the VFAs organic acid addition flow, more accurate addition control can be achieved, thereby improving the sewage treatment efficiency and realizing the intelligent control of the sewage treatment system for kitchen waste VFAs.

Description

Technical Field

[0001] The present application relates to the field of intelligent sewage treatment, and specifically to a control method for a sewage treatment system for kitchen waste VFAs. Background Art

[0002] The rapid development of urbanization has led to an increase in urban sewage discharge, and the demand for sewage treatment has risen accordingly. Currently, there are traditional fossil carbon source substitutes with good biodegradability on the market, such as brewery wastewater, sugar wastewater, food processing wastewater, etc. Although some substitutes have improved biodegradability and reduced impurities through biochemical technology, they are still mainly used to adjust the nutrient ratio and fail to fully realize the resource potential. In the biochemical system of the sewage plant, the nutrients in the substitutes may compete with other microorganisms, affect the carbon source uptake of denitrifying bacteria, and reduce the denitrification efficiency.

[0003] VFA, whose Chinese name is Volatile Fatty Acid, is a type of fatty acid. Kitchen waste VFAs is a weakly acidic organic complex solution with a pH between 4 and 6. It is rich in organic components and can meet the nutritional needs of various microorganisms for growth and metabolism. It is suitable for various sewage treatment systems that utilize microbial degradation.

[0004] However, traditional VFAs organic acid dosing control often relies on preset static models and empirical parameters. In particular, excessive VFAs organic acid dosing will cause the sewage collection and treatment system to overload, which can easily lead to excess carbon sources. The remaining carbon source will become a nutrient source for microorganisms in other treatment links, resulting in increased sludge production and higher operating costs; insufficient dosing will cause incomplete denitrification and deammonification reactions, resulting in low deammonification efficiency of the entire sewage treatment system, and unable to fully utilize the external carbon source, reducing the value of the carbon source. In other words, traditional control cannot capture short-term fluctuations in sewage flow or immediate changes in VFAs demand, resulting in inaccurate dosing strategies, affecting sewage treatment efficiency and cost-effectiveness.

[0005] Therefore, an optimized control scheme for food waste VFAs wastewater treatment system is desired. Summary of the invention

[0006] This application is made in view of the above problems. One object of this application is to provide a control method for a sewage treatment system for kitchen waste VFAs.

[0007] The embodiment of the present application provides a control method for a wastewater treatment system for kitchen waste VFAs, which includes:

[0008] Acquire a time series of sewage inflow flow collected by a first flow meter;

[0009] Obtaining a time series of VFAs organic acid addition flow rate collected by a second flow meter;

[0010] Input the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow into a sequence encoder respectively to obtain a sequence of local time series correlation feature vectors of the sewage inflow flow and a sequence of local time series correlation feature vectors of the VFAs organic acid addition flow;

[0011] Input each corresponding group of local time series correlation feature vectors of sewage inflow flow and local time series correlation feature vectors of VFAs organic acid addition flow in the sequence of local time series correlation feature vectors of sewage inflow flow and the sequence of local time series correlation feature vectors of VFAs organic acid addition flow into an adaptive characteristic response analyzer to obtain a sequence of sewage flow-addition flow local time series response representation vectors;

[0012] Inputting the sequence of the sewage flow-dosage flow local time series response representation vector into the node feature propagation network based on the node energy decay mechanism to obtain the sewage flow-dosage flow full time series response aggregation representation vector as the sewage flow-dosage flow full time series response aggregation representation feature;

[0013] Based on the sewage flow-dosage flow full time series response aggregation representation characteristics, a control instruction is obtained, and the control instruction is used to indicate whether the VFAs organic acid dosing flow at the current time point should increase, decrease or remain unchanged.

[0014] For example, according to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow are respectively input into the sequence encoder to obtain the sequence of the local time series correlation feature vector of the sewage inflow flow and the sequence of the local time series correlation feature vector of the VFAs organic acid addition flow, including:

[0015] The time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow are respectively input into the sequence encoder based on the 1D-CNN model to obtain the sequence of the local time series correlation feature vectors of the sewage inflow flow and the sequence of the local time series correlation feature vectors of the VFAs organic acid addition flow.

[0016] For example, according to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, each group of corresponding sewage inflow flow local time series correlation feature vectors and VFAs organic acid addition flow local time series correlation feature vectors in the sequence of the sewage inflow flow local time series correlation feature vectors and the sequence of the VFAs organic acid addition flow local time series correlation feature vectors is input into the adaptive characteristic response analyzer to obtain a sequence of sewage flow-addition flow local time series response representation vectors, including:

[0017] Calculate the position-by-position response between the local time series correlation characteristic vector of the sewage inflow flow and the local time series correlation characteristic vector of the VFAs organic acid addition flow to obtain the position-by-position response characteristic vector of the sewage inflow flow-VFAs organic acid addition flow;

[0018] Using a Softmax function to normalize the sewage inflow flow-VFAs organic acid addition flow position-by-position response characteristic vector to obtain a normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response characteristic vector;

[0019] Inputting the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector into a learnable gating function to obtain a sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector;

[0020] Calculate the sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector and the normalized sewage inflow flow-VFAs organic acid addition flow location response feature vector to obtain a sewage inflow flow-VFAs organic acid addition flow location response distinguishable weight mask vector;

[0021] Calculate the position-by-position response of the sewage inflow flow-VFAs organic acid addition flow, and multiply the distinguishable weight mask vector by the position-by-position response feature vector of the sewage inflow flow-VFAs organic acid addition flow to obtain the sewage flow-addition flow local time series response representation vector.

[0022] For example, according to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector is input into a learnable gating function to obtain a sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector, including:

[0023] The negative number of each position characteristic value in the position-by-position response characteristic vector of the normalized sewage inflow flow-VFAs organic acid addition flow is used as the exponent of the natural constant to calculate the exponential function value with e as the base according to the position to obtain the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response class support vector;

[0024] The inverse of the sum of the eigenvalues ​​of each position in the position-by-position response class support vector of the normalized sewage inflow flow-VFAs organic acid addition flow and the constant one is calculated to obtain the sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector.

[0025] For example, according to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, the sequence of the sewage flow-dosing flow local time series response representation vector is input into the node feature propagation network based on the node energy decay mechanism to obtain the sewage flow-dosing flow full time series response aggregation representation vector, including:

[0026] Based on the maximum value, average value and variance of each sewage flow-dosage flow local time series response representation vector in the sequence of sewage flow-dosage flow local time series response representation vector, the energy statistical paradigm value of each sewage flow-dosage flow local time series response representation vector is calculated to obtain a sequence of sewage flow-dosage flow local time series response energy statistical paradigm values, wherein the energy statistical paradigm value corresponding to the current sewage flow-dosage flow local time series response representation vector in the sequence of sewage flow-dosage flow local time series response energy statistical paradigm values ​​is used as the current node energy statistical paradigm value and other energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of current sewage flow-dosage flow local time series response energy statistical paradigm values ​​and historical sewage flow-dosage flow local time series response energy statistical paradigm values;

[0027] Counting the node propagation space span values ​​between each other sewage flow-dosage flow local time series response representation vector in the sequence of sewage flow-dosage flow local time series response representation vector and the current sewage flow-dosage flow local time series response representation vector to obtain a sequence of sewage flow-dosage flow local time series response propagation space span values;

[0028] Based on the sequence of the propagation spatial span values ​​of the local time series response of sewage flow-dosing flow and the sequence of the energy statistical paradigm values ​​of the local time series response of historical sewage flow-dosing flow, determine the energy propagation attenuation coefficient values ​​of other local time series response representation vectors of sewage flow-dosing flow in the sequence of local time series response representation vectors of sewage flow-dosing flow to obtain a sequence of energy propagation attenuation coefficient values ​​of the local time series response of sewage flow-dosing flow, wherein the energy propagation attenuation coefficient value of the local time series response of sewage flow-dosing flow is inversely correlated with the energy statistical paradigm value of the local time series response of historical sewage flow-dosing flow;

[0029] Taking the sequence of energy propagation attenuation coefficient values ​​of the local time series response of sewage flow-dosing flow as the weight sequence, calculating the weighted sum of all other local time series response representation vectors of sewage flow-dosing flow in the sequence of local time series response representation vectors of sewage flow-dosing flow to obtain the energy attenuation time series aggregation feature vector of the historical local time series response of sewage flow-dosing flow;

[0030] The current sewage flow-dosing flow local time series response energy statistical paradigm value is integrated to calculate the weighted sum of the historical sewage flow-dosing flow local time series response energy attenuation time series aggregation feature vector and the current sewage flow-dosing flow local time series response representation vector to obtain the sewage flow-dosing flow full time series response aggregation representation vector.

[0031] For example, according to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, based on the maximum value, average value and variance of each sewage flow-dosing flow local time series response representation vector in the sequence of the sewage flow-dosing flow local time series response representation vector, the energy statistical paradigm value of each sewage flow-dosing flow local time series response representation vector is calculated to obtain a sequence of sewage flow-dosing flow local time series response energy statistical paradigm values, including:

[0032] Extracting the maximum value of the sewage flow-dosage flow local time series response representation vector to obtain the maximum value of the sewage flow-dosage flow local time series response;

[0033] Respectively calculating the average value and variance of the sewage flow-dosage flow local time series response representation vector to obtain the average value of the sewage flow-dosage flow local time series response and the variance of the sewage flow-dosage flow local time series response;

[0034] The value obtained by adding the variance of the local time series response of the sewage flow-dosage flow and the preset hyperparameter is multiplied by four to obtain the first energy statistical paradigm value of the local time series response of the sewage flow-dosage flow;

[0035] Calculate the square of the difference between the maximum value of the local time series response of the sewage flow-dosing flow and the average value of the local time series response of the sewage flow-dosing flow to obtain the difference value of the local time series response of the sewage flow-dosing flow;

[0036] The second energy statistical paradigm value of the sewage flow-dosage flow local time series response is obtained by multiplying the sewage flow-dosage flow local time series response variance by two, adding the modulated sewage flow-dosage flow local time series response variance obtained by multiplying the preset hyperparameter by two and the sewage flow-dosage flow local time series response difference value;

[0037] The first energy statistical paradigm value of the local time series response of sewage flow-dosing flow is divided by the second energy statistical paradigm value of the local time series response of sewage flow-dosing flow to obtain the energy statistical paradigm value of the local time series response of sewage flow-dosing flow.

[0038] For example, according to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, based on the sequence of the sewage flow-dosing flow local time series response propagation space span values ​​and the sequence of the historical sewage flow-dosing flow local time series response energy statistical paradigm values, the energy propagation attenuation coefficient values ​​of other sewage flow-dosing flow local time series response representation vectors in the sequence of the sewage flow-dosing flow local time series response representation vectors are determined to obtain the sequence of sewage flow-dosing flow local time series response energy propagation attenuation coefficient values, including:

[0039] Multiplying each sewage flow-dosage flow local time series response propagation space span value in the sequence of sewage flow-dosage flow local time series response propagation space span values ​​by a first weight hyperparameter to obtain a sequence of first sewage flow-dosage flow weighted propagation space span values;

[0040] Using each sewage flow-dosage flow local time series response propagation space span value in the sequence of sewage flow-dosage flow local time series response propagation space span value as an exponent of a natural constant to calculate an exponential function value with e as the base according to position to obtain a sequence of sewage flow-dosage flow local time series response propagation class support space span values;

[0041] Multiply each sewage flow-dosage flow local time series response propagation class support space span value in the sequence of sewage flow-dosage flow local time series response propagation class support space span value by a second weight hyperparameter to obtain a sequence of second sewage flow-dosage flow weighted propagation space span values;

[0042] Calculate the positional sum of the first sewage flow-dosage flow weighted propagation space span value sequence and the second sewage flow-dosage flow weighted propagation space span value sequence to obtain a sewage flow-dosage flow weighted sum propagation space span value sequence;

[0043] Divide the sequence of the historical sewage flow-injection flow local time series response energy statistical paradigm values ​​and the sequence of the sewage flow-injection flow weighted sum propagation space span values ​​corresponding to each group of historical sewage flow-injection flow local time series response energy statistical paradigm values ​​and the sewage flow-injection flow weighted sum propagation space span values ​​to obtain the sequence of the sewage flow-injection flow local time series response energy propagation attenuation coefficient values.

[0044] For example, according to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, the control instruction is obtained based on the sewage flow-dosing flow full time series response aggregation representation characteristics, including:

[0045] The sewage flow-dosage flow full time series response aggregation representation vector is input into a classifier-based parameter controller to obtain the control instruction.

[0046] According to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, it is possible to achieve more accurate addition control by real-time monitoring of the sewage inflow flow and the VFAs organic acid addition flow, thereby improving the sewage treatment efficiency and realizing the intelligent control of the kitchen waste VFAs sewage treatment system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0048] Figure 1 A schematic diagram of the application architecture of a control method for a kitchen waste VFAs sewage treatment system according to an embodiment of the present application is shown;

[0049] Figure 2 A flow chart showing a method for controlling a wastewater treatment system for kitchen waste VFAs in an embodiment of the present application is shown;

[0050] Figure 3 A flowchart of sub-step S540 of the method for controlling the kitchen waste VFAs sewage treatment system in an embodiment of the present application is shown;

[0051] Figure 4 A flowchart showing sub-step S550 of the method for controlling the kitchen waste VFAs sewage treatment system in an embodiment of the present application; and

[0052] Figure 5 A schematic diagram of the structure of a control system for a wastewater treatment system for kitchen waste VFAs in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0054] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0055] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0056] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0057] Figure 1 A schematic diagram of the application architecture of a control method for a sewage treatment system for kitchen waste VFAs in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0058] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0059] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0060] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the program of the control method of the sewage treatment system for food waste VFAs in the embodiment of the present application.

[0061] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any method for controlling a wastewater treatment system for kitchen waste VFAs in the embodiments of the present application according to the obtained program instructions.

[0062] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0063] The following is a non-restrictive description of the control method for the sewage treatment system for kitchen waste VFAs provided in accordance with at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0064] In response to the above technical problems, the technical concept of the present application is to collect the time series of the sewage inflow flow rate by the first flow meter, and collect the time series of the VFAs organic acid addition flow rate by the second flow meter, and use data analysis and processing technology based on deep learning to extract and associate the local time series features of the sewage inflow flow rate and the VFAs organic acid addition flow rate, so as to adaptively adjust the VFAs organic acid addition flow rate at the current time point according to the aggregate response representation characteristics of the sewage inflow flow rate and the VFAs organic acid addition flow rate in the full time domain. In this way, by real-time monitoring of the sewage inflow flow rate and the VFAs organic acid addition flow rate, more accurate addition control can be achieved to avoid excessive or insufficient addition, while reducing the impact of the sewage treatment system caused by flow fluctuations, thereby improving the sewage treatment efficiency and realizing the intelligent control of the sewage treatment system for kitchen waste VFAs.

[0065] Figure 2 The flowchart of the control method of the kitchen waste VFAs sewage treatment system in the embodiment of the present application is shown. For example, the control method of the kitchen waste VFAs sewage treatment system can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 2As shown, according to the control method of the kitchen waste VFAs sewage treatment system of the embodiment of the present application, the steps include: S510, obtaining the time series of the sewage inflow flow collected by the first flow meter; S520, obtaining the time series of the VFAs organic acid addition flow collected by the second flow meter; S530, inputting the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow into the sequence encoder respectively to obtain the sequence of the local time series correlation feature vector of the sewage inflow flow and the sequence of the local time series correlation feature vector of the VFAs organic acid addition flow; S540, each group of corresponding The local time series correlation feature vector of the sewage inflow flow and the local time series correlation feature vector of the VFAs organic acid addition flow are input into the adaptive feature response analyzer to obtain a sequence of sewage flow-addition flow local time series response representation vectors; S550, the sequence of sewage flow-addition flow local time series response representation vectors is input into the node feature propagation network based on the node energy decay mechanism to obtain the sewage flow-addition flow full time series response aggregation representation vector as the sewage flow-addition flow full time series response aggregation representation feature; S560, based on the sewage flow-addition flow full time series response aggregation representation feature, a control instruction is obtained, and the control instruction is used to indicate whether the VFAs organic acid addition flow at the current time point should increase, decrease or remain unchanged.

[0066] Specifically, in the technical solution of the present application, first, the time series of the sewage inflow flow collected by the first flow meter is obtained, and the time series of the VFAs organic acid addition flow collected by the second flow meter is obtained. Next, it is considered that the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow have local temporal correlation characteristics in local time, such as local increase of sewage flow. Based on this, in the technical solution of the present application, the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow are processed using a sequence encoder of a 1D-CNN model that is particularly good at extracting local pattern features from one-dimensional sequence data to respectively capture and extract the temporal correlation relationship between the sewage inflow flow and the VFAs organic acid addition flow in a local time period, thereby obtaining a sequence of local temporal correlation feature vectors of the sewage inflow flow and a sequence of local temporal correlation feature vectors of the VFAs organic acid addition flow. In particular, the sequence encoder of the 1D-CNN model here is a one-dimensional convolutional neural network model.

[0067] Correspondingly, in step S530, the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow are respectively input into the sequence encoder to obtain a sequence of local time series correlation feature vectors of the sewage inflow flow and a sequence of local time series correlation feature vectors of the VFAs organic acid addition flow, including: inputting the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow into the sequence encoder based on the 1D-CNN model respectively to obtain a sequence of local time series correlation feature vectors of the sewage inflow flow and a sequence of local time series correlation feature vectors of the VFAs organic acid addition flow.

[0068] Then, considering that there is a mutual response and interactive relationship in local time series between the local time series correlation characteristics of the sewage inflow flow and the local time series correlation characteristics of the VFAs organic acid addition flow, therefore, in order to more accurately and clearly understand the interactive correlation relationship in each local time period, in the technical solution of the present application, each group of corresponding local time series correlation feature vectors of sewage inflow flow and local time series correlation feature vectors of VFAs organic acid addition flow in the sequence of local time series correlation feature vectors of the sewage inflow flow and the sequence of local time series correlation feature vectors of the VFAs organic acid addition flow is input into the adaptive feature response analyzer to obtain a sequence of sewage flow-addition flow local time series response representation vectors. It is worth mentioning that the adaptive feature response analyzer considers the ability to capture key local time series information by performing responsiveness calculations on the elements at each position in the local time series correlation feature vectors of the sewage inflow flow and the local time series correlation feature vectors of the VFAs organic acid addition flow. Then, the features obtained by responsiveness calculation are adaptively weighted to dynamically adjust the feature weights according to the real-time interactive response data between the two, ensuring that the model can adapt to the key significant information between the local temporal changes of sewage inflow flow and the adjustment of VFAs addition flow, thereby significantly improving the expression ability and overall performance of the temporal interaction between the two and achieving more effective sewage treatment process control.

[0069] Accordingly, in step S540, if Figure 3As shown, each group of corresponding local time series correlation feature vectors of sewage inflow flow and local time series correlation feature vectors of VFAs organic acid addition flow in the sequence of local time series correlation feature vectors of sewage inflow flow and the sequence of local time series correlation feature vectors of VFAs organic acid addition flow is input into an adaptive characteristic response analyzer to obtain a sequence of sewage flow-addition flow local time series response representation vectors, including: S541, calculating the position-by-position response between the local time series correlation feature vectors of sewage inflow flow and the local time series correlation feature vectors of VFAs organic acid addition flow to obtain a sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector; S542, using a Softmax function to normalize the sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector to obtain a normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector. response characteristic vector; S543, inputting the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response characteristic vector into a learnable gating function to obtain a sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector; S544, calculating the position point multiplication between the sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector and the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response characteristic vector to obtain a sewage inflow flow-VFAs organic acid addition flow position-by-position distinguishable weight mask vector; S545, calculating the position point multiplication between the sewage inflow flow-VFAs organic acid addition flow position-by-position distinguishable weight mask vector and the sewage inflow flow-VFAs organic acid addition flow position-by-position response characteristic vector to obtain the sewage flow-addition flow local time series response representation vector.

[0070] Among them, in step S543, the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector is input into a learnable gating function to obtain a sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector, including: using the negative number of the characteristic value of each position in the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector as the exponent of the natural constant to calculate the exponential function value with base e according to the position to obtain the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response class support vector; calculating the reciprocal of the sum of the characteristic value of each position in the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response class support vector and the constant one to obtain the sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector.

[0071] In a specific example, each group of corresponding local time series correlation feature vectors of sewage inflow flow and local time series correlation feature vectors of VFAs organic acid addition flow in the sequence of local time series correlation feature vectors of sewage inflow flow and the sequence of local time series correlation feature vectors of VFAs organic acid addition flow is input into an adaptive characteristic response analyzer to obtain a sequence of local time series response representation vectors of sewage flow-addition flow, including: inputting the local time series correlation feature vectors of sewage inflow flow and the local time series correlation feature vectors of VFAs organic acid addition flow into the adaptive characteristic response analyzer, and processing them with the following response formula to obtain the local time series response representation vector of sewage flow-addition flow; wherein the response formula is:

[0072] v r =v 2 / v 1

[0073] v n =softmax(v r )

[0074]

[0075] v r =v r ⊙(v n ⊙v a )

[0076] Among them, v 2 and v 1 are the local time series correlation characteristic vector of the sewage inflow flow and the local time series correlation characteristic vector of the VFAs organic acid addition flow, v r represents the sewage inflow flow-VFAs organic acid addition flow rate response feature vector, sofmtax(·) is the softmax function, v n is the position-by-position response characteristic vector of the normalized sewage inflow flow-VFAs organic acid addition flow, exp(·) is an exponential function with the natural constant e as the base, v a is the sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector, ⊙ is the point multiplication by location, x r It is the vector representing the local time series response of sewage flow-dosage flow.

[0077] Further, considering that each sewage flow-addition flow local time series response representation vector in the sequence of the sewage flow-addition flow local time series response representation vector represents the key characteristic information of the sewage flow-addition flow time series response in a local time period, therefore, in order to effectively integrate the sewage flow-addition flow in the full time series response information to capture the dynamic changes and long-term dependencies of the entire sequence, further, in the technical solution of the present application, the sequence of the sewage flow-addition flow local time series response representation vector is input into the node feature propagation network based on the node energy decay mechanism to obtain the sewage flow-addition flow full time series response aggregation representation vector. That is to say, the node feature propagation network based on the node energy decay mechanism utilizes the node energy decay principle, monitors the energy state of the node features in real time, effectively integrates the node features of the time series data, enhances the key time series features of the model and the understanding of the dynamics of the data, and ignores unimportant information, ensuring the accuracy and efficiency of identifying and responding to key data points in the task. Specifically, firstly, the sequence of the sewage flow-addition flow local time series response representation vector is energy-measured, and its energy level is evaluated based on the maximum value, average value and variance of the vector. Next, the propagation space span between the representation vectors of the local time series response of the sewage flow-dosage flow is analyzed to reveal the connection pattern and spatial layout between nodes. In particular, these span values ​​are combined with the energy measurement of the historical nodes to determine the decay rate of energy propagation, which is inversely proportional to the energy of the historical nodes, reflecting the natural decay of energy over time and spatial distribution. Using these decay rates as weighting factors, the energy statistical paradigm values ​​are weighted and summed to form a time series aggregate feature vector of the energy decay of the historical nodes, thereby aggregating the node characteristics of the local time series of the historical sewage flow-dosage flow. Finally, combined with the energy measurement of the current node, the historical aggregate feature vector is weighted and merged with the current node characteristics to generate a sewage flow-dosage flow full time series response aggregate representation vector that combines the time series evolution and spatial network structure.

[0078] Accordingly, in step S550, if Figure 4As shown, the sequence of the sewage flow-addition flow local time series response representation vectors is input into the node feature propagation network based on the node energy decay mechanism to obtain the sewage flow-addition flow full time series response aggregation representation vector, including: S551, based on the maximum value, average value and variance of each sewage flow-addition flow local time series response representation vector in the sequence of the sewage flow-addition flow local time series response representation vector, the energy statistical paradigm value of each sewage flow-addition flow local time series response representation vector is calculated to obtain a sequence of sewage flow-addition flow local time series response energy statistical paradigm values, wherein the current value in the sequence of the sewage flow-addition flow local time series response energy statistical paradigm values ​​is The energy statistical paradigm value corresponding to the sewage flow-dosage flow local time series response representation vector is used as the current node energy statistical paradigm value and other energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of the current sewage flow-dosage flow local time series response energy statistical paradigm value and the historical sewage flow-dosage flow local time series response energy statistical paradigm value; S552, counting the node propagation space span value between each other sewage flow-dosage flow local time series response representation vector in the sequence of the sewage flow-dosage flow local time series response representation vector and the current sewage flow-dosage flow local time series response representation vector to obtain the sewage flow-dosage flow local time series response propagation space span S553, based on the sequence of the propagation spatial span values ​​of the local time series response of sewage flow-dosing flow and the sequence of the energy statistical paradigm values ​​of the local time series response of historical sewage flow-dosing flow, determine the energy propagation attenuation coefficient values ​​of other local time series response representation vectors of sewage flow-dosing flow in the sequence of local time series response representation vectors of sewage flow-dosing flow to obtain a sequence of energy propagation attenuation coefficient values ​​of the local time series response of sewage flow-dosing flow, wherein the energy propagation attenuation coefficient value of the local time series response of sewage flow-dosing flow is inversely correlated with the energy statistical paradigm value of the local time series response of historical sewage flow-dosing flow; S554, with the sewage flow-dosing flow The sequence of energy propagation attenuation coefficient values ​​of the local time series response of the added flow is used as the weight sequence, and the weighted sum of all other local time series response representation vectors of sewage flow-dosage flow in the sequence of the local time series response representation vector of sewage flow-dosage flow is calculated to obtain the historical sewage flow-dosage flow local time series response energy attenuation time series aggregation feature vector; S555, the weighted sum of the historical sewage flow-dosage flow local time series response energy attenuation time series aggregation feature vector and the current sewage flow-dosage flow local time series response representation vector are calculated by fusing the current sewage flow-dosage flow local time series response energy statistical paradigm value to obtain the sewage flow-dosage flow full time series response aggregation representation vector.

[0079] Wherein, in step S551, based on the maximum value, average value and variance of each sewage flow-dosage flow local time series response representation vector in the sequence of the sewage flow-dosage flow local time series response representation vector, the energy statistical paradigm value of each sewage flow-dosage flow local time series response representation vector is calculated to obtain a sequence of energy statistical paradigm values ​​of the sewage flow-dosage flow local time series response, including: extracting the maximum value of the sewage flow-dosage flow local time series response representation vector to obtain the maximum value of the sewage flow-dosage flow local time series response; respectively calculating the average value and variance of the sewage flow-dosage flow local time series response representation vector to obtain the average value of the sewage flow-dosage flow local time series response and the variance of the sewage flow-dosage flow local time series response; multiplying the value obtained by adding the variance of the sewage flow-dosage flow local time series response and the preset hyperparameter Four is used to obtain the first energy statistical paradigm value of the local timing response of sewage flow-dosing flow; the square of the difference between the maximum value of the local timing response of sewage flow-dosing flow and the average value of the local timing response of sewage flow-dosing flow is calculated to obtain the difference value of the local timing response of sewage flow-dosing flow; the variance of the local timing response of sewage flow-dosing flow is multiplied by two to obtain the modulated local timing response variance of sewage flow-dosing flow, the value obtained by multiplying the preset hyperparameter by two and the difference value of the local timing response of sewage flow-dosing flow are added to obtain the second energy statistical paradigm value of the local timing response of sewage flow-dosing flow; the first energy statistical paradigm value of the local timing response of sewage flow-dosing flow is divided by the second energy statistical paradigm value of the local timing response of sewage flow-dosing flow to obtain the energy statistical paradigm value of the local timing response of sewage flow-dosing flow.

[0080] Wherein, in step S553, based on the sequence of the propagation spatial span values ​​of the local time series response of sewage flow-dosage flow and the sequence of the energy statistical paradigm values ​​of the local time series response of historical sewage flow-dosage flow, the energy propagation attenuation coefficient values ​​of other local time series response representation vectors of sewage flow-dosage flow in the sequence of local time series response representation vectors of sewage flow-dosage flow are determined to obtain a sequence of energy propagation attenuation coefficient values ​​of the local time series response of sewage flow-dosage flow, including: multiplying the propagation spatial span values ​​of the local time series response of sewage flow-dosage flow in the sequence of the propagation spatial span values ​​of the local time series response of sewage flow-dosage flow by the first weight hyperparameter to obtain a sequence of first sewage flow-dosage flow weighted propagation spatial span values; using the propagation spatial span values ​​of the local time series response of sewage flow-dosage flow in the sequence of the propagation spatial span values ​​of the local time series response of sewage flow-dosage flow as the exponent of the natural constant to calculate the exponential function value with base e according to the position to obtain the sewage flow-dosage flow. The method comprises the following steps: obtaining a sequence of spatial span values ​​supported by the propagation class of the local time series response of the sewage flow-dosage flow; multiplying each spatial span value supported by the propagation class of the local time series response of the sewage flow-dosage flow by the second weight hyperparameter to obtain a sequence of second sewage flow-dosage flow weighted propagation spatial span values; calculating the positional sum of the sequence of the first sewage flow-dosage flow weighted propagation spatial span values ​​and the sequence of the second sewage flow-dosage flow weighted propagation spatial span values ​​to obtain a sequence of sewage flow-dosage flow weighted sum propagation spatial span values; dividing the sequence of the local time series response energy statistical paradigm values ​​of the historical sewage flow-dosage flow and the sequence of the weighted sum propagation spatial span values ​​of the sewage flow-dosage flow by each corresponding group of historical sewage flow-dosage flow local time series response energy statistical paradigm values ​​and the sewage flow-dosage flow weighted sum propagation spatial span values ​​to obtain a sequence of sewage flow-dosage flow local time series response energy propagation attenuation coefficient values.

[0081] In a specific example, the sequence of the sewage flow-dosage flow local time series response representation vector is input into the node feature propagation network based on the node energy decay mechanism to obtain the sewage flow-dosage flow full time series response aggregation representation vector, including: inputting the sequence of the sewage flow-dosage flow local time series response representation vector into the node feature propagation network based on the node energy decay mechanism, and processing it with the following propagation formula to obtain the sewage flow-dosage flow full time series response aggregation representation vector; wherein the propagation formula is:

[0082] X={x 1 ,x 2 ,...x k-1 ,x k}

[0083]

[0084] Wherein, X is the sequence of vectors representing the local time series response of the sewage flow-dosage flow, x k-1 and x k are the k-1th and kth sewage flow-dosing flow local time series response representation vectors in the sequence of the sewage flow-dosing flow local time series response representation vectors, k represents the number of characteristic vectors in the sequence of the sewage flow-dosing flow local time series response representation vectors, μ i and σ 2 i are the mean and variance of the i-th sewage flow-dosage flow local time series response representation vector in the sequence of the sewage flow-dosage flow local time series response representation vector, max(x i ) is the maximum value of the local time series response representation vector of the i-th sewage flow-dosage flow, ∈ is the preset hyperparameter, is the energy statistical paradigm value of the i-th sewage flow-dosage flow local time series response representation vector, is the eigenvalue of the sth position in the i-th sewage flow-dosage flow local time series response representation vector, L is the number of eigenvalues ​​in each sewage flow-dosage flow local time series response representation vector, and are the energy statistical norm values ​​of the kth and ith sewage flow-dosage flow local time series response representation vectors, Count(x k →x i ) represents x k With x i The number of feature vectors between them, α, β, γ and δ are weight hyperparameters, and x c It is the aggregate representation vector of the full time series response of the sewage flow-injection flow.

[0085] Then, the sewage flow-addition flow full time series response aggregation representation vector is input into the parameter controller based on the classifier to obtain a control instruction, and the control instruction is used to indicate whether the VFAs organic acid addition flow at the current time point should be increased, decreased, or kept unchanged. That is, the sewage flow-addition flow full time series response aggregation representation feature obtained by full time domain propagation of the sequence of the sewage flow-addition flow local time series response representation vector is classified and processed, so as to adaptively adjust the VFAs organic acid addition flow at the current time point. In this way, by real-time monitoring of the sewage inflow flow and the VFAs organic acid addition flow, more accurate addition control can be achieved to avoid excessive or insufficient addition, while reducing the impact of the sewage treatment system caused by flow fluctuations, thereby improving the sewage treatment efficiency and realizing the intelligent control of the sewage treatment system for kitchen waste VFAs.

[0086] Accordingly, in step S560, a control instruction is obtained based on the aggregate representation characteristics of the full time series response of the sewage flow-dosage flow, including: inputting the aggregate representation vector of the full time series response of the sewage flow-dosage flow into a classifier-based parameter controller to obtain the control instruction.

[0087] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification. However, this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.

[0088] The present application takes into account that the sequence of local time series correlation feature vectors of sewage inflow flow and the sequence of local time series correlation feature vectors of VFAs organic acid addition flow respectively express the local time series correlation characteristics of sewage inflow flow and VFAs organic acid addition flow. In this way, when an adaptive feature response based on local time domain distribution is performed on them and feature propagation aggregation is performed through time node energy attenuation, the obtained sewage flow-addition flow full time series response aggregation representation vector will also cause insufficient feature propagation aggregation coverage due to the energy difference of the adaptive response nodes of the local time domain feature distribution, thereby leading to outlier class reasoning mapping deviation, affecting the accuracy of the control instructions obtained by the sewage flow-addition flow full time series response aggregation representation vector input into the classifier-based parameter controller.

[0089] Based on this, in a preferred example, the sewage flow-dosage flow full time series response aggregate representation vector is input into a classifier-based parameter controller to obtain the control instruction, including: calculating the characteristic mean of the sewage flow-dosage flow full time series response aggregate representation vector, and dividing the characteristic mean by the difference between the maximum eigenvalue and the minimum eigenvalue of the sewage flow-dosage flow full time series response aggregate representation vector to obtain the sewage flow-dosage flow full time series response aggregate representation distribution representation value; subtracting the sewage flow-dosage flow full time series response aggregate representation distribution representation value and dividing it by the sewage flow-dosage flow full time series response aggregate representation distribution representation value to obtain the sewage flow-dosage flow full time series response aggregate representation distribution modulation value; activating the sewage flow-dosage flow full time series response aggregate representation vector through a probabilistic function to obtain a probabilistic sewage flow-dosage flow full time series response aggregate representation vector; performing point subtraction between the probabilistic sewage flow-dosage flow full time series response aggregate representation vector and the sewage flow-dosage flow full time series response aggregate representation distribution modulation value, and taking the absolute value. And calculate the negative of the logarithmic value with base 2 to obtain the probabilistic sewage flow-dosage flow full time series response aggregation representation distribution modulation information vector; divide the sewage flow-dosage flow full time series response aggregation representation distribution characterization value by one minus the difference of each eigenvalue of the probabilistic sewage flow-dosage flow full time series response aggregation representation vector, sum all eigenvalues ​​of the probabilistic sewage flow-dosage flow full time series response aggregation representation vector, and divide it by the length of the sewage flow-dosage flow full time series response aggregation representation vector to obtain the probabilistic sewage flow-dosage flow full time series response aggregation representation distribution modulation bias value; perform dot addition on the probabilistic sewage flow-dosage flow full time series response aggregation representation distribution modulation information vector and the probabilistic sewage flow-dosage flow full time series response aggregation representation distribution modulation bias value and the weight as a hyperparameter to obtain the optimized sewage flow-dosage flow full time series response aggregation representation vector; input the optimized sewage flow-dosage flow full time series response aggregation representation vector into the classifier-based parameter controller to obtain the control instruction to obtain the control result.

[0090] The calculation process of the optimized sewage flow-dosage flow full time series response aggregation representation vector is expressed by the following optimization formula:

[0091]

[0092] Among them, V is the probabilistic sewage flow-dosage flow full time series response aggregation representation vector, p is the sewage flow-dosage flow full time series response aggregation representation distribution representation value, is subtraction by location point, log represents the logarithmic function value with base 2, ⊕ is addition by location point, ε is the weight as a hyperparameter, L is the length of the aggregate representation vector of the full time series response of the sewage flow-dosage flow, and v i is the characteristic value of each position in the probabilistic sewage flow-dosage flow full time series response aggregation representation vector, is the characteristic mean of the aggregate representation vector of the full time series response of the sewage flow-dosage flow, v max and v min are respectively the maximum eigenvalue and the minimum eigenvalue of the aggregate representation vector of the full time series response of the sewage flow-dosage flow, and V' is the aggregate representation vector of the full time series response of the optimized sewage flow-dosage flow.

[0093] That is, in the above preferred example, the probability information distribution planning of the sewage flow-dosage flow full time-series response aggregate representation vector based on the eigenvalue is carried out through the Bernoulli probability modulation distribution of the sewage flow-dosage flow full time-series response aggregate representation vector relative to the eigenvalue distribution, and the probability reverse mapping of the probability characteristics of the sewage flow-dosage flow full time-series response aggregate representation vector as a whole is used as the extended coverage of the set mapping space of the sewage flow-dosage flow full time-series response aggregate representation vector, so as to independently understand the interaction path between the intuitive probability information distribution and the abstract probability space mapping of the sewage flow-dosage flow full time-series response aggregate representation vector, so as to improve the accuracy of the control instructions obtained by the optimized sewage flow-dosage flow full time-series response aggregate representation vector input based on the classifier parameter controller by avoiding the counterfactual reasoning mapping of the outlier feature distribution of the sewage flow-dosage flow full time-series response aggregate representation vector to the class probability. In this way, by real-time monitoring of the sewage inflow flow and the VFAs organic acid addition flow, more precise addition control can be achieved to avoid excessive or insufficient addition, while reducing the impact of flow fluctuations on the sewage treatment system, thereby improving sewage treatment efficiency and realizing the intelligent control of the sewage treatment system for kitchen waste VFAs.

[0094] Furthermore, it is worth mentioning that the main effective components of VFAs prepared from kitchen waste are volatile organic compounds with 1-6 carbon atoms, such as ethanol, acetic acid, propionic acid, butyric acid, etc. After complete hydrolysis and acidification, the COD equivalent of kitchen waste acidified liquid can reach 150,000-250,000 mg / L, but the suspended solids (SS) concentration is also nearly 50,000 mg / L. This acidified liquid cannot meet the requirements of being directly added to the denitrification tank as a substitute for traditional fossil carbon sources. Therefore, further purification and concentration are required.

[0095] VFAs prepared from kitchen waste are directly added to the denitrification tank as a substitute for traditional fossil carbon sources. They should not only play the properties and functions of carbon sources, but also should not have a negative impact on other operating parameters of the sewage treatment system. Therefore, the acidified liquid after complete hydrolysis and acidification is purified and concentrated by multi-stage filtration to obtain a stable product of VFAs organic acid with SS <100 mg / L, COD = 50,000-80,000 mg / L, and B / C = 0.7. As a substitute for traditional fossil carbon sources, this product can be directly added to the water inlet of the denitrification tank with reference to traditional fossil carbon sources to replace traditional fossil carbon sources and play the role of carbon source. According to long-term actual addition feedback, when VFAs organic acids are used as substitutes for traditional fossil carbon sources, the carbon source properties of VFAs can play an efficient role when the ratio of their addition amount to the water inlet flow of the denitrification tank of the municipal sewage treatment system is about 1:1000. When the addition ratio of VFAs organic acids is greater than 1:1000, it is easy to cause excess carbon source, and the remaining carbon source will become a nutrient source for microorganisms in other treatment links, resulting in increased sludge production and increased operating costs; severe excess will cause an increase in COD treatment load, affecting the concentration of effluent water quality indicators. When the addition ratio of VFAs organic acids is less than 1:1000, the added carbon source is insufficient, and the denitrification reaction is incomplete, resulting in low denitrification efficiency of the entire municipal sewage treatment system, and the inability to give full play to the effect of the external carbon source.

[0096] The addition ratio of VFAs organic acids prepared from kitchen waste as a substitute for traditional fossil carbon sources in the municipal sewage system is 1:1000, which is close to the actual ratio of kitchen waste generation to domestic sewage generation (1:1118). This shows that within the urban jurisdiction, VFAs organic acids prepared from kitchen waste can fully meet the demand of urban domestic sewage treatment systems for external supplementary carbon sources, and promote the high-quality construction and development of "zero-waste cities".

[0097] Furthermore, in order to ensure that VFAs can be directly added to the denitrification tank, the SS of VFAs produced from kitchen waste should be less than 100 mg / L. Therefore, the COD equivalent of VFAs produced from kitchen waste is 5×10 4 ~8×10 4 mg / L.

[0098] Further, based on the above embodiments, refer to Figure 5As shown, it is a structural schematic diagram of a control system 800 for a wastewater treatment system for kitchen waste VFAs in an embodiment of the present application. The control system 800 for the wastewater treatment system for kitchen waste VFAs includes: a first flow acquisition module 810, used to obtain the time series of the sewage inflow flow collected by the first flow meter; a second flow acquisition module 820, used to obtain the time series of the VFAs organic acid addition flow collected by the second flow meter; a sequence encoding module 830, used to input the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow into a sequence encoder respectively to obtain a sequence of local time series correlation feature vectors of the sewage inflow flow and a sequence of local time series correlation feature vectors of the VFAs organic acid addition flow; an adaptive characteristic response analysis module 840, used to input the sequence of local time series correlation feature vectors of the sewage inflow flow and the sequence of local time series correlation feature vectors of the VFAs organic acid addition flow Each group of corresponding local time series correlation feature vectors of sewage inflow flow and local time series correlation feature vectors of VFAs organic acid addition flow in the column is input into the adaptive feature response analyzer to obtain a sequence of local time series response representation vectors of sewage flow-addition flow; a full time series response aggregation module 850 is used to input the sequence of local time series response representation vectors of sewage flow-addition flow into the node feature propagation network based on the node energy decay mechanism to obtain the full time series response aggregation representation vector of sewage flow-addition flow as the full time series response aggregation representation feature of sewage flow-addition flow; a control module 860 is used to obtain a control instruction based on the full time series response aggregation representation feature of sewage flow-addition flow, and the control instruction is used to indicate whether the VFAs organic acid addition flow at the current time point should increase, decrease or remain unchanged.

[0099] In one example, in the control system 800 of the above-mentioned kitchen waste VFAs sewage treatment system, the sequence encoding module 830 is used to: input the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow into a sequence encoder based on the 1D-CNN model respectively to obtain a sequence of local time series correlation feature vectors of the sewage inflow flow and a sequence of local time series correlation feature vectors of the VFAs organic acid addition flow.

[0100] Here, those skilled in the art will appreciate that the specific functions and operations of each module in the control system 800 of the food waste VFAs sewage treatment system have been described in the above reference. Figures 2 to 4 The control method of food waste VFAs using a sewage treatment system has been described in detail, and therefore, its repeated description will be omitted.

[0101] Based on the above embodiments, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of the control method of the kitchen waste VFAs sewage treatment system in the above embodiments can be implemented when the processor executes the program.

[0102] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0103] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by the processor, the control method for the kitchen waste VFAs sewage treatment system according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0104] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the control method of the kitchen waste VFAs sewage treatment system according to the embodiment of the present application.

[0105] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0106] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0107] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0108] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0109] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A method for controlling a wastewater treatment system for kitchen waste VFAs, characterized in that: include: Acquire a time series of sewage inflow flow collected by a first flow meter; Obtaining a time series of VFAs organic acid addition flow rate collected by a second flow meter; Input the time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow into a sequence encoder respectively to obtain a sequence of local time series correlation feature vectors of the sewage inflow flow and a sequence of local time series correlation feature vectors of the VFAs organic acid addition flow; Input each corresponding group of local time series correlation feature vectors of sewage inflow flow and local time series correlation feature vectors of VFAs organic acid addition flow in the sequence of local time series correlation feature vectors of sewage inflow flow and the sequence of local time series correlation feature vectors of VFAs organic acid addition flow into an adaptive characteristic response analyzer to obtain a sequence of sewage flow-addition flow local time series response representation vectors; Inputting the sequence of the sewage flow-dosage flow local time series response representation vector into the node feature propagation network based on the node energy decay mechanism to obtain the sewage flow-dosage flow full time series response aggregation representation vector as the sewage flow-dosage flow full time series response aggregation representation feature; Based on the sewage flow-dosage flow full time series response aggregation representation characteristics, a control instruction is obtained, and the control instruction is used to indicate whether the VFAs organic acid dosing flow at the current time point should increase, decrease, or remain unchanged; The sequence of the sewage flow-dosage flow local time series response representation vector is input into the node feature propagation network based on the node energy decay mechanism to obtain the sewage flow-dosage flow full time series response aggregation representation vector, including: Based on the maximum value, average value and variance of each sewage flow-dosage flow local time series response representation vector in the sequence of sewage flow-dosage flow local time series response representation vector, the energy statistical paradigm value of each sewage flow-dosage flow local time series response representation vector is calculated to obtain a sequence of sewage flow-dosage flow local time series response energy statistical paradigm values, wherein the energy statistical paradigm value corresponding to the current sewage flow-dosage flow local time series response representation vector in the sequence of sewage flow-dosage flow local time series response energy statistical paradigm values ​​is used as the current node energy statistical paradigm value and other energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of current sewage flow-dosage flow local time series response energy statistical paradigm values ​​and historical sewage flow-dosage flow local time series response energy statistical paradigm values; Counting the node propagation space span values ​​between each other sewage flow-dosage flow local time series response representation vector in the sequence of sewage flow-dosage flow local time series response representation vector and the current sewage flow-dosage flow local time series response representation vector to obtain a sequence of sewage flow-dosage flow local time series response propagation space span values; Based on the sequence of the propagation spatial span values ​​of the local time series response of sewage flow-dosing flow and the sequence of the energy statistical paradigm values ​​of the local time series response of historical sewage flow-dosing flow, determine the energy propagation attenuation coefficient values ​​of other local time series response representation vectors of sewage flow-dosing flow in the sequence of local time series response representation vectors of sewage flow-dosing flow to obtain a sequence of energy propagation attenuation coefficient values ​​of the local time series response of sewage flow-dosing flow, wherein the energy propagation attenuation coefficient value of the local time series response of sewage flow-dosing flow is inversely correlated with the energy statistical paradigm value of the local time series response of historical sewage flow-dosing flow; Taking the sequence of energy propagation attenuation coefficient values ​​of the local time series response of sewage flow-dosing flow as the weight sequence, calculating the weighted sum of all other local time series response representation vectors of sewage flow-dosing flow in the sequence of local time series response representation vectors of sewage flow-dosing flow to obtain the energy attenuation time series aggregation feature vector of the historical local time series response of sewage flow-dosing flow; The current sewage flow-dosing flow local time series response energy statistical paradigm value is integrated to calculate the weighted sum of the historical sewage flow-dosing flow local time series response energy attenuation time series aggregation feature vector and the current sewage flow-dosing flow local time series response representation vector to obtain the sewage flow-dosing flow full time series response aggregation representation vector.

2. The control method of the kitchen waste VFAs sewage treatment system according to claim 1, characterized in that: The time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow are respectively input into a sequence encoder to obtain a sequence of local time series correlation feature vectors of the sewage inflow flow and a sequence of local time series correlation feature vectors of the VFAs organic acid addition flow, including: The time series of the sewage inflow flow and the time series of the VFAs organic acid addition flow are respectively input into the sequence encoder based on the 1D-CNN model to obtain the sequence of the local time series correlation feature vectors of the sewage inflow flow and the sequence of the local time series correlation feature vectors of the VFAs organic acid addition flow.

3. The control method of the kitchen waste VFAs sewage treatment system according to claim 2, characterized in that: Input each corresponding group of local time series correlation feature vectors of sewage inflow flow and local time series correlation feature vectors of VFAs organic acid addition flow in the sequence of local time series correlation feature vectors of sewage inflow flow and the sequence of local time series correlation feature vectors of VFAs organic acid addition flow into an adaptive characteristic response analyzer to obtain a sequence of sewage flow-addition flow local time series response representation vectors, including: Calculate the position-by-position response between the local time series correlation characteristic vector of the sewage inflow flow and the local time series correlation characteristic vector of the VFAs organic acid addition flow to obtain the position-by-position response characteristic vector of the sewage inflow flow-VFAs organic acid addition flow; Using a Softmax function to normalize the sewage inflow flow-VFAs organic acid addition flow position-by-position response characteristic vector to obtain a normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response characteristic vector; Inputting the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector into a learnable gating function to obtain a sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector; Calculate the sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector and the normalized sewage inflow flow-VFAs organic acid addition flow location response feature vector to obtain a sewage inflow flow-VFAs organic acid addition flow location response distinguishable weight mask vector; Calculate the position-by-position response of the sewage inflow flow-VFAs organic acid addition flow, and multiply the distinguishable weight mask vector by the position-by-position response feature vector of the sewage inflow flow-VFAs organic acid addition flow to obtain the sewage flow-addition flow local time series response representation vector.

4. The control method of the kitchen waste VFAs sewage treatment system according to claim 3, characterized in that: The normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response feature vector is input into a learnable gating function to obtain a sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector, including: The negative number of each position characteristic value in the position-by-position response characteristic vector of the normalized sewage inflow flow-VFAs organic acid addition flow is used as the exponent of the natural constant to calculate the exponential function value with e as the base according to the position to obtain the normalized sewage inflow flow-VFAs organic acid addition flow position-by-position response class support vector; The reciprocal of the sum of the eigenvalues ​​of each position in the position-by-position response class support vector of the normalized sewage inflow flow-VFAs organic acid addition flow and the constant one is calculated to obtain the sewage inflow flow-VFAs organic acid addition flow response screening weight mask vector.

5. The control method of the kitchen waste VFAs sewage treatment system according to claim 4, characterized in that: Based on the maximum value, average value and variance of each sewage flow-dosage flow local time series response representation vector in the sequence of the sewage flow-dosage flow local time series response representation vector, the energy statistical paradigm value of each sewage flow-dosage flow local time series response representation vector is calculated to obtain a sequence of sewage flow-dosage flow local time series response energy statistical paradigm values, including: Extracting the maximum value of the sewage flow-dosage flow local time series response representation vector to obtain the maximum value of the sewage flow-dosage flow local time series response; Respectively calculating the average value and variance of the sewage flow-dosage flow local time series response representation vector to obtain the average value of the sewage flow-dosage flow local time series response and the variance of the sewage flow-dosage flow local time series response; The value obtained by adding the variance of the local time series response of the sewage flow-dosage flow and the preset hyperparameter is multiplied by four to obtain the first energy statistical paradigm value of the local time series response of the sewage flow-dosage flow; Calculate the square of the difference between the maximum value of the local time series response of the sewage flow-dosing flow and the average value of the local time series response of the sewage flow-dosing flow to obtain the difference value of the local time series response of the sewage flow-dosing flow; The second energy statistical paradigm value of the sewage flow-dosage flow local time series response is obtained by multiplying the sewage flow-dosage flow local time series response variance by two, adding the modulated sewage flow-dosage flow local time series response variance obtained by multiplying the preset hyperparameter by two and the sewage flow-dosage flow local time series response difference value; The first energy statistical paradigm value of the local time series response of sewage flow-dosing flow is divided by the second energy statistical paradigm value of the local time series response of sewage flow-dosing flow to obtain the energy statistical paradigm value of the local time series response of sewage flow-dosing flow.

6. The control method of the kitchen waste VFAs sewage treatment system according to claim 5, characterized in that: Based on the sequence of the propagation spatial span values ​​of the local time series response of the sewage flow-dosing flow and the sequence of the energy statistical paradigm values ​​of the local time series response of the historical sewage flow-dosing flow, the energy propagation attenuation coefficient values ​​of other local time series response representation vectors of the sewage flow-dosing flow in the sequence of the local time series response representation vectors of the sewage flow-dosing flow are determined to obtain a sequence of energy propagation attenuation coefficient values ​​of the local time series response of the sewage flow-dosing flow, including: Multiplying each sewage flow-dosage flow local time series response propagation space span value in the sequence of sewage flow-dosage flow local time series response propagation space span values ​​by a first weight hyperparameter to obtain a sequence of first sewage flow-dosage flow weighted propagation space span values; Using each sewage flow-dosage flow local time series response propagation space span value in the sequence of sewage flow-dosage flow local time series response propagation space span value as an exponent of a natural constant to calculate an exponential function value with e as the base according to position to obtain a sequence of sewage flow-dosage flow local time series response propagation class support space span values; Multiply each sewage flow-dosage flow local time series response propagation class support space span value in the sequence of sewage flow-dosage flow local time series response propagation class support space span value by a second weight hyperparameter to obtain a sequence of second sewage flow-dosage flow weighted propagation space span values; Calculate the positional sum of the first sewage flow-dosage flow weighted propagation space span value sequence and the second sewage flow-dosage flow weighted propagation space span value sequence to obtain a sewage flow-dosage flow weighted sum propagation space span value sequence; Divide the sequence of the historical sewage flow-injection flow local time series response energy statistical paradigm values ​​and the sequence of the sewage flow-injection flow weighted sum propagation space span values ​​corresponding to each group of historical sewage flow-injection flow local time series response energy statistical paradigm values ​​and the sewage flow-injection flow weighted sum propagation space span values ​​to obtain the sequence of the sewage flow-injection flow local time series response energy propagation attenuation coefficient values.

7. The control method of the kitchen waste VFAs sewage treatment system according to claim 6, characterized in that: Based on the sewage flow-dosing flow full time series response aggregation representation characteristics, a control instruction is obtained, including: The sewage flow-dosage flow full time series response aggregation representation vector is input into a classifier-based parameter controller to obtain the control instruction.

Images