Method and device for detecting grease content in data-driven sewer network

Abstract

This application relates to the field of environmental monitoring technology and provides a data-driven method and apparatus for detecting grease content in drainage pipe networks. The method includes: acquiring various indicator data of the drainage pipe network under test at a first moment, including water quality indicator data, water quantity indicator data, and air quality indicator data of the pipe network; inputting the indicator data into a preset grease content prediction model to obtain the grease content of the wastewater in the drainage pipe network under test at the first moment. This method is simple to implement and can quickly obtain the detection results of grease content.

Description

Technical Field

[0001] This application belongs to the field of environmental testing technology, and in particular relates to a data-driven method and apparatus for detecting grease content in drainage pipe networks. Background Technology

[0002] As a crucial municipal infrastructure, urban drainage networks are responsible for the collection and transportation of urban sewage, and their operation directly impacts the city's production and living environment. However, with the increasing number and density of commercial, food industry, and service facilities in cities year by year, the amount of grease in drainage networks is also increasing annually. Grease in sewage easily reacts with other components to form insoluble solid deposits. These deposits exhibit strong adhesion within the drainage network and accumulate and grow larger as the grease content increases, ultimately reducing the network's water conveyance capacity and leading to overflow problems.

[0003] To better prevent the accumulation of grease deposits, it is necessary to monitor the grease content in drainage pipe networks. Traditional methods for detecting grease content include Fourier transform infrared analysis, thermogravimetric analysis, and gas chromatography-flame ionization detector analysis. These methods have advantages such as high stability, high sensitivity, and high accuracy. However, due to the time-consuming sampling and chemical analysis processes and the high cost of the detection instruments, real-time monitoring of grease content is not possible.

[0004] Therefore, in order to achieve the tracking and monitoring of grease content in drainage pipe networks, it is urgent to develop a rapid, simple and easy-to-operate method for detecting grease content in urban drainage pipe networks. Summary of the Invention

[0005] This application provides a data-driven method for detecting grease content in drainage pipe networks, which can quickly detect the grease content in sewage in drainage pipe networks.

[0006] In a first aspect, embodiments of this application provide a data-driven method for detecting grease content in drainage pipe networks, comprising:

[0007] The data of various indicators of the drainage network under test at the first moment are obtained, including water quality indicators, water quantity indicators and network air quality indicators.

[0008] The data of each indicator are input into a preset oil content prediction model to obtain the oil content of the sewage in the drainage network to be tested at the first moment.

[0009] The data-driven method for detecting grease content in drainage pipe networks in this embodiment of the application obtains the corresponding grease content by inputting various index data of the drainage pipe network to be tested into a preset grease content prediction model. This method is simple and easy to implement, and can quickly obtain the detection results of grease content. Furthermore, since the input data indexes include water quality index data, water quantity index data, and pipe network air index data, the detection results are correlated with multi-source index data, thus achieving higher accuracy.

[0010] In one possible implementation of the first aspect, the water quality data includes pH, conductivity, and water temperature; the water quantity data includes flow rate and liquid level; and the network air data includes relative humidity and air pressure.

[0011] In one possible implementation of the first aspect, the oil content prediction model is a GRU deep learning model, an RNN deep learning model, or an LSTM deep learning model.

[0012] For example, the oil content prediction model is a GRU deep learning model, which includes an input layer, a hidden layer and an output layer. The hidden layer has 4 layers, and each hidden layer includes 58 GRU neural network structures.

[0013] In one possible implementation of the first aspect, the oil content prediction model is obtained by training a deep learning model using an original feature dataset V; wherein the original feature dataset V = {(v 11 ,v 12 ,v 13 ,v 14 ,v 15 ,v 16 ,v 17 ,v 18 ),(v 21 ,v 22 ,v 23 ,v 24 ,v 25 ,v 26 ,v 27 ,v 28 ),……,(v n1 ,v n2 ,v n3 ,v n4 ,v n5 ,v n6 ,v n7 ,v n8 )}, where n represents the number of record sets in the original feature dataset V; where v m1 v m2 v m3 v m4 vm5 v m6 v m7 and v m8 The values ​​represent pH, conductivity, water temperature, flow rate, liquid level, relative humidity, air pressure, and oil content corresponding to the m-th set of recorded data, respectively; where each set of data corresponds to the same moment at a collection point of the drainage pipe network to be tested, and the data in the original feature dataset are arranged in order.

[0014] In one possible implementation of the first aspect, the data in the original feature dataset is collected from multiple collection points of the drainage network to be tested, and adjacent data groups for the same collection point have the same collection time difference.

[0015] For example, the training method of the deep learning model includes:

[0016] The original feature dataset is obtained from the collection points of the drainage pipe network to be tested;

[0017] The data in the original feature dataset are normalized to obtain the processed dataset;

[0018] The initial deep learning model is trained using the dataset to obtain the oil content prediction model;

[0019] The normalization process includes normalizing the data in the original feature dataset using the following formula:

[0020]

[0021] Among them, v ij v represents the data of the i-th indicator in the j-th group of records. i ′ j This represents the data of the i-th indicator after normalization in the j-th group of records, max(v i ) and min(v i ) represent the maximum value and minimum value of the i-th indicator in the original feature dataset, respectively.

[0022] Secondly, embodiments of this application provide a data-driven device for detecting grease content in drainage pipe networks, comprising:

[0023] The data acquisition module is used to acquire various indicator data of the drainage network under test at the first moment. The various indicator data include water quality indicator data, water quantity indicator data and network air indicator data.

[0024] The oil content acquisition module is used to input the data of the various indicators into a preset oil content prediction model to obtain the oil content of the sewage in the drainage network to be tested at the first moment.

[0025] Thirdly, embodiments of this application provide a data processing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the detection method described in any one of the first aspects above.

[0026] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the detection method described in any one of the first aspects above.

[0027] Fifthly, embodiments of this application provide a computer program product that, when run on a data processing device, causes the data processing device to perform the detection method described in any one of the first aspects.

[0028] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a schematic diagram of a data-driven grease content detection system in a drainage network provided in an embodiment of this application;

[0031] Figure 2 This is a schematic flowchart of a data-driven method for detecting grease content in a drainage network according to an embodiment of this application;

[0032] Figure 3 This is a schematic diagram of the structure of a GRU neural network provided in an embodiment of this application;

[0033] Figure 4 This is a graph showing the model evaluation results of a GRU neural network-based oil content prediction model provided in an embodiment of this application on the training and test sets;

[0034] Figure 5This is a comparison chart of the predicted and actual values ​​of the oil content prediction model based on the GRU neural network provided in an embodiment of this application on the test set;

[0035] Figure 6 This is a schematic diagram of the structure of a data-driven device for detecting grease content in a drainage network according to an embodiment of this application;

[0036] Figure 7 This is a schematic diagram of the structure of a data processing device provided in an embodiment of this application. Detailed Implementation

[0037] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0038] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0039] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0040] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0041] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0042] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0043] Current methods for detecting grease in urban drainage pipe networks suffer from slow detection speeds and difficulty in achieving real-time monitoring. To address these issues, this application provides a data-driven method for detecting grease content in drainage pipe networks. The method acquires various indicator data from the drainage pipe network under test, inputs this data into a pre-defined grease content prediction model for processing, and outputs the liquid grease content in the drainage. This method is simple to implement and can quickly obtain grease content detection results, facilitating real-time monitoring.

[0044] This application provides a data-driven method for detecting grease content in drainage pipe networks, which can be applied to, for example... Figure 1 The data-driven grease content detection system shown is used to detect the grease content in drainage network 1.

[0045] like Figure 1 As shown, the detection system includes a data processing device 10 and multiple data acquisition devices 20, wherein the multiple data acquisition devices 20 are respectively installed at multiple key nodes of the drainage pipe network 1. The data acquisition devices 20 are used to collect indicator data at their respective key nodes and send the collected indicator data to the data processing device 10. The data processing device 10 has a built-in preset oil content prediction model, which inputs the received indicator data into the preset oil content prediction model for processing, thereby obtaining the oil content result corresponding to each data acquisition device 20. For example, the data processing device 10 and the data acquisition devices 20 transmit data wirelessly.

[0046] In this embodiment of the application, the data acquisition device 20 can periodically measure the indicator data of the key node and send it to the data processing device 10. The data processing device 10 processes the indicator data received at each time to obtain the oil content of the key node where the data acquisition device 20 is located at each time.

[0047] In this embodiment, the key nodes are determined based on the topology of the drainage network under test, such as the junctions or intersections of important pipelines. Preferably, to facilitate equipment installation and testing, the key nodes can be the locations of manholes (also known as inspection wells) in the drainage network.

[0048] In some embodiments, the data processing device 10 can also generate and display a grease content change curve corresponding to each data acquisition device 20, thereby making the change in grease content in the drainage network 1 more intuitive. For example, the horizontal axis of the grease content change curve represents time points, and the vertical axis represents grease content.

[0049] The data acquisition device 20 in this embodiment is a device that integrates multiple data acquisition functions. For example, the data acquisition device 20 can be used to collect water quality index data, water quantity index data and air index data in the drainage pipe network.

[0050] Figure 2 This application provides a schematic flowchart illustrating the data-driven method for detecting grease content in drainage networks. (See attached diagram.) Figure 2 The method includes:

[0051] Step S210: Obtain the data of various indicators of the drainage network under test at the first moment. The data of various indicators include water quality indicators, water quantity indicators and network air quality indicators.

[0052] In this embodiment, the first moment can be any moment within any time period, and the various indicator data refer to the indicator data of the same node in the drainage network under test at a certain moment. Among them, the water quality data refers to the data related to the quality indicators of liquids (e.g., sewage) in the drainage network; the water quantity indicator data refers to the data related to the water flow rate; and the network air indicator data refers to the data related to the air in the network.

[0053] In one embodiment, water quality indicators include pH, conductivity, and water temperature; water quantity indicators include flow rate and liquid level; and network air indicators include relative humidity and air pressure. All of these indicators are readily available and quickly obtainable, ensuring the timeliness of data acquisition and enabling real-time detection of oil content.

[0054] Step S220: Input the data of various indicators into the preset oil content prediction model to obtain the oil content of the sewage in the drainage network to be tested at the first moment.

[0055] In this embodiment, the oil content prediction model is pre-trained. The output oil content has a certain time attribute; the model's output is the oil content corresponding to the collection time of the various input indicator data.

[0056] In this embodiment, the training samples for the oil content prediction model can be obtained by collecting index data and detecting oil content in the drainage pipe network to be tested, or by collecting index data and detecting oil content in a drainage pipe network with a similar topology to the drainage pipe network to be tested. The oil content in the training samples can be detected using traditional methods such as Fourier transform infrared analysis, thermogravimetric analysis, or gas chromatography-flame ionization detector analysis.

[0057] In one embodiment, the oil content prediction model is a GRU (Gate Recurrent Unit) deep learning model, an RNN (Recurrent Neural Network) deep learning model, or an LSTM (Long Short-Term Memory) deep learning model. Different deep learning models can be trained using different datasets and conventional training methods, which will not be elaborated upon here.

[0058] For example, the oil content prediction model is obtained by training a deep learning model on the original feature dataset V; where the original feature dataset V = {(v 11 ,v 12 ,v 13 ,v 14 ,v 15 ,v 16 ,v 17 ,v 18 ),(v 21 ,v 22 ,v 23 ,v 24 ,v 25 ,v 26 ,v 27 ,v 28 ),……,(v n1 ,v n2 ,v n3 ,v n4 ,v n5 ,v n6 ,v n7 ,v n8 )}, where n represents the number of record sets in the original feature dataset V; where v m1 v m2 v m3 v m4 v m5 v m6 v m7 and v m8The values ​​represent pH, conductivity, water temperature, flow rate, liquid level, relative humidity, air pressure, and grease content corresponding to the m-th set of recorded data, respectively; each set of data corresponds to the same moment at a collection point of the drainage pipe network to be tested, and the data in the original feature dataset are arranged in order.

[0059] In one embodiment, the data in the original feature dataset is collected from multiple collection points of the drainage pipe network to be tested, and adjacent data groups for the same collection point have the same collection time difference. In this embodiment, the training samples for the grease content prediction model are obtained by collecting various indicator data and detecting the corresponding grease content at multiple time points from multiple collection points of the drainage pipe network to be tested, with the same time difference between adjacent time points. Using collection data from multiple collection points of the drainage pipe network to be tested with the same time difference as training samples ensures that the training data is highly representative in both the spatial and temporal dimensions of the pipe network to be tested. This guarantees that the trained model has a good application effect on the drainage pipe network to be tested, and ensures high accuracy of the model for each node and each detection time point of the drainage pipe network to be tested.

[0060] In one embodiment, a method for training a deep learning model includes:

[0061] The original feature dataset is obtained from the collection points of the drainage pipe network to be tested;

[0062] The data in the original feature dataset is normalized to obtain the processed dataset;

[0063] The initial deep learning model was trained using the dataset to obtain a fat content prediction model;

[0064] The data normalization process includes normalizing the data in the original feature dataset using the following formula:

[0065]

[0066] Among them, v ij v represents the data of the i-th indicator in the j-th group of records. i ′ j This represents the data of the i-th indicator after normalization in the j-th group of records, max(v i ) and min(v i ) represent the maximum and minimum values ​​of the i-th indicator in the original feature dataset, respectively. In this embodiment, after normalizing the data in the original feature dataset, the processed dataset is used for model training to avoid interference to the model that may be caused by differences in data amplitude.

[0067] In one embodiment, when various indicator data are input into a preset oil content prediction model, the preset oil content prediction model will normalize the input data based on existing data (data during model training and data input in previous applications) and input data; when outputting oil content, the output data will be back-normalized to output oil content.

[0068] In one embodiment, the oil content prediction model is a GRU deep learning model, which includes an input layer, a hidden layer, and an output layer. The hidden layer has four layers, and each hidden layer includes 58 GRU neural network structures.

[0069] The following example uses the GRU deep learning model to illustrate the training process of the fat content prediction model:

[0070] Step (1): Establish an online monitoring system at key nodes of the drainage network to collect data on various indicators, including the pH, conductivity, water temperature, flow rate, liquid level, relative humidity, and air pressure of the wastewater in the drainage network. Simultaneously, collect the grease content in the drainage network during the same time period. Obtain the original feature dataset V, V = {(v...} 11 ,v 12 ,v 13 ,v 14 ,v 15 ,v 16 ,v 17 ,v 18 ),(v 21 ,v 22 ,v 23 ,v 24 ,v 25 ,v 26 ,v 27 ,v 28 ),……,(v n1 ,v n2 ,v n3 ,v n4 ,v n5 ,v n6 ,v n7 ,v n8 )}, where n is the number of records in the original feature dataset V. In this embodiment, data collection and grease content detection are performed on a total of 6 key nodes of the drainage network to be tested. The data collection period is 70 days, with each day consisting of 4-hour periods, and 6 sets of data are collected each day; therefore, n is 2520 in this embodiment.

[0071] Step (2) normalizes the data in the original dataset using the following formula:

[0072]

[0073] Among them, v ij v represents the data of the i-th indicator in the j-th group of records. i ′ j This represents the data of the i-th indicator after normalization in the j-th group of records, max(v i ) and min(v i Let $\mathbf{i}$ and $\mathbf{i}$ represent the maximum and minimum values ​​of the $i$-th indicator in the original feature dataset, respectively. After normalization, the standardized dataset is obtained as: $V' = {(v'\mathbf{i}}$. 11 ,v' 12 ,v' 13 ,v' 14 ,v' 15 ,v' 16 ,v' 17 ,v' 18 ),(v' 21 ,v' 22 ,v' 23 ,v' 24 ,v' 25 ,v' 26 ,v' 27 ,v' 28 ),…,(v' n1 ,v' n2 ,v' n3 ,v' n4 ,v' n5 ,v' n6 ,v' n7 ,v' n8 Then, the dataset V' is divided into a training set and a test set in a 7:3 ratio. Preferably, the training set and the test set are divided in chronological order; for example, if the dataset includes data from 100 time points: the dataset can be divided into a training set for the first 70 time points and a test set for the last 30 time points; or the dataset can also be divided into a test set for the first 30 time points and a training set for the last 70 time points.

[0074] Step (3) Construct a prediction model for oil content in drainage pipe network based on GRU neural network structure, wherein the oil content prediction model includes an input layer, a hidden layer and an output layer.

[0075] Figure 3 This is a schematic diagram of a GRU neural network structure. A Gated Recurrent Unit (GRU) neural network contains a reset gate (R... t ), an update gate (Z t The training process is as follows: Input variable (X) t ) and hidden state vector (H t-1 R is obtained by combining calculations. tand Z t The threshold value is set and activated by the sigmoid(σ) function, which sets X to... t R t and H t-1 Combining, and activating the combination through the tanh function, generates an intermediate candidate hidden state vector H. t Finally, H t 'and H t-1 The weighted average through Z t Dynamic control yields the final output Y. t For detailed calculations, please refer to the following formula:

[0076] R t =σ(X) t w xr +H t-1 w hr +b r )

[0077] Z t =σ(X) t w xz +H t-1 w hz +b z )

[0078] H t =tanh(X) t w xh +(R t *H t-1 )w hh +b h )

[0079] H t =Z t *H t-1 +(1-Z t )*H t ′

[0080] Y t =σ(H t w o +b o )

[0081]

[0082]

[0083] Among them, X t H is the input variable of the GRU network. t '、H t H t-1 These are the candidate hidden state vector, the candidate hidden state vector, and the hidden state vector, respectively; Y tThe output variable of the model is represented by σ; σ is the sigmoid activation function, and tanh is the hyperbolic tangent activation function; w xr w hr w xz w hz w xh 、 and w hh These represent the weights between each layer; b r b z b h and b o These are the bias vectors for the reset gate, update gate, candidate hidden state, and output layer, respectively; * indicates the product of two vectors.

[0084] Step (4), set the input of the model to X. t =(v' n1 ,v' n2 ,v' n3 ,v' n4 ,v' n5 ,v' n6 ,v'n7), the output is Y t =(v' n8 If ), then the correspondence between the input and output of the model for each set of training data is: Y t =f(X) t The GRU model is trained using the backpropagation algorithm, continuously optimizing the corresponding parameter matrix. Training stops when the loss function value reaches the required level. Figure 4 As shown, in this embodiment, the model converges when the number of iterations is 2000.

[0085] The optimal oil content prediction model obtained at the end has 4 hidden layers and 58 GRU neural network structures in the hidden layers.

[0086] Step (5), model evaluation, applying root mean square error (RMSE) and R0 2 The root mean square error (RMSE) and the coefficient of determination (R²) are used as evaluation criteria for the accuracy of model predictions. 2 The calculation formula is as follows:

[0087]

[0088]

[0089] Among them, Y t act and Y t pre Let represent the actual and predicted oil content of the t-th sample, respectively. and ... 2 The larger the value, the more accurate the prediction result.

[0090] In this embodiment, as Figure 4 and Figure 5 As shown, the root mean square error (RMSE) is 0.003, and the coefficient of determination (R²) is... 2 =0.952. This indicates that the oil content detection model has good predictive ability.

[0091] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0092] Corresponding to the method for detecting grease content in drainage pipe networks described in the above embodiments, Figure 6 The diagram shows a structural block diagram of a device for detecting grease content in a drainage network according to an embodiment of this application. For ease of explanation, only the parts related to the embodiment of this application are shown.

[0093] Reference Figure 6 The grease content detection device 60 in the drainage pipe network includes:

[0094] The data acquisition module 61 is used to acquire various indicator data of the drainage network under test at the first moment. The various indicator data include water quality indicator data, water quantity indicator data and network air indicator data.

[0095] The grease content acquisition module 62 is used to input various indicator data into the preset grease content prediction model to obtain the grease content of the sewage in the drainage network to be tested at the first moment.

[0096] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0097] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0098] Figure 7 This is a schematic diagram of the structure of a data processing device provided in one embodiment of this application. Figure 7 As shown, the data processing device 10 of this embodiment includes: at least one processor 11 ( Figure 7 (Only one is shown) a processor, a memory 12, and a computer program 13 stored in the memory 12 and executable on the at least one processor 11, wherein the processor 11 executes the computer program 13 to implement the steps in the embodiments of the methods for detecting grease content in any of the above-described drainage networks.

[0099] The data processing device 10 can be a desktop computer, laptop, handheld computer, or cloud server, etc. This data processing device may include, but is not limited to, a processor 11 and a memory 12. Those skilled in the art will understand that... Figure 7 This is merely an example of the data processing device 10 and does not constitute a limitation on the data processing device 10. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0100] The processor 11 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0101] In some embodiments, the memory 12 may be an internal storage unit of the data processing device 10, such as a hard disk or memory of the data processing device 10. In other embodiments, the memory 12 may be an external storage device of the data processing device 10, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the data processing device 10. Furthermore, the memory 12 may include both internal and external storage units of the data processing device 10. The memory 12 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 12 can also be used to temporarily store data that has been output or will be output.

[0102] This application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the steps in any of the above method embodiments.

[0103] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0104] This application provides a computer program product that, when run on a data processing device, enables the data processing device to implement the steps described in the various method embodiments above.

[0105] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / data processing equipment, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0106] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0107] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0108] In the embodiments provided in this application, it should be understood that the disclosed apparatus / data processing device and method can be implemented in other ways. For example, the apparatus / data processing device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0109] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0110] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A data-driven method for detecting grease content in drainage pipe networks, characterized in that, include: The system acquires various index data of the drainage network under test at the first moment. The various index data include water quality index data, water quantity index data, and network air index data. The water quality index data includes pH, conductivity, and water temperature. The water quantity index data includes flow rate and liquid level. The network air index data includes relative humidity and air pressure. The data of each indicator are input into a preset oil content prediction model to obtain the oil content of the sewage in the drainage network to be tested at the first moment.

2. The detection method as described in claim 1, characterized in that, The oil content prediction model is a GRU deep learning model, an RNN deep learning model, or an LSTM deep learning model.

3. The detection method as described in claim 1, characterized in that, The oil content prediction model is a GRU deep learning model, which includes an input layer, a hidden layer, and an output layer. The hidden layer has four layers, and each hidden layer includes 58 GRU neural network structures.

4. The detection method according to any one of claims 1 to 3, characterized in that, The oil content prediction model is obtained by training a deep learning model on the original feature dataset V; where the original feature dataset V = {(v 11 ,v 12 ,v 13 ,v 14 ,v 15 ,v 16 ,v 17 ,v 18 ),(v 21 ,v 22 ,v 23 ,v 24 ,v 25 ,v 26 ,v 27 ,v 28 ),……,(v n1 ,v n2 ,v n3 ,v n4 ,v n5 ,v n6 ,v n7 ,v n8 )}, where n represents the number of record sets in the original feature dataset V; where v m1 v m2 v m3 v m4 v m5 v m6 v m7 and v m8 The values ​​represent pH, conductivity, water temperature, flow rate, liquid level, relative humidity, air pressure, and oil content corresponding to the m-th set of recorded data, respectively; where each set of data corresponds to the same moment at a collection point of the drainage pipe network to be tested, and the data in the original feature dataset are arranged in chronological order.

5. The detection method as described in claim 4, characterized in that, The data in the original feature dataset is collected from multiple collection points of the drainage pipe network to be tested, and adjacent data groups for the same collection point have the same collection time difference.

6. The detection method as described in claim 4, characterized in that, The training methods for the deep learning model include: The original feature dataset is obtained from the collection points of the drainage pipe network to be tested; The data in the original feature dataset are normalized to obtain the processed dataset; The initial deep learning model is trained using the dataset to obtain the oil content prediction model; The normalization process includes normalizing the data in the original feature dataset using the following formula: Among them, v ij v represents the data of the i-th indicator in the j-th group of records. i ′ j This represents the data of the i-th indicator after normalization in the j-th group of records, max(v i ) and min(v i ) represent the maximum value and minimum value of the i-th indicator in the original feature dataset, respectively.

7. A data-driven device for detecting grease content in a drainage network, characterized in that, include: The data acquisition module is used to acquire various indicator data of the drainage network under test at the first moment. The various indicator data include water quality indicator data, water quantity indicator data and network air indicator data. The water quality indicator data includes pH, conductivity and water temperature. The water quantity indicator data includes flow rate and liquid level. The network air indicator data includes relative humidity and air pressure. The oil content acquisition module is used to input the data of the various indicators into a preset oil content prediction model to obtain the oil content of the sewage in the drainage network to be tested at the first moment.

8. A data processing apparatus, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the detection method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the detection method as described in any one of claims 1 to 6.

Images