Smart urban sewage treatment system

Abstract

The invention relates to the field of smart city sewage treatment, and discloses a smart city sewage treatment system which comprises the following steps: arranging a water quality sensor in a city sewage detection area, and performing parameter treatment on water quality parameters detected by the water quality sensor to realize data fusion; meanwhile, a water quality fluctuation prediction model is constructed based on the fused water quality parameters, and real-time prediction of the water quality parameters is achieved. And finally, performing sewage treatment on the urban sewage detection area based on a real-time prediction result. By predicting the water quality change, the water quality condition and the development trend of the urban sewage can be known in time, scientific data support is provided for a water resource management department, and the purpose of ensuring sustainable utilization of water resources is achieved.

Description

Technical Field

[0001] The present invention relates to the field of smart city sewage treatment, and in particular to a smart city sewage treatment system. Background Art

[0002] Due to the existence of pollution sources such as factory pollution and domestic water pollution in cities, the sewage flowing in urban pipelines has extremely high pollution. If sewage prediction and treatment are not carried out, pollutants in the water body cannot be detected and treated in time, which will lead to continuous deterioration of water quality. The deterioration of water quality will damage the ecological balance, resulting in problems such as the death of aquatic organisms and the reduction of biodiversity, which will cause irreversible damage to the entire ecosystem. At the same time, the deterioration of water quality will affect the development of industries such as agricultural production and fishery farming, causing economic losses. Accurately predicting water quality changes helps to detect and treat pollutants in the water body in time, prevent pollutants from damaging the ecological environment, and helps to maintain the ecological balance. A smart city sewage treatment system can formulate a sewage treatment plan more accurately by predicting water quality changes, improve the efficiency and quality of sewage treatment, and ensure the sustainable use of water resources and the healthy development of the ecological environment. Summary of the Invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a smart city sewage treatment system.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] The first aspect of the present invention provides a smart city sewage treatment method, including the following steps:

[0006] Arrange water quality sensors in the urban sewage detection area, and connect and fuse the water quality parameters of the urban sewage detection area collected by different water quality sensors by combining an anti-interference dynamic routing algorithm and a multi-modal data fusion algorithm;

[0007] Construct a water quality fluctuation prediction model by fusing urban meteorological data and urban pipe network load data through a spatio-temporal convolutional neural network;

[0008] Combine the water quality fluctuation prediction model to predict the water quality change situation in the urban sewage detection area, and treat the sewage in combination with the water quality change situation in the urban sewage detection area.

[0009] Further, in a preferred embodiment of the present invention, the step of arranging water quality sensors in the urban sewage detection area, and connecting and fusing the water quality parameters of the urban sewage detection area collected by different water quality sensors by combining an anti-interference dynamic routing algorithm and a multi-modal data fusion algorithm is specifically as follows:

[0010] Determine the areas within the city that require sewage detection and sewage treatment, calibrate the urban sewage detection areas, introduce a big data network, and based on the big data network, determine the locations in all urban sewage detection areas where water quality sensors are appropriately installed, and calibrate them as water quality sensor installation locations;

[0011] Install water quality sensors at all water quality sensor installation locations. Among them, the water quality sensors are used to detect water quality parameters at the urban sewage detection areas, including dissolved oxygen content, pH value, TP value content, and ammonia nitrogen content;

[0012] Import an anti-interference dynamic routing algorithm into all water quality sensors. Among them, the anti-interference dynamic routing algorithm is used to connect all water quality sensors to form a water quality sensor array network;

[0013] Among them, the method for the anti-interference dynamic routing algorithm to connect all water quality sensors is as follows: The anti-interference dynamic routing algorithm controls the LoRaWAN modulation module in the water quality sensors to realize controlling the water quality sensors to adaptively transmit communication signals to other water quality sensors, and in the urban sewage detection area, install a cluster head node within 200 meters around the water quality sensor;

[0014] Among them, the cluster head node is used to receive communication signals, combine all cluster head nodes to construct the network topology of the urban sewage detection area, and the cluster head node serves as a relay point to continue transmitting communication signals to other water quality sensors through the network topology of the urban sewage detection area to form a water quality sensor array network;

[0015] Real-time detect the water quality parameters in the urban sewage detection area through the water quality sensors, combine the water quality sensor array network, and at the same time introduce a multi-modal data fusion algorithm to perform fusion processing on all water quality parameters to obtain fused water quality parameters.

[0016] Furthermore, in a preferred embodiment of the present invention, the combination of the water quality sensor array network and the simultaneous introduction of a multi-modal data fusion algorithm to perform fusion processing on all water quality parameters to obtain fused water quality parameters is specifically as follows:

[0017] Calibrate the water quality parameters in the urban sewage detection area collected by the water quality sensors as real-time water quality parameters, and perform spline interpolation filling and normalization processing on the real-time water quality parameters to obtain real-time preprocessed water quality parameters;

[0018] Install a spatio-temporal convolutional block in the water quality sensor, and perform real-time spatio-temporal convolution processing on the real-time preprocessed water quality parameters based on the spatio-temporal convolutional block. Among them, the real-time spatio-temporal convolution processing is to traverse all real-time preprocessed water quality parameters through the spatio-temporal convolutional block and synchronize the detection time and detection environment of all real-time preprocessed water quality parameters during the traversal process to obtain spatio-temporally synchronized water quality parameters;

[0019] Determine the sampling rate of the spatio-temporally synchronized water quality parameters detected by the water quality sensor, introduce the Kalman filter optimization algorithm, perform adaptive Kalman filtering on the spatio-temporally synchronized water quality parameters, and generate the fused water quality parameters;

[0020] Among them, the weights of the spatio-temporally synchronized water quality parameters detected by different water quality sensors in the fused water quality parameters are different, and the weights are determined according to the sampling rate of the spatio-temporally synchronized water quality parameters detected by the water quality sensor.

[0021] Furthermore, in a preferred embodiment of the present invention, the spatio-temporal convolutional neural network is used to fuse the urban meteorological data and the urban pipe network load data to construct a water quality fluctuation prediction model, specifically:

[0022] Introduce a meteorological platform, retrieve the real-time meteorological data of the urban sewage detection area based on the meteorological platform, and determine the detection time point and detection location of the real-time meteorological data;

[0023] Introduce a water quality configuration module, connect the water quality sensor array network to the water quality configuration module, and construct a target water quality configuration module. Among them, the target water quality configuration module can combine the water quality sensor array network to construct a neural network model for predicting water quality fluctuations;

[0024] Connect a spatio-temporal feature extraction network and a multi-source LSTM fusion network to the water quality configuration module, and construct a blank water quality fluctuation prediction model in the water quality configuration module;

[0025] Import the current fused water quality parameters into the spatio-temporal feature extraction network, extract the spatio-temporal features of the fused water quality parameters based on the spatio-temporal feature extraction network, and at the same time extract the spatio-temporal features of the real-time meteorological data based on the detection time point and detection location of the real-time meteorological data;

[0026] Retrieve the load data at different positions in the urban pipe network from the big data network. Among them, the load data includes the flow rate and pressure value at different positions in the urban pipe network. In the multi-source LSTM fusion network, fuse the spatio-temporal features of the fused water quality parameters, the spatio-temporal features of the real-time meteorological data, and the load data at different positions in the urban pipe network, and import them into the blank water quality fluctuation prediction model for model training to obtain the water quality fluctuation prediction model;

[0027] Among them, the steps of the model training are to calculate the multi-objective weighted loss value and the spatio-temporal continuity constraint for all the data after parameter fusion, generate the dynamic learning rate of the fused parameters based on the calculation results, and update the dynamic learning rate of the fused parameters to the blank water quality fluctuation prediction model to complete the model training.

[0028] Furthermore, in a preferred embodiment of the present invention, the combined water quality fluctuation prediction model is used to predict the water quality changes in the urban sewage detection area, and the sewage is treated in combination with the water quality changes in the urban sewage detection area, specifically as follows:

[0029] Preset the water quality change prediction time, and use the water quality fluctuation prediction model to predict the water quality parameters of different urban sewage detection areas after the water quality change prediction time, which are calibrated as predicted water quality parameters;

[0030] At the same time, analyze the predicted water quality parameters and the integrated water quality parameters of the urban sewage detection area, and preset the standard water quality parameter range;

[0031] If both the integrated water quality parameters and the predicted water quality parameters are maintained within the standard water quality parameter range, the urban sewage detection area is classified as a qualified urban sewage detection area, where the qualified urban sewage detection area is the urban sewage detection area that does not require sewage treatment;

[0032] If the integrated water quality parameters are maintained within the standard water quality parameter range, but the predicted water quality parameters are not maintained within the standard water quality parameter range, the urban sewage detection area is classified as a predicted unqualified urban sewage detection area;

[0033] Retrieve all pollution sources connected to the predicted unqualified urban sewage detection area in the urban pipe network through the big data network, which are calibrated as first-class pollution sources, and at the same time determine the types and contents of chemical substances discharged in the first-class pollution sources;

[0034] In the urban pipe network, determine the intersection of the first-class pollution source and the unqualified urban sewage detection area, and retrieve the solutions for the types and contents of chemical substances discharged in the first-class pollution source in the big data network, which are calibrated as the first-class pollution source solutions;

[0035] Output the first-class pollution source solutions at the intersection of the first-class pollution source and the unqualified urban sewage detection area until both the integrated water quality parameters and the predicted water quality parameters are maintained within the standard water quality parameter range, and a qualified urban sewage monitoring area is obtained;

[0036] If the integrated water quality parameters are not maintained within the standard water quality parameter range, the urban sewage detection area is classified as the current unqualified urban sewage detection area, and the current unqualified urban sewage detection area is treated for sewage.

[0037] Furthermore, in a preferred embodiment of the present invention, when the integrated water quality parameters are not maintained within the standard water quality parameter range, the urban sewage detection area is classified as the current unqualified urban sewage detection area, and the current unqualified urban sewage detection area is treated for sewage, specifically as follows:

[0038] Determine the types and contents of chemical substances in the sewage in the current unqualified urban sewage detection area, and calibrate them as the types and contents of real-time chemical substances in the sewage;

[0039] First, enclose the current unqualified urban sewage detection area to prevent the current unqualified urban sewage detection area from flowing through the urban pipeline, clean up the pollutants in the current unqualified urban sewage detection area, and at the same time perform sewage sedimentation on the current unqualified urban sewage detection area after the pollutants are cleaned up;

[0040] Preset the sewage sedimentation time. After the sewage sedimentation time, analyze the sewage in the current unqualified urban sewage detection area to determine whether the integrated water quality parameters are maintained within the standard water quality parameter range. If so, remove the enclosure of the current unqualified urban sewage detection area to obtain a qualified urban sewage detection area;

[0041] If not, retrieve and output solutions for the types and contents of real-time chemical substances in the sewage in the big data network.

[0042] The second aspect of the present invention also provides a smart city sewage treatment system. The smart city sewage treatment system includes a memory and a processor. The memory stores a smart city sewage treatment method. When the smart city sewage treatment method is executed by the processor, the following steps are implemented:

[0043] Deploy water quality sensors in the urban sewage detection area, and combine the anti-interference dynamic routing algorithm and the multi-modal data fusion algorithm to connect the water quality sensors and fuse the water quality parameters of the urban sewage detection area collected by different water quality sensors;

[0044] Through the spatio-temporal convolutional neural network, fuse the urban meteorological data and the urban pipe network load data to construct a water quality fluctuation prediction model;

[0045] Combine the water quality fluctuation prediction model to predict the water quality change situation in the urban sewage detection area, and treat the sewage in combination with the water quality change situation in the urban sewage detection area.

[0046] The present invention solves the technical defects existing in the background technology. The present invention has the following beneficial effects: By deploying water quality sensors in the urban sewage detection area and performing parameter processing on the water quality parameters detected by the water quality sensors to achieve data fusion, and at the same time constructing a water quality fluctuation prediction model based on the fused water quality parameters to achieve real-time prediction of water quality parameters. Finally, based on the real-time prediction results, sewage treatment is carried out in the urban sewage detection area. By predicting the water quality change, the water quality status and its development trend of urban sewage can be timely understood, providing scientific data support for the water resource management department and achieving the goal of ensuring the sustainable utilization of water resources. Description of the Drawings

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings of embodiments can also be obtained based on these drawings.

[0048] Figure 1 Shows a flowchart of a smart city sewage treatment method;

[0049] Figure 2 Shows a flowchart of a method for collecting water quality parameters in the urban sewage detection area by combining an anti-interference dynamic routing algorithm and a multi-modal data fusion algorithm;

[0050] Figure 3 Shows a program view of a smart city sewage treatment system. Detailed implementation manners

[0051] In order to be able to more clearly understand the above-mentioned objects, features, and advantages of the present invention, the present invention will be further described in detail below in conjunction with the drawings and specific implementation manners. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0052] Many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0053] Figure 1 Shows a flowchart of a smart city sewage treatment method, including the following steps:

[0054] S102: Deploy water quality sensors in the urban sewage detection area, and combine an anti-interference dynamic routing algorithm and a multi-modal data fusion algorithm to connect and fuse the water quality sensors and collect the water quality parameters of the urban sewage detection area collected by different water quality sensors;

[0055] S104: Through a spatio-temporal convolutional neural network, fuse urban meteorological data and urban pipe network load data to construct a water quality fluctuation prediction model;

[0056] S106: Combine the water quality fluctuation prediction model to predict the water quality change situation in the urban sewage detection area, and treat the sewage in combination with the water quality change situation in the urban sewage detection area.

[0057] Further, in a preferred embodiment of the present invention, the water quality fluctuation prediction model is constructed by fusing urban meteorological data and urban pipe network load data through a spatio-temporal convolutional neural network, specifically as follows:

[0058] Introduce a meteorological platform, retrieve real-time meteorological data of the urban sewage detection area based on the meteorological platform, and determine the detection time point and detection location of the real-time meteorological data;

[0059] Introduce a water quality configuration module, connect the water quality sensor array network with the water quality configuration module, and construct a target water quality configuration module. Among them, the target water quality configuration module can combine the water quality sensor array network to construct a neural network model for predicting water quality fluctuations;

[0060] Connect a spatio-temporal feature extraction network and a multi-source LSTM fusion network to the water quality configuration module, and construct a blank water quality fluctuation prediction model in the water quality configuration module;

[0061] Import the current fused water quality parameters into the spatio-temporal feature extraction network, extract the spatio-temporal features of the fused water quality parameters based on the spatio-temporal feature extraction network, and at the same time, extract the spatio-temporal features of the real-time meteorological data based on the detection time point and detection location of the real-time meteorological data;

[0062] Retrieve the load data of different positions in the urban pipe network from the big data network. Among them, the load data includes the flow rate and pressure value of different positions in the urban pipe network. In the multi-source LSTM fusion network, fuse the spatio-temporal features of the fused water quality parameters, the spatio-temporal features of the real-time meteorological data, and the load data of different positions in the urban pipe network, and import them into the blank water quality fluctuation prediction model for model training to obtain the water quality fluctuation prediction model;

[0063] Among them, the steps of the model training are to calculate the multi-objective weighted loss value and the spatio-temporal continuity constraint for all the data after parameter fusion, generate a dynamic learning rate of the fused parameters based on the calculation results, and update the dynamic learning rate of the fused parameters to the blank water quality fluctuation prediction model to complete the model training.

[0064] It should be noted that the sewage is affected by conditions such as temperature, humidity, and rainfall in meteorological data, as well as by other pollution sources connected in the urban pipeline. Therefore, the meteorological data at the sewage detection area and the load data at different positions in the urban pipe network are determined to construct a water quality fluctuation prediction model. Among them, the water quality fluctuation prediction model is a model that can predict the water quality parameters of the sewage detection area according to meteorological data, current water quality parameters, and parameters of other pollution sources in the pipeline, and it is a convolutional neural network model. The spatio-temporal feature extraction network and the multi-source LSTM fusion network are essential algorithms for constructing the model. The spatio-temporal feature extraction network is used to extract the spatio-temporal features of different data to ensure the unity of the conditional parameters in the model, while the LSTM fusion network is an algorithm that fuses the spatio-temporal features of all data. When the model is trained, the dynamic learning rate of the fused parameters is generated, and then it can be imported into the blank model for model training. Among them, the dynamic learning rate is an important parameter for the model to automatically predict water quality parameters, which is obtained by calculating the multi-objective weighted loss value and the spatio-temporal continuity constraint calculation.

[0065] Furthermore, in a preferred embodiment of the present invention, the water quality change situation of the urban sewage detection area is predicted by combining the water quality fluctuation prediction model, and the sewage is treated in combination with the water quality change situation of the urban sewage detection area. Specifically:

[0066] Preset the water quality change prediction time, and predict the water quality parameters of different urban sewage detection areas after the water quality change prediction time through the water quality fluctuation prediction model, which are calibrated as predicted water quality parameters;

[0067] At the same time, analyze the predicted water quality parameters and the fused water quality parameters of the urban sewage detection area, and preset the standard water quality parameter range;

[0068] If both the fused water quality parameters and the predicted water quality parameters are maintained within the standard water quality parameter range, the urban sewage detection area is classified as a qualified urban sewage detection area, where the qualified urban sewage detection area is an urban sewage detection area that does not require sewage treatment;

[0069] If the fused water quality parameters are maintained within the standard water quality parameter range, but the predicted water quality parameters are not maintained within the standard water quality parameter range, the urban sewage detection area is classified as a predicted unqualified urban sewage detection area;

[0070] Retrieve all pollution sources connected to the predicted unqualified urban sewage detection area in the urban pipe network through the big data network, which are calibrated as category I pollution sources, and at the same time determine the types and contents of chemical substances discharged in the category I pollution sources;

[0071] In the urban pipe network, determine the intersection of a type of pollution source and the unqualified urban sewage detection area, and retrieve the types and contents of chemical substances discharged from the type of pollution source in the big data network. The solution is designated as the type of pollution source solution;

[0072] Output the type of pollution source solution at the intersection of the type of pollution source and the unqualified urban sewage detection area until both the integrated water quality parameters and the predicted water quality parameters are maintained within the standard water quality parameter range, and obtain a qualified urban sewage monitoring area;

[0073] If the integrated water quality parameters are not maintained within the standard water quality parameter range, divide the urban sewage detection area into the current unqualified urban sewage detection area, and treat the sewage in the current unqualified urban sewage detection area.

[0074] It should be noted that the water quality parameter anomalies are divided into current anomalies and post-prediction time anomalies. If there are no anomalies both currently and after the prediction event, it proves that the water quality parameters of the urban sewage detection area are at normal levels both in the current state and after the prediction time, and no sewage treatment is required. If the current is normal and the post-prediction time is abnormal, it proves that there are external factors affecting the monitoring area resulting in water quality parameter anomalies after the prediction time, which is usually judged to be caused by pollution sources in the pipeline. It is necessary to find the source of the pollution source and carry out corresponding pollution treatment to ensure that the chemical substances in the pollution source do not flow into the detection area, so as to ensure that the predicted water quality parameters in the detection area after the prediction time are maintained within the standard water quality parameter range and obtain a qualified urban sewage monitoring area.

[0075] Further, in a preferred embodiment of the present invention, when the integrated water quality parameters are not maintained within the standard water quality parameter range, the urban sewage detection area is divided into the current unqualified urban sewage detection area, and the sewage in the current unqualified urban sewage detection area is treated as follows:

[0076] Determine the types and contents of chemical substances in the sewage in the current unqualified urban sewage detection area, and designate them as the real-time chemical substance types and contents of the sewage;

[0077] First, enclose the current unqualified urban sewage detection area to prevent the current unqualified urban sewage detection area from flowing through the urban pipeline, clean up the pollutants in the current unqualified urban sewage detection area, and at the same time perform sewage sedimentation on the current unqualified urban sewage detection area after the pollutants are cleaned up;

[0078] Preset the sewage sedimentation time. After the sewage sedimentation time, analyze the sewage in the current unqualified urban sewage detection area to determine whether the integrated water quality parameters are maintained within the standard water quality parameter range. If so, remove the enclosure of the current unqualified urban sewage detection area to obtain a qualified urban sewage detection area;

[0079] If not, retrieve the solutions for the types and contents of real-time chemical substances in sewage in the big data network and output them.

[0080] It should be noted that if the current water quality parameters are abnormal, sewage treatment needs to be carried out immediately to prevent the deterioration of water quality parameters. Before sewage treatment, the pipeline needs to be sealed to prevent sewage from flowing to other locations along the pipeline, and the pollutants that can be salvaged in the sewage need to be salvaged and cleaned. Finally, the sewage is precipitated to separate the sewage pollutants from the sewage. If the water quality parameters of the sewage are still unqualified at this time, solutions need to be retrieved in the big data network, combined with the types and contents of chemical substances in the sewage, and the solutions are obtained and output, and the enclosure of the sewage detection area is removed.

[0081] Figure 2 The method flow chart for collecting water quality parameters in the urban sewage detection area by combining the anti-interference dynamic routing algorithm and the multi-modal data fusion algorithm is shown, including the following steps:

[0082] S202: Deploy water quality sensors in the urban sewage detection area, and combine the anti-interference dynamic routing algorithm and the multi-modal data fusion algorithm to connect and fuse the water quality sensors to collect the water quality parameters of the urban sewage detection area collected by different water quality sensors;

[0083] S204: Combine the water quality sensor array network, and at the same time introduce the multi-modal data fusion algorithm to perform fusion processing on all water quality parameters to obtain the fused water quality parameters.

[0084] Further, in a preferred embodiment of the present invention, the step of deploying water quality sensors in the urban sewage detection area, and combining the anti-interference dynamic routing algorithm and the multi-modal data fusion algorithm to connect and fuse the water quality sensors to collect the water quality parameters of the urban sewage detection area collected by different water quality sensors is specifically as follows:

[0085] Determine the areas in the city that need sewage detection and treatment, calibrate the urban sewage detection area, introduce the big data network, and based on the big data network, determine the locations suitable for deploying water quality sensors in all urban sewage detection areas, and calibrate them as the installation locations of water quality sensors;

[0086] Install water quality sensors at all the installation locations of water quality sensors. Among them, the water quality sensors are used to detect the water quality parameters at the urban sewage detection area, including dissolved oxygen content, pH value, TP value content, and ammonia nitrogen content;

[0087] Import the anti-interference dynamic routing algorithm into all water quality sensors. Among them, the anti-interference dynamic routing algorithm is used to connect all water quality sensors to form a water quality sensor array network;

[0088] Among them, the method for the anti-interference dynamic routing algorithm to connect all water quality sensors is as follows: The anti-interference dynamic routing algorithm controls the LoRaWAN modulation module in the water quality sensor to enable the water quality sensor to adaptively transmit communication signals to other water quality sensors. And within a range of 200 meters around the water quality sensor in the urban sewage detection area, a cluster head node is installed.

[0089] Among them, the cluster head node is used to receive communication signals, combine all cluster head nodes to construct the network topology of the urban sewage detection area. And the cluster head node serves as a relay point to continue transmitting communication signals to other water quality sensors through the network topology of the urban sewage detection area, forming a water quality sensor array network.

[0090] The water quality parameters in the urban sewage detection area are detected in real time by the water quality sensors. Combining with the water quality sensor array network and introducing a multi-modal data fusion algorithm at the same time, all water quality parameters are fused to obtain the fused water quality parameters.

[0091] It should be noted that after determining the sewage detection area, since the sewage detection area may be a relatively large range, it is necessary to determine the positions where water quality sensors can be installed for installing water quality sensors. The anti-interference dynamic routing algorithm using LoRaWAN modulation technology is an algorithm that can construct a signal transmission network and can adaptively transmit communication signals between different water quality sensors. Because the water quality at different positions may be different, it is necessary to comprehensively analyze by combining water quality sensors at multiple positions to give a solution. The purpose of setting the cluster head node around the water quality sensor is to construct the network topology. The node is a component of the link in the network topology and serves as a relay point for signals to ensure that the signals will not be lost. Combining the water quality sensors and the cluster head nodes can construct a water quality sensor array network. In the water quality sensor array network, the signals transmitted by different sensors are real-time water quality parameters, but the water quality parameters need to be fused and analyzed.

[0092] Further, in a preferred embodiment of the present invention, the step of combining the water quality sensor array network, introducing a multi-modal data fusion algorithm at the same time, and fusing all water quality parameters to obtain the fused water quality parameters is specifically as follows:

[0093] Calibrate the water quality parameters in the urban sewage detection area collected by the water quality sensors as real-time water quality parameters, and perform spline interpolation filling and normalization processing on the real-time water quality parameters to obtain real-time preprocessed water quality parameters.

[0094] Install a spatio-temporal convolutional block in the water quality sensor, and perform real-time spatio-temporal convolution processing on the real-time preprocessed water quality parameters based on the spatio-temporal convolutional block. Among them, the real-time spatio-temporal convolution processing is to traverse all the real-time preprocessed water quality parameters through the spatio-temporal convolutional block, and synchronize the detection time and detection environment of all the real-time preprocessed water quality parameters during the traversal process to obtain spatio-temporally synchronized water quality parameters;

[0095] Determine the sampling rate of the spatio-temporally synchronized water quality parameters detected by the water quality sensor, introduce the Kalman filter optimization algorithm, perform adaptive Kalman filtering on the spatio-temporally synchronized water quality parameters, and generate fused water quality parameters;

[0096] Among them, the weights of the spatio-temporally synchronized water quality parameters detected by different water quality sensors in the fused water quality parameters are different, and the weights are determined according to the sampling rate of the spatio-temporally synchronized water quality parameters detected by the water quality sensor.

[0097] It should be noted that after the water quality parameters are detected, they need to be preprocessed to achieve the filling of interpolation to ensure data integrity, and normalized processing to achieve data unity. The role of the spatio-temporal convolutional block is to unify the detection time and detection environment of the real-time preprocessed water quality parameters, that is, to unify the data in the time and space dimensions to ensure the rigor of the data. The Kalman filter optimization algorithm is an algorithm that unifies the sampling rates of different data. Only when the sampling rates are unified can the data be fused and analyzed uniformly.

[0098] As Figure 3 shown, the second aspect of the present invention also provides a smart city sewage treatment system. The smart city sewage treatment system includes a memory 31 and a processor 32. The memory 31 stores a smart city sewage treatment method. When the smart city sewage treatment method is executed by the processor 32, the following steps are implemented:

[0099] Deploy water quality sensors in the urban sewage detection area, and combine the anti-interference dynamic routing algorithm and the multi-modal data fusion algorithm to connect the water quality sensors and fuse the water quality parameters of the urban sewage detection area collected by different water quality sensors;

[0100] Through the spatio-temporal convolutional neural network, fuse the urban meteorological data and the urban pipe network load data to construct a water quality fluctuation prediction model;

[0101] Combine the water quality fluctuation prediction model to predict the water quality change situation in the urban sewage detection area, and treat the sewage in combination with the water quality change situation in the urban sewage detection area.

[0102] The above are only the specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A smart city sewage treatment method, characterized in that: The following steps are involved: Water quality sensors are deployed in the urban sewage detection area, and combined with the anti-interference dynamic routing algorithm and the multimodal data fusion algorithm, the water quality sensors are connected and the water quality parameters of the urban sewage detection area collected by different water quality sensors are integrated; Through the spatiotemporal convolutional neural network, urban meteorological data and urban pipe network load data are integrated to build a water quality fluctuation prediction model; Combined with the water quality fluctuation prediction model, the water quality changes in the urban sewage detection area are predicted, and the sewage is treated based on the water quality changes in the urban sewage detection area.

2. According to a smart city sewage treatment method as described in claim 1, it is characterized in that: The water quality sensors are arranged in the urban sewage detection area, and the anti-interference dynamic routing algorithm and the multimodal data fusion algorithm are combined to connect the water quality sensors and fuse the water quality parameters of the urban sewage detection area collected by different water quality sensors, specifically: Determine the areas in the city that need sewage detection and sewage treatment, calibrate the urban sewage detection areas, introduce a big data network, determine the locations where water quality sensors are appropriately deployed in all urban sewage detection areas based on the big data network, and calibrate them as water quality sensor installation locations; Water quality sensors are installed in all water quality sensor installation locations, wherein the water quality sensors are used to detect water quality parameters in the urban sewage detection area, including dissolved oxygen content, pH value, TP value content and ammonia nitrogen content; Introducing an anti-interference dynamic routing algorithm into all water quality sensors, wherein the anti-interference dynamic routing algorithm is used to connect all water quality sensors to form a water quality sensor array network; Among them, the method of connecting all water quality sensors by the anti-interference dynamic routing algorithm is as follows: the anti-interference dynamic routing algorithm controls the LoRaWAN modulation module in the water quality sensor to realize the control of the water quality sensor to adaptively transmit communication signals to other water quality sensors, and installs a cluster head node within 200 meters of the water quality sensor in the urban sewage detection area; The cluster head node is used to receive communication signals, and to combine all cluster head nodes to construct a network topology for urban sewage detection area. The cluster head node acts as a relay point, and continues to transmit communication signals to other water quality sensors through the urban sewage detection area network topology to form a water quality sensor array network. Water quality sensors are used to detect water quality parameters in the urban sewage detection area in real time. Combined with the water quality sensor array network, a multimodal data fusion algorithm is introduced to fuse all water quality parameters to obtain fused water quality parameters.

3. A smart city sewage treatment method according to claim 2, characterized in that: The water quality sensor array network is combined with a multimodal data fusion algorithm to fuse all water quality parameters to obtain fused water quality parameters, specifically: The water quality parameters in the urban sewage detection area collected by the water quality sensor are calibrated as real-time water quality parameters, and the real-time water quality parameters are filled with spline differences and normalized to obtain real-time pre-processed water quality parameters; A spatiotemporal convolution block is installed in the water quality sensor, and real-time spatiotemporal convolution processing is performed on the real-time pre-processed water quality parameters based on the spatiotemporal convolution block, wherein the real-time spatiotemporal convolution processing is to traverse all real-time pre-processed water quality parameters through the spatiotemporal convolution block, and synchronize the detection time and detection environment of all real-time pre-processed water quality parameters during the traversal process to obtain spatiotemporal synchronized water quality parameters; Determine the sampling rate of the spatiotemporal synchronized water quality parameters detected by the water quality sensor, introduce the Kalman filter optimization algorithm, perform adaptive Kalman filtering on the spatiotemporal synchronized water quality parameters, and generate fused water quality parameters; Among them, the weights of the spatiotemporal synchronized water quality parameters detected by different water quality sensors in the fused water quality parameters are different, and the weights are determined according to the sampling rates of the spatiotemporal synchronized water quality parameters detected by the water quality sensors.

4. A smart city sewage treatment method according to claim 1, characterized in that: The water quality fluctuation prediction model is constructed by integrating urban meteorological data and urban pipe network load data through spatiotemporal convolutional neural network, specifically: Introducing a meteorological platform, and retrieving real-time meteorological data of the urban sewage detection area based on the meteorological platform, and determining the detection time point and detection location of the real-time meteorological data; Introducing a water quality configuration module, and connecting the water quality sensor array network with the water quality configuration module to construct a target water quality configuration module, wherein the target water quality configuration module can be combined with the water quality sensor array network to construct a neural network model for predicting water quality fluctuations; In the water quality configuration module, the spatiotemporal feature extraction network and the multi-source LSTM fusion network are connected, and a blank model for water quality fluctuation prediction is constructed in the water quality configuration module; Import the current fused water quality parameters into the spatiotemporal feature extraction network, extract the spatiotemporal features of the fused water quality parameters based on the spatiotemporal feature extraction network, and extract the spatiotemporal features of the real-time meteorological data based on the detection time point and detection position of the real-time meteorological data; Retrieve load data at different locations in the urban pipe network from the big data network, wherein the load data includes flow and pressure values ​​at different locations in the urban pipe network, fuse the spatiotemporal characteristics of fused water quality parameters, the spatiotemporal characteristics of real-time meteorological data, and the load data at different locations in the urban pipe network in a multi-source LSTM fusion network, perform parameter fusion, and import them into a blank water quality fluctuation prediction model for model training to obtain a water quality fluctuation prediction model; Among them, the steps of model training are to calculate the multi-objective weighted loss value and the spatiotemporal continuity constraint for all the data after parameter fusion, and generate the dynamic learning rate of the fused parameters based on the calculation results, and update the dynamic learning rate of the fused parameters to the blank model for water quality fluctuation prediction to complete the model training.

5. A smart city sewage treatment method according to claim 1, characterized in that: The water quality fluctuation prediction model is combined to predict the water quality changes in the urban sewage detection area, and the sewage is treated in combination with the water quality changes in the urban sewage detection area, specifically: Preset the predicted time of water quality change, and use the water quality fluctuation prediction model to predict the water quality parameters of different urban sewage detection areas after the predicted time of water quality change, and calibrate them as predicted water quality parameters; At the same time, the predicted water quality parameters and integrated water quality parameters of the urban sewage detection area are analyzed, and the standard water quality parameter range is preset; If the fused water quality parameters and the predicted water quality parameters are both maintained within the standard water quality parameter range, the urban sewage detection area is divided into a qualified urban sewage detection area, wherein the qualified urban sewage detection area is an urban sewage detection area that does not require sewage treatment; If the integrated water quality parameters are maintained within the standard water quality parameter range, but the predicted water quality parameters are not maintained within the standard water quality parameter range, the urban sewage detection area is divided into a predicted unqualified urban sewage detection area; Through the big data network, all pollution sources in the urban pipe network that are connected to the predicted unqualified urban sewage detection area are retrieved and marked as Class I pollution sources. At the same time, the types and contents of chemical substances discharged in Class I pollution sources are determined; In the urban pipe network, determine the intersection of a type of pollution source and the unqualified urban sewage detection area, and retrieve the type and content of chemical substances discharged from the type of pollution source in the big data network, and mark it as a type of pollution source solution; Outputting a type of pollution source solution at the intersection of the type of pollution source and the unqualified urban sewage detection area until the fused water quality parameters and the predicted water quality parameters are maintained within the standard water quality parameter range to obtain a qualified urban sewage monitoring area; If the integrated water quality parameters are not maintained within the standard water quality parameter range, the urban sewage detection area will be divided into the current unqualified urban sewage detection area, and sewage treatment will be carried out in the current unqualified urban sewage detection area.

6. A smart city sewage treatment method according to claim 5, characterized in that: If the fused water quality parameters are not maintained within the standard water quality parameter range, the urban sewage detection area is divided into the current unqualified urban sewage detection area, and the current unqualified urban sewage detection area is subjected to sewage treatment, specifically: Determine the types and contents of chemical substances in the sewage in the current unqualified urban sewage detection area, and calibrate them as the real-time types and contents of chemical substances in the sewage; First, the unqualified urban sewage detection area is enclosed to prevent the unqualified urban sewage detection area from flowing through the urban pipeline, and the pollutants in the unqualified urban sewage detection area are cleaned, and the unqualified urban sewage detection area after the pollutants are cleaned is subjected to sewage sedimentation; Preset sewage settling time, after the sewage settling time, analyze the sewage in the current unqualified urban sewage detection area to determine whether the fused water quality parameters are maintained within the standard water quality parameter range, if so, remove the enclosure of the current unqualified urban sewage detection area to obtain the qualified urban sewage detection area; If not, the solution of real-time chemical substance types and contents in wastewater is retrieved and output in the big data network.

7. A smart city sewage treatment system, characterized in that: The smart city sewage treatment system includes a memory and a processor, wherein a smart city sewage treatment method program is stored in the memory. When the smart city sewage treatment method program is executed by the processor, the smart city sewage treatment method steps as described in any one of claims 1-6 are implemented.

Images