Distributed stress quantification sewage interception and drainage measurement method, system, medium and equipment

Abstract

The present invention discloses a method, system, medium and equipment for measuring sewage interception and flood control with distributed stress quantification. By collecting real-time sensing information from multiple sensors, including stress information and water flow velocity information, and using advanced data processing technologies such as empirical mode decomposition and blind source separation, it is possible to effectively extract features reflecting stress states from complex sensing signals. By using the PSD algorithm and the time difference positioning model, the stress frequency domain characteristics and stress source location information can be accurately obtained to obtain the stress distribution characteristics of various parts of the system. Combined with the structural parameter information of the system, a dynamic distributed stress quantification model is established, which can simulate the service life changes of the system under stress. According to the stress simulation results, targeted structural improvement suggestions are proposed, and a variety of flood scenarios are adapted to achieve stress state simulation under different states; thus, comprehensive monitoring and dynamic evaluation of the stress state of the sewage interception and flood control measurement system are achieved.

Description

Technical Field

[0001] The present invention relates to the field of sewage interception and flood drainage measurement, and in particular to a sewage interception and flood drainage measurement method, system, medium and equipment for distributed stress quantification. Background Art

[0002] At present, large-capacity swirl interception sewage and drainage rainwater outlet device is increasingly widely used in engineering as the main device for rapid sewage discharge to achieve rainwater and sewage separation. This device can not only solve the problems of mixed rainwater and sewage, poor sewage discharge, and two-way flow; it can also adjust the drainage speed; at the same time, it can comprehensively coordinate rainwater outlets, overflow outlets, inspection wells, filter boxes, rainwater discard devices, sewage interception baskets, infiltration wells, monitoring wells and many other equipment to exert coordinated control. It is one of the current advanced technologies for rapid sewage interception and drainage.

[0003] However, the swirl sewage interception and drainage rainwater inlet device with a large rotation radius needs to withstand greater water flow impact and dynamic loads, and needs to withstand greater wind loads in extreme weather such as wind disasters. Therefore, dynamic stress testing is required to verify its structural stability and safety and provide a basis for design and manufacturing. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose a dynamic stress testing method, system, medium and equipment for a large-capacity cyclone sewage interception and flood drainage stormwater inlet device to verify the structural stability and safety of the device and guide engineering practice.

[0005] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:

[0006] In a first aspect, the present invention provides a method for measuring sewage interception and flood drainage by quantifying distributed stress, which is applicable to a sewage interception and flood drainage measurement system, and the method comprises:

[0007] Obtain parameter information of the sewage interception and drainage measurement system, including material information and structural information of the sewage interception and drainage measurement system, obtain sensor information collected by multiple sensors in real time, including stress information and water flow velocity information, and map and store the sensor information with the position information of the sensor in the sewage interception and drainage measurement system;

[0008] Preprocessing the sensor information one by one, the preprocessing including empirical mode decomposition and blind source separation, to obtain a first sensor feature data set, the first sensor feature data set including a plurality of first sensor features;

[0009] Performing data fusion on the first sensor feature data set to obtain a first weight value of each first sensor feature, and performing secondary assignment on the first sensor feature according to the first weight value to obtain a second sensor feature data set, wherein the second sensor feature data set includes a plurality of second sensor features at the same time node;

[0010] Arrange the second sensor features in the plurality of second sensor feature data sets according to the position information and the time sequence to form a position feature sequence corresponding to each position information;

[0011] The position feature sequences are input into the PSD algorithm model one by one to obtain the stress frequency domain features;

[0012] And, inputting the position feature sequences into the time difference positioning model one by one to obtain the stress source position information corresponding to each position feature sequence;

[0013] According to the stress frequency domain characteristics, the location information of the stress source and the parameter information of the sewage interception and drainage measurement system, a distributed stress quantification model of the current sewage interception and drainage measurement system is constructed;

[0014] According to the time variable, the distributed stress quantification model is simulated for dynamic stress changes to obtain stress simulation results, which include reference service life of different positions of the current sewage interception and drainage measurement system under stress impact;

[0015] A structural assessment report is generated based on the stress simulation results. The structural assessment report includes structural risk information of the current sewage interception and drainage measurement system under stress and structural improvement suggestions.

[0016] In some embodiments, the sensing information is preprocessed one by one, and the preprocessing includes empirical mode decomposition and blind source separation to obtain a first sensing feature data set, and the first sensing feature data set includes a plurality of first sensing features including:

[0017] The sensor information is subjected to empirical mode decomposition one by one to obtain the IMF signal group and residual value of each sensor information, which are expressed by formula (1). Formula (1) is as follows:

[0018] ;

[0019] In formula (1), For sensor information, For the The intrinsic mode functions, is the residual term corresponding to the sensor information, For time, , is a positive integer;

[0020] Perform frequency domain conversion on the multi-layer IMF signal of each sensing information to obtain the first frequency signal corresponding to each of the multi-layer IMF signals, and convert the frequency domain into Fourier transform or wavelet transform;

[0021] The following steps are performed for each layer of first frequency signal corresponding to each sensing information:

[0022] Determine one by one whether the frequency range to which the first frequency signal belongs is within the range of a preset threshold, where the preset threshold includes a preset high frequency threshold and a preset low frequency threshold;

[0023] When the first frequency signal is within the range of a preset high frequency threshold, the first frequency signal is removed;

[0024] and, when the first frequency signal is within a range of a preset low-frequency threshold, removing the first frequency signal;

[0025] The remaining IMF signals corresponding to the first frequency signals are summed and accumulated with the residual value to obtain the first processing feature. The first processing feature is the vector representation of the sensor information after empirical mode decomposition, which is expressed by formula (2). Formula (2) is as follows:

[0026] ;

[0027] In formula (2), is the first processing feature, is the set to which the first frequency signal within the range of the preset high frequency threshold belongs, is a set to which the first frequency signal placed within the range of a preset low-frequency threshold belongs;

[0028] Performing blind source separation on the first processing features to obtain at least one first source feature contained in each first processing feature, mapping and storing the first processing features and the at least one first source feature to obtain a first sensing feature;

[0029] A plurality of first sensory features are aggregated into a first sensory feature data set.

[0030] In some embodiments, performing blind source separation on the first processing features to obtain at least one first source feature contained in each first processing feature includes:

[0031] Initialize a first separation matrix for blind source separation, and construct a blind source separation function of the first processing feature and the current first source feature;

[0032] Iteratively optimizing the first separation matrix by using a BSS algorithm until a preset convergence threshold is reached, where the BSS algorithm is an ICA algorithm or an NMF algorithm;

[0033] Substitute the first processing feature into the first separation matrix after iteration to obtain the first source feature, which is expressed by formula (3). Formula (3) is as follows:

[0034] ;

[0035] In formula (3), is the current first source feature, is the first separation matrix after iteration, is the first processing feature.

[0036] In some embodiments, performing data fusion on the first sensor feature data set to obtain a first weight value of each first sensor feature includes:

[0037] Inputting the first sensor feature data set into the DS algorithm model for data fusion to obtain the confidence of each first sensor feature under a preset feature category, where the preset feature categories include a first category, a second category, a third category, and a fourth category;

[0038] The first sensing features are grouped once according to the confidence under the preset feature category to obtain a first preset feature group, a second preset feature group, a third preset feature group and a fourth preset feature group, and the confidence of each first sensing feature is converted into its corresponding first weight value:

[0039] The first sensor feature is revalued according to the first weight value to obtain a second sensor feature data set, where the second sensor feature data set includes a plurality of second sensor features including:

[0040] Secondarily grouping the first sensor feature according to the position information of the sensor in the sewage interception and drainage determination system to obtain a plurality of regional feature groups, each of which contains at least one first sensor feature, each of which has a position feature, and each of which corresponds to a second sensor feature;

[0041] Calculating a second weight value of each first sensing feature in the same regional feature group;

[0042] A first sensing feature belonging to a first preset feature group in the same regional feature group is weighted according to a first weight value and a second weight value to obtain a first feature value corresponding to the first preset feature group in the same regional feature group;

[0043] A first sensing feature belonging to a second preset feature group in the same regional feature group is weighted according to a first weight value and a second weight value to obtain a second feature value corresponding to the second preset feature group in the same regional feature group;

[0044] The first sensing feature belonging to the third preset feature group in the same regional feature group is weighted according to the first weight value and the second weight value to obtain a third feature value corresponding to the third preset feature group in the same regional feature group;

[0045] A first sensing feature belonging to a fourth preset feature group in the same regional feature group is weighted according to a first weight value and a second weight value to obtain a fourth feature value corresponding to the fourth preset feature group in the same regional feature group;

[0046] splicing the first characteristic value of the first category, the second characteristic value of the second category, the third characteristic value of the third category, and the fourth characteristic value of the fourth category of the same regional characteristic group into a second sensing characteristic;

[0047] Repeat the above steps until a plurality of regional feature groups have a second sensing feature;

[0048] The plurality of second sensing features and the position features are mapped and stored, and collected into a second sensing feature data set.

[0049] In some embodiments, the position feature sequences are input into the PSD algorithm model one by one to obtain stress frequency domain features including:

[0050] Dividing the position feature sequence into a plurality of overlapping segments, each overlapping segment having at least an overlapping area within a preset threshold range;

[0051] A window function is applied to each overlapping segment, and an FFT transformation is performed on the overlapping segment after the window function is applied to obtain the power spectrum density corresponding to each overlapping segment and the frequency axis corresponding to each power spectrum density. The power spectrum density is expressed by formula (4), which is as follows:

[0052] ;

[0053] In formula (4), is the power spectral density, is the length of the overlapping segment, is the result of FFT transformation, is the frequency axis corresponding to the overlapping segments;

[0054] The frequency axis is expressed by formula (5), which is as follows:

[0055] ;

[0056] In formula (5), is the sampling time interval, is the index of the frequency component corresponding to the current overlapping segment;

[0057] Extract stress frequency domain features from power spectral density.

[0058] In some embodiments, the position feature sequences are input into the time difference positioning model one by one, and the stress source position information corresponding to each position feature sequence is obtained, including:

[0059] Each position feature sequence is represented by a matrix and expressed by formula (6). Formula (6) is as follows:

[0060] ;

[0061] In formula (6), For the Position feature sequences, is the first eigenvalue of the first time node, is the second eigenvalue of the first time node, is the third eigenvalue of the first time node, is the fourth eigenvalue of the first time node, For the The first eigenvalue of the time node, For the The second eigenvalue of the time node, For the The third eigenvalue of the time node, For the The fourth eigenvalue of the time node;

[0062] The feature difference of the feature sequence at adjacent positions is calculated to obtain the feature difference value, which includes the first feature difference value, the second feature difference value, the third feature difference value and the fourth feature difference value, which are expressed by formula (7). Formula (7) is as follows:

[0063] ;

[0064] In formula (7), For the The time node The characteristic difference, ;

[0065] The time difference between two adjacent position feature sequences is calculated according to the first feature difference, the second feature difference, the third feature difference, and the fourth feature difference, which is expressed by formula (8). Formula (8) is as follows:

[0066] ;

[0067] In formula (8), For the The first position feature sequence The feature difference is the time difference between the two position feature sequences, For the The rate of change of the preset feature category to which the feature value belongs;

[0068] Assume the stress source location information is , the position information of the sensor corresponding to each position feature sequence is expressed as , the equation group of stress source position information is obtained, as shown in formula (9), which is as follows:

[0069] ;

[0070] In formula (9), For the Position information of position feature sequences;

[0071] Solve the equation group shown in formula (9) to obtain the stress source location information.

[0072] In some embodiments, the distributed stress quantization model is subjected to dynamic stress change simulation according to the time variable to obtain stress simulation results including:

[0073] The stress simulation results are plotted as radar charts and displayed;

[0074] And, mapping and storing the risk position information corresponding to the stress area exceeding the stress risk threshold with the risk stress value and the duration of the risk stress value, and calculating the reference service life of the risk position information. The stress risk threshold is obtained by converting the parameter information of the sewage interception and drainage measurement system;

[0075] The structural assessment report generated based on the stress simulation results includes:

[0076] The stress simulation results are input into the trained neural network model to obtain a structural assessment report.

[0077] In a second aspect, the present invention further provides a distributed stress quantified sewage interception and flood drainage measurement system, which is applicable to the method described in the first aspect, and the system includes: a main body, a first guide plate, a first sewage interception cylinder, a second sewage interception cylinder, a first dredging pipe and a first overflow pipe, the main body includes a first chamber, the first chamber is used to store water, and a first opening is provided above the main body; the first guide plate is spirally distributed at the first opening from top to bottom, the first guide plate has a first inner edge and a first outer edge, the height of the first inner edge in the same radial direction is higher than the first outer edge, the first guide plate also includes a first end and a second end, the first end and the water chamber where the first opening is located The first and second chambers are parallel in height, and the second end portion is arranged lower than the first end portion; the first sewage interception cylinder is arranged at the central axis of the first opening, the side wall of the first sewage interception cylinder is provided with a first sewage interception grid, and the outer side wall of the first sewage interception cylinder is connected to the first inner edge; the second sewage interception cylinder is arranged at the second end portion, and the second sewage interception cylinder is placed in the first chamber, the side wall of the second sewage interception cylinder is provided with a second sewage interception grid, the first chamber and the first opening are separated by the first sewage interception cylinder, the second sewage interception cylinder and the first guide plate; the first dredging pipe is connected to the first chamber; the first overflow pipe is connected to the first chamber, and the first dredging pipe and the first overflow pipe are arranged at different heights.

[0078] In a third aspect, the present invention further provides a computer-readable storage medium storing computer program instructions, which implement the method described in the first aspect when executed by a processor.

[0079] In a fourth aspect, the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method described in the first aspect.

[0080] By adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0081] Different from the existing technology, the above technical solution can effectively extract the characteristics reflecting the stress state from the complex sensor signals by collecting real-time sensor information from multiple sensors, including stress information and water flow velocity information, and using advanced data processing technologies such as empirical mode decomposition and blind source separation. At the same time, by using the PSD algorithm and the time difference positioning model, the stress frequency domain characteristics and stress source location information can be accurately obtained, providing reliable basic data for the establishment of the stress quantification model, and obtaining the stress distribution characteristics of various parts of the system. Combined with the structural parameter information of the system, a dynamic distributed stress quantification model is established, which can simulate the service life changes of the system under stress and provide a basis for structural evaluation and optimization. According to the stress simulation results, the structural risk information of various parts of the system can be grasped in time, and targeted structural improvement suggestions can be put forward, providing effective decision support for the safe operation and extended service life of the system. The comprehensive monitoring and dynamic evaluation of the stress state of the sewage interception and drainage measurement system is realized, and it is adapted to a variety of flood scenarios, and the stress state simulation under different states is realized, which is helpful to improve the sewage interception and drainage measurement system. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0083] Figure 1 It is a schematic diagram of steps S101 to S109 of the sewage interception and flood drainage determination method described in the specific implementation method.

[0084] Figure 2 It is a schematic diagram of steps S201 to S209 of the sewage interception and flood drainage determination method described in the specific implementation method.

[0085] Figure 3It is the first structural schematic diagram of the sewage interception and flood drainage measurement system described in the specific implementation method.

[0086] Figure 4 It is a second structural schematic diagram of the sewage interception and flood drainage measurement system described in the specific implementation method.

[0087] Reference numerals:

[0088] 1. Ontology;

[0089] 2. The first guide plate;

[0090] 3. The first sewage tank;

[0091] 31. The first sewage interception grid;

[0092] 4. The second sewage barrel;

[0093] 41. Second sewage interception grid;

[0094] 5. Sensor. DETAILED DESCRIPTION

[0095] The present invention will be further described in detail below in conjunction with the accompanying drawings and examples. It is particularly noted that the following examples are only used to illustrate the present invention, but are not intended to limit the scope of the present invention. Similarly, the following examples are only partial embodiments of the present invention rather than all embodiments, and all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0096] See also Figure 1 In a first aspect, this embodiment provides a method for measuring sewage interception and flood drainage by using distributed stress quantification, which is applicable to a sewage interception and flood drainage measurement system, and the method includes:

[0097] S101, obtaining parameter information of a sewage interception and drainage measurement system, the parameter information including material information and structural information of the sewage interception and drainage measurement system, obtaining sensor information collected by multiple sensors in real time, the sensor information including stress information and water flow velocity information, and mapping and storing the sensor information with the position information of the sensor in the sewage interception and drainage measurement system;

[0098] S102, preprocessing the sensor information one by one, where the preprocessing includes empirical mode decomposition and blind source separation, to obtain a first sensor feature data set, where the first sensor feature data set includes a plurality of first sensor features;

[0099] S103, performing data fusion on the first sensor feature data set to obtain a first weight value of each first sensor feature, and performing secondary value assignment on the first sensor feature according to the first weight value to obtain a second sensor feature data set, where the second sensor feature data set includes multiple second sensor features at the same time node;

[0100] S104, arranging the second sensor features in the plurality of second sensor feature data sets according to the position information and the time sequence to form a position feature sequence corresponding to each position information;

[0101] S105, inputting the position feature sequences into the PSD algorithm model one by one to obtain stress frequency domain features;

[0102] S106, and inputting the position feature sequences into the time difference positioning model one by one to obtain the stress source position information corresponding to each position feature sequence;

[0103] S107, constructing a distributed stress quantization model of the current sewage interception and flood drainage measurement system according to the stress frequency domain characteristics, the stress source location information and the parameter information of the sewage interception and flood drainage measurement system;

[0104] S108, simulating the dynamic stress change of the distributed stress quantification model according to the time variable to obtain a stress simulation result, wherein the stress simulation result includes a reference service life of different positions of the current sewage interception and drainage measurement system under stress impact;

[0105] S109. Generate a structural assessment report based on the stress simulation results. The structural assessment report includes structural risk information of the current sewage interception and drainage measurement system under stress and structural improvement suggestions.

[0106] It should be noted that the sewage interception and drainage measurement system can be the sewage interception and drainage measurement system described later, or it can be an existing sewage interception and drainage measurement system. Optionally, the sewage interception and drainage measurement system can be a sewage interception and drainage measurement system that has been built and is in operation, or it can be model data of the sewage interception and drainage measurement system. The measurement method shown in this embodiment can measure and simulate the stress of the sewage interception and drainage measurement system during the experiment or simulation process, thereby improving the structure, operation mode or flood drainage method of the existing sewage interception and drainage measurement system.

[0107] Specifically, before executing the aforementioned method, corresponding sensors, including pressure sensors, flow rate sensors, etc., can be installed in the sewage interception and drainage measurement system. The specific installation location can be set according to actual needs. In this embodiment, it can be selected to be distributed in the sewage interception and drainage measurement system in the form of a spider web, that is, the entire sewage interception and drainage measurement system is set with multiple horizontal reference layers distributed in the vertical direction according to the area through which water flows, and multiple sensors are arranged in each horizontal reference layer according to the connection nodes on the spider web, so as to comprehensively collect the stress data of multiple structures in the sewage interception and drainage measurement system caused by water flow in space; the specific sensor distribution density can be set according to actual needs.

[0108] In step S101, the parameter information includes material information and structural information of the sewage interception and drainage measurement system. It can be understood that the material information is the material of the entire system in the sewage interception and drainage measurement system. For example, taking the sewage interception and drainage measurement system described below as an example, the body can be made of cast steel, and the first guide plate can be made of steel plate or alloy plate. Different material information corresponds to different yield stress coefficients, and the wear life in the sewage interception and drainage process is also different. Structural information can be understood as parameters on various dimensions, such as the wall thickness of the body, the wall thickness of the first guide plate, the inclination angle of the first guide plate, etc. In this embodiment, the installation position of the sensor in the sewage interception and drainage measurement system will be converted into a unified position storage format and represented by position information. In this embodiment, it is also necessary to obtain the sensor information collected by multiple sensors in real time. It should be noted that real-time acquisition can be understood as acquisition at a certain frequency, and the sensor information of multiple sensors in the entire sewage interception and drainage measurement system is obtained at the same time. This method can realize the dynamic change acquisition of raw data for subsequent analysis. The sensing information includes stress information and water flow velocity information, wherein the stress information can be understood as the stress information brought by the water flow when it flows through the area where the sensor is located, and the water flow velocity information can be understood as the flow velocity information of the water flow when it flows through the area where the sensor is located. The sensing information is mapped and stored with the position information of the sensor in the sewage interception and drainage measurement system, that is, the sensing information and the position information are mapped one by one, which is convenient for subsequent modeling and analysis.

[0109] In step S102, the sensor information needs to be preprocessed, and the preprocessing includes empirical mode decomposition and blind source separation, wherein empirical mode decomposition can remove noise information in the sensor information to better preserve the accuracy of the information in the sensor information. Further, blind source separation can separate multiple independent feature information contained in the sensor information. For example, stress information can include stress information of impurities and stress information of the water flow liquid itself. Similarly, water flow velocity information can include sediment velocity information with dense sediment distribution and sewage velocity information with sparse sediment distribution. Blind source separation can further subdivide the sensor information and adapt to more specific simulation scenarios. After preprocessing, the sensor information at the same timestamp is regularized to obtain a first sensor feature data set, which includes multiple preprocessed sensor information, that is, the first sensor feature.

[0110] In step S103, data fusion is performed on multiple first sensor features in the first sensor feature data set to obtain a first weight value for each first sensor feature, and a secondary assignment is performed on the first sensor feature according to the first weight value. The secondary assignment is to assign importance to the first sensor feature according to the first weight value. For example, if the first sensor feature is more obvious in the impurity stress part, the first sensor feature is assigned to the impurity stress category and the data evaluation value corresponding to the impurity stress category is assigned. The specific steps can be understood by referring to the following text. After the multiple first sensor features in the same first sensor feature data set are assigned, a second sensor feature data set can be obtained, and the second sensor feature data set contains multiple second sensor features.

[0111] In step S104, multiple second sensor feature data sets correspond to multiple acquisition times, so the multiple second sensor features are sorted according to the position information and time sequence to obtain a position feature sequence corresponding to each position information. This position feature sequence represents the data of the second sensor features in different position information changing with time.

[0112] In step S105, the PSD algorithm model is a power spectrum density analysis method. Power spectrum density is used to describe the power distribution of a signal at different frequencies. The PSD algorithm model can help identify the frequency components present in a signal. By inputting the position feature sequence into the PSD algorithm model, frequency domain features related to stress, i.e., stress frequency domain features, can be obtained. These features can reflect the stress response at different frequencies and help evaluate the performance of the sewage interception and drainage measurement system under dynamic conditions.

[0113] In step S106, the position feature sequence is input into the time difference positioning model. The time difference positioning model can solve the position of the stress source by calculating the signal arrival time difference in different position feature sequences, thereby obtaining a spatial stress source area. For example, the position change of the stress source caused by the vortex flow, swirl flow, and impact flow generated during the drainage process is recorded as stress source position information for the convenience of expression.

[0114] In step S107, a distributed stress quantification model of the current sewage interception and drainage measurement system is constructed based on the stress frequency domain characteristics, the stress source location information, and the parameter information of the sewage interception and drainage measurement system. Specifically, ANSYS and Adams can be used to perform joint dynamic load simulation and modeling to facilitate system analysis. When constructing the distributed stress quantification model, the parameter information of the sewage interception and drainage measurement system can be modified to facilitate the measurement of the sewage interception and drainage measurement system under different materials. Furthermore, the structural parameters can be modified to find the optimal structural parameters of the sewage interception and drainage measurement system.

[0115] In step S108, the dynamic stress change simulation of the distributed stress quantification model is performed according to the time variable, that is, the drainage process is simulated with time as the variable to obtain the stress simulation result. The stress simulation result includes the reference service life of different positions of the current sewage interception and drainage measurement system under stress impact. Optionally, it can also include the stress distribution and dynamic changes of the current sewage interception and drainage measurement system at different positions, and the peak stress and peak-valley stress can be further extracted to facilitate subsequent structural evaluation.

[0116] In step S109, a structural assessment report is generated based on the stress simulation results. The structural assessment report is the structural risk information of the sewage interception and drainage measurement system under the current parameter information, as well as structural improvement suggestions. The structural improvement suggestions include specific numerical adjustment opinions on the structural parameters in the peak stress area, such as increasing the wall thickness of a structure, reducing the bending angle of a structure, etc. It can also include improvements to material information. For example, the structure with the shortest reference service life is selected in the peak stress, and the yield stress range in the structure can be further analyzed to generate corresponding alternative yield stress ranges and corresponding materials, which are listed in the form of a list to facilitate users to select according to actual needs.

[0117] This embodiment collects stress and water flow information from multiple sensors in real time, and combines the material and structural parameters of the system to establish a dynamic distributed stress quantification model. It can accurately describe the stress distribution characteristics of the system at different positions and times, and can also predict the service life changes of the system under stress, providing an important basis for structural optimization and maintenance. This embodiment can also effectively extract key features reflecting the stress state from complex sensor signals through empirical mode decomposition and blind source separation, greatly improving the accuracy and reliability of the analysis. This embodiment uses the PSD algorithm and the time difference positioning model to accurately obtain stress frequency domain characteristics and stress source location information, providing reliable basic data for the establishment of a distributed stress quantification model. The method shown in this embodiment can comprehensively monitor and dynamically evaluate the stress state of the sewage interception and drainage measurement system, provide important technical support for the safe operation and service life management of the system, and facilitate users to adjust some parameters and simulate various sewage interception and drainage conditions according to actual needs, thereby improving the design and modification efficiency of the sewage interception and drainage measurement system.

[0118] In some embodiments, the sensing information is preprocessed one by one, and the preprocessing includes empirical mode decomposition and blind source separation to obtain a first sensing feature data set, and the first sensing feature data set includes a plurality of first sensing features including:

[0119] The sensor information is subjected to empirical mode decomposition one by one to obtain the IMF signal group and residual value of each sensor information, which are expressed by formula (1). Formula (1) is as follows:

[0120] ;

[0121] In formula (1), For sensor information, For the The intrinsic mode functions, is the residual term corresponding to the sensor information, For time, , is a positive integer;

[0122] Perform frequency domain conversion on the multi-layer IMF signal of each sensing information to obtain the first frequency signal corresponding to each of the multi-layer IMF signals, and convert the frequency domain into Fourier transform or wavelet transform;

[0123] The following steps are performed for each layer of first frequency signal corresponding to each sensing information:

[0124] Determine one by one whether the frequency range to which the first frequency signal belongs is within the range of a preset threshold, where the preset threshold includes a preset high frequency threshold and a preset low frequency threshold;

[0125] When the first frequency signal is within the range of a preset high frequency threshold, the first frequency signal is removed;

[0126] and, when the first frequency signal is within a range of a preset low-frequency threshold, removing the first frequency signal;

[0127] The remaining IMF signals corresponding to the first frequency signals are summed and accumulated with the residual value to obtain the first processing feature. The first processing feature is the vector representation of the sensor information after empirical mode decomposition, which is expressed by formula (2). Formula (2) is as follows:

[0128] ;

[0129] In formula (2), is the first processing feature, is the set to which the first frequency signal within the range of the preset high frequency threshold belongs, is a set to which the first frequency signal placed within the range of a preset low-frequency threshold belongs;

[0130] Performing blind source separation on the first processing features to obtain at least one first source feature contained in each first processing feature, mapping and storing the first processing features and the at least one first source feature to obtain a first sensing feature;

[0131] A plurality of first sensory features are aggregated into a first sensory feature data set.

[0132] In this embodiment, after the sensor information is subjected to empirical mode decomposition, multiple intrinsic mode functions, i.e., IMFs, can be obtained. Each intrinsic mode function is denoted as IMF. ; Each IMF The following two conditions need to be met:

[0133] (1) In the entire data set, the number of maximum points and minimum points is equal, or the difference is at most 1;

[0134] (2) At any point, the average value of the IMF is 0:

[0135] ;

[0136] The decomposition process of EMD can be expressed as:

[0137] Initial signal:

[0138] ;

[0139] Continue to break down:

[0140] ;

[0141] ;

[0142]

[0143] ;

[0144] When the residual The decomposition stops when it becomes a monotonic function (such as a linear trend) or a constant.

[0145] Finally, the original signal Broken down into multiple IMFs and a residual , providing multi-level features of the signal.

[0146] That is, after the above steps, each sensing information can be split into multiple IMF functions. By performing frequency domain conversion on the multi-layer IMF signal of each sensing information, multiple first frequency signals corresponding to each sensing information can be obtained. The frequency domain conversion can use Fourier transform or wavelet transform, which is not limited in this embodiment.

[0147] The steps performed on the multiple layers of first frequency signals corresponding to each sensing information can remove unimportant or noisy first frequency signals. In this embodiment, first frequency signals containing certain high-frequency signals or certain low-frequency signals are removed to retain the characteristic information of each sensing information and remove noise information.

[0148] Specifically, a preset high-frequency threshold and a preset low-frequency threshold are obtained. The specific preset high-frequency threshold and preset low-frequency threshold can be determined after multiple tests to meet the noise data contained in the current sewage interception and drainage measurement system. That is, the preset high-frequency threshold represents the high-frequency noise area contained in the current sewage interception and drainage measurement system, and the preset low-frequency threshold represents the low-frequency noise area contained in the current sewage interception and drainage measurement system.

[0149] Determine the frequency range to which the first frequency signal belongs one by one. The frequency range here can be understood as the overall fluctuation range of the first frequency signal, or as the local fluctuation range of the first frequency signal. The judgment process can be calculated according to the percentage. For example, when more than 80% of the area of ​​the first frequency signal is within the range specified by the preset high-frequency threshold, it is determined that the first frequency signal belongs to the range of the preset high-frequency threshold. Similarly, when more than 80% of the area of ​​the first frequency signal is within the range specified by the preset low-frequency threshold, it is determined that the first frequency signal belongs to the range of the preset low-frequency threshold. In any of the two cases, the first frequency signal is eliminated, indicating that the first frequency signal is a noise signal.

[0150] After the judgment is completed, the remaining IMF signals corresponding to the first frequency signals are summed to finally obtain the first processing feature, that is, the vector representation of the sensor information after empirical mode decomposition and removal of corresponding noise. For details, please refer to formula (2) for understanding.

[0151] Further, the aforementioned steps are repeated until each sensing information corresponds to a first processing feature. On this basis, the first processing features are blindly source separated to obtain the first source features contained in each processing feature. In this embodiment, the first source feature can be understood as an independent signal source in the first processing feature. A first processing feature can include multiple first source features or only one first source feature. The number of first source features can be determined based on the specific results obtained after blind source separation.

[0152] Specifically, in some embodiments, blind source separation is performed on the first processing features to obtain at least one first source feature contained in each first processing feature, including:

[0153] Initialize a first separation matrix for blind source separation, and construct a blind source separation function of the first processing feature and the current first source feature;

[0154] Iteratively optimizing the first separation matrix by using a BSS algorithm until a preset convergence threshold is reached, where the BSS algorithm is an ICA algorithm or an NMF algorithm;

[0155] Substitute the first processing feature into the first separation matrix after iteration to obtain the first source feature, which is expressed by formula (3). Formula (3) is as follows:

[0156] ;

[0157] In formula (3), is the current first source feature, is the first separation matrix after iteration, is the first processing feature.

[0158] In this embodiment, the process of blind source separation is specifically as follows:

[0159] The first processing feature is expressed as:

[0160] ;

[0161] Assume that the first source signal of the unknown state contained in the first processing feature is expressed as:

[0162] ;

[0163] The relationship between the first processing feature and the first source signal can be expressed as:

[0164] ;

[0165] in, is the mixing matrix.

[0166] The first processed features are subjected to preprocessing operations such as normalization and decentralization to improve the stability of subsequent algorithms.

[0167] Randomly initialize the first separation matrix W to satisfy conditions.

[0168] According to the selected BSS algorithm, the first separation matrix W is updated iteratively so that the estimated source signal is:

[0169] ;

[0170] The selected BSS algorithm can be independent component analysis (ICA), non-negative matrix factorization (NMF), etc.

[0171] Check whether the first separation matrix W converges to a stable value, or the estimated first source signal Whether the independence, non-Gaussianity and other conditions are met, if not, repeat the above steps until the first separation matrix W converges to a stable value, and the first separation matrix iteration is considered to be completed, recorded as .

[0172] Output the final estimated first source signal:

[0173] .

[0174] After obtaining the first source feature, the first source feature and the first processing feature are mapped and stored, so as to obtain the first sensing feature. Then, multiple first sensing features are collected into a first sensing feature data set for subsequent analysis.

[0175] In this embodiment, empirical mode decomposition can effectively extract multiple intrinsic mode functions and residual terms from complex sensor information, fully reflecting the multi-level characteristics of the sensor information. By performing frequency domain conversion and noise removal on the IMF signal, high-frequency noise and low-frequency noise can be effectively removed, retaining the key information contained in the sensor information. The application of blind source separation technology further extracts the independent signal source in each sensor information, so that the intrinsic characteristics of the sensor information can be captured and described more accurately. This embodiment greatly improves the accuracy and reliability of the subsequent stress analysis model through in-depth analysis and feature extraction of sensor information.

[0176] See also Figure 2 In some embodiments, performing data fusion on the first sensor feature data set to obtain a first weight value of each first sensor feature includes:

[0177] Inputting the first sensor feature data set into the DS algorithm model for data fusion to obtain the confidence of each first sensor feature under a preset feature category, where the preset feature categories include a first category, a second category, a third category, and a fourth category;

[0178] The first sensing features are grouped once according to the confidence under the preset feature category to obtain a first preset feature group, a second preset feature group, a third preset feature group and a fourth preset feature group, and the confidence of each first sensing feature is converted into its corresponding first weight value:

[0179] The first sensor feature is revalued according to the first weight value to obtain a second sensor feature data set, where the second sensor feature data set includes a plurality of second sensor features including:

[0180] S201, performing secondary grouping of the first sensor features according to the position information of the sensor in the sewage interception and drainage measurement system to obtain a plurality of regional feature groups, each of which contains at least one first sensor feature, each of which has a position feature, and each of which corresponds to a second sensor feature;

[0181] S202, calculating a second weight value of each first sensing feature in the same regional feature group;

[0182] S203, performing a weighted operation on the first sensing feature belonging to the first preset feature group in the same regional feature group according to the first weight value and the second weight value to obtain a first feature value corresponding to the first preset feature group in the same regional feature group;

[0183] S204, performing a weighted operation on the first sensing feature belonging to the second preset feature group in the same regional feature group according to the first weight value and the second weight value to obtain a second feature value corresponding to the second preset feature group in the same regional feature group;

[0184] S205, performing a weighted operation on the first sensing feature belonging to the third preset feature group in the same regional feature group according to the first weight value and the second weight value to obtain a third feature value corresponding to the third preset feature group in the same regional feature group;

[0185] S206, performing a weighted operation on the first sensing feature belonging to the fourth preset feature group in the same regional feature group according to the first weight value and the second weight value to obtain a fourth feature value corresponding to the fourth preset feature group in the same regional feature group;

[0186] S207, splicing the first characteristic value of the first category, the second characteristic value of the second category, the third characteristic value of the third category, and the fourth characteristic value of the fourth category of the same regional characteristic group into a second sensing characteristic;

[0187] S208, repeat the above steps until the plurality of regional feature groups have a second sensing feature;

[0188] S209: Map and store the plurality of second sensing features and the position features, and aggregate them into a second sensing feature data set.

[0189] In this embodiment, the DS algorithm model is used to perform data fusion on the same first sensor feature data set to obtain the confidence of each first sensor feature under the preset feature category. Specifically, the DS algorithm model defines the basic belief unit, constructs and builds the evidence set, and further calculates the synthetic belief of the current first sensor feature according to the evidence set, so as to obtain the confidence of each first sensor feature under the preset feature category. Specifically, it can be understood in conjunction with examples: the preset feature category can be divided into the first category, the second category, the third category, and the fourth category according to actual needs. For example, the first category can be defined as a high stress value, the second category can be defined as a low stress value, the third category can be defined as a high water flow rate, and the fourth category can be defined as a low water flow rate. In this process, high and low can be assigned to more specific numerical ranges. In this embodiment, the confidence under the preset feature category indicates the authenticity of the current first sensor feature in different categories. For example, the confidence of the first sensor feature under the first category is 0.9. Similarly, the confidence of the first sensor feature under the second category is 0.2, then the first sensor feature belongs to the preset category of high stress value. By analogy, the DS algorithm model is used to fuse and match different data in terms of features, and finally the first preset feature group, the second preset feature group, the third preset feature group, and the fourth preset feature group can be obtained. Correspondingly, the confidence of each first sensing feature is converted into the first weight value corresponding to the first sensing feature.

[0190] Furthermore, the first sensing feature is secondary valued according to the first weight value, and this process also includes a secondary grouping step of the first sensing feature according to the position information.

[0191] Specifically, in step S201, the first sensor features are grouped twice according to the position information of the sensor in the sewage interception and drainage measurement system, that is, the first sensor features belonging to the same position information are grouped into the same group, thereby obtaining multiple regional feature groups, and the regional feature groups are distinguished by the position information. It should be noted that this process is to obtain the second weight value of the first sensor feature, which is different from the position feature sequence described above.

[0192] In step S202, the second weight value of each first sensing feature in the same regional feature group is calculated. It can be understood that the second weight value can be understood as the acquisition error coefficient of the position information to the first sensing feature in the regional feature group. For example, in some embodiments, there are small eddies generated by corner structures, and the sensors in this area will be affected by the small eddies, which will cause a certain deviation in the water flow velocity information in the collected sensing information, and so on. The acquisition error coefficient can be obtained based on multiple test data, or it can be derived based on the structure and water flow characteristics. In step S202, the second weight value of the same regional feature group can be the same, or it can be based on the geometric center point of the region as the zero point, and the second weight value can be dynamically allocated as the distance between the sensor in the region and the geometric center point is close. This embodiment is not limited to this.

[0193] In steps S203 to S206, the first weight value and the second weight value are used to calculate the characteristic value of the first sensor feature. The difference is that step S203 corresponds to the calculation of the first sensor feature in the first category, step S204 corresponds to the calculation of the first sensor feature in the second category, step S205 corresponds to the calculation of the first sensor feature in the third category, and step S206 corresponds to the calculation of the first sensor feature in the fourth category.

[0194] Finally, in step S207, the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue are concatenated to form a second sensor feature, that is, each second sensor feature contains at least one of the first sensor feature of the first category, the first sensor feature of the second category, the first sensor feature of the third category, and the first sensor feature of the fourth category, so that the fusion of the specific feature categories of the first sensor feature to the second sensor feature can be achieved.

[0195] In step S208, the above steps are repeated until each regional feature group has a second sensing feature.

[0196] In step S209, multiple second sensor features and position features are mapped and stored and collected into a second sensor feature data set. It should be noted that the above steps are for collecting first sensor features in the same time period, and the formed second sensor feature data set contains multiple second sensor features at the same time node.

[0197] In this embodiment, the data fusion process effectively integrates the information in the first sensor feature data set through the DS algorithm model, significantly improving the accuracy and reliability of feature extraction. The authenticity of each first sensor feature under the preset feature category is evaluated by confidence, so that the first sensor features of different categories can be accurately classified and weighted, and the position information is further introduced to form multiple regional feature groups, which enhances the ability to identify the stress and flow rate conditions corresponding to different position information in the sewage interception and drainage measurement system. After weighted calculation and splicing, the constructed second sensor feature data set not only retains important information, but also provides a more comprehensive and accurate feature view, laying a solid foundation for subsequent stress analysis and system optimization, and improving the intelligence level of the system.

[0198] In some embodiments, the position feature sequences are input into the PSD algorithm model one by one to obtain stress frequency domain features including:

[0199] Dividing the position feature sequence into a plurality of overlapping segments, each overlapping segment having at least an overlapping area within a preset threshold range;

[0200] A window function is applied to each overlapping segment, and an FFT transformation is performed on the overlapping segment after the window function is applied to obtain the power spectrum density corresponding to each overlapping segment and the frequency axis corresponding to each power spectrum density. The power spectrum density is expressed by formula (4), which is as follows:

[0201] ;

[0202] In formula (4), is the power spectral density, is the length of the overlapping segment, is the result of FFT transformation, is the frequency axis corresponding to the overlapping segments;

[0203] The frequency axis is expressed by formula (5), which is as follows:

[0204] ;

[0205] In formula (5), is the sampling time interval, is the index of the frequency component corresponding to the current overlapping segment;

[0206] Extract stress frequency domain features from power spectral density.

[0207] In this embodiment, the position feature sequence is divided into a plurality of overlapping segments, and the preset threshold range can be set according to actual needs. For example, the preset threshold range is set to 60% to 90%, and the position feature sequence can be randomly divided into a plurality of overlapping segments according to the preset threshold range. Further, a window function is applied to each overlapping segment, and the window function can be a Hanning window or a Hamming window. This method can reduce the convenience effect. After performing the windowing step, an FFT transformation is performed on each overlapping segment to obtain the power spectrum density corresponding to each overlapping segment and the frequency axis corresponding to each power spectrum density.

[0208] On the basis of obtaining the power spectrum density and the frequency axis, the stress frequency domain characteristics can be extracted according to actual needs, which can specifically include main frequency extraction, extraction of total spectrum power, calculation of average value of weighted frequency, bandwidth, spectrum entropy and other parameters. This embodiment does not limit the stress frequency domain characteristics, and the parameters of the current stress frequency domain characteristics can be set according to the needs of the actual measurement plan.

[0209] This embodiment significantly improves the extraction accuracy and effectiveness of each stress frequency domain feature by dividing each position feature sequence into multiple overlapping segments and applying a window function for FFT transformation. The setting of overlapping segments and the application of window functions effectively reduce the spectrum leakage phenomenon and ensure the stability and reliability of frequency domain analysis. By calculating the power spectrum density and the corresponding frequency axis, the frequency characteristics of the stress signal can be deeply analyzed to provide key data for subsequent dynamic stress assessment. By extracting parameters such as the main frequency, total spectrum power, and average value of the weighted frequency, the stress response characteristics of the system at different frequencies can be fully understood. It provides a scientific basis for optimizing the design and operation of the sewage interception and drainage measurement system and enhances the safety and reliability of the system.

[0210] In some embodiments, the position feature sequences are input into the time difference positioning model one by one, and the stress source position information corresponding to each position feature sequence is obtained, including:

[0211] Each position feature sequence is represented by a matrix and expressed by formula (6). Formula (6) is as follows:

[0212] ;

[0213] In formula (6), For the Position feature sequences, is the first eigenvalue of the first time node, is the second eigenvalue of the first time node, is the third eigenvalue of the first time node, is the fourth eigenvalue of the first time node, For the The first eigenvalue of the time node, For the The second eigenvalue of the time node, For the The third eigenvalue of the time node, For the The fourth eigenvalue of the time node;

[0214] The feature difference of the feature sequence at adjacent positions is calculated to obtain the feature difference value, which includes the first feature difference value, the second feature difference value, the third feature difference value and the fourth feature difference value, which are expressed by formula (7). Formula (7) is as follows:

[0215] ;

[0216] In formula (7), For the The time node The characteristic difference, ;

[0217] The time difference between two adjacent position feature sequences is calculated according to the first feature difference, the second feature difference, the third feature difference, and the fourth feature difference, which is expressed by formula (8). Formula (8) is as follows:

[0218] ;

[0219] In formula (8), For the The first position feature sequence The feature difference is the time difference between the two position feature sequences, For the The rate of change of the preset feature category to which the feature value belongs;

[0220] Assume the stress source location information is , the position information of the sensor corresponding to each position feature sequence is expressed as , the equation group of stress source position information is obtained, as shown in formula (9), which is as follows:

[0221] ;

[0222] In formula (9), For the Position information of position feature sequences;

[0223] Solve the equation group shown in formula (9) to obtain the stress source location information.

[0224] In this embodiment, multiple position feature sequences are represented in the form of a matrix to facilitate the calculation of the subsequent time difference positioning model. It can be understood that in this embodiment, the differential calculation of the data under the first category, the second category, the third category and the fourth category is introduced, so that the calculation results of the four stress source position information can be obtained when solving formula (9). Preferably, the area divided by the four stress source position information is used as the representation of the final stress source position information, which can facilitate the construction accuracy of the subsequent distributed stress quantization model and reduce the error problem caused by multiple data calculations.

[0225] It should be noted that the position information of the position feature sequence is the position information of the corresponding sensor in the sewage interception and drainage measurement system. For the The rate of change of the preset feature category to which the feature value belongs can be understood as follows: In The value range is ,when hour, That is to say , which is the rate of change of the first eigenvalue, and the first eigenvalue belongs to the first category, so For the The rate of change of the preset feature category to which the feature value belongs can be understood here as is the rate of change of the first category to which the first eigenvalue belongs; similarly, when When , the second category of the second eigenvalue, the third category of the third eigenvalue, and the fourth category of the fourth eigenvalue are substituted accordingly.

[0226] This embodiment represents the position feature sequence in matrix form, making full use of the data contained in multiple processed sensor information at each time node, calculating the feature difference between adjacent position feature sequences, and accurately calculating the time difference between two adjacent position feature sequences based on the feature difference and the preset rate of change, laying the foundation for establishing the relationship equation between position information and time difference. Finally, by solving this set of equations, the precise coordinate position of the stress source can be obtained. The conversion from the time domain to the space domain is realized, which is convenient for the subsequent construction of the distributed stress quantification model.

[0227] In some embodiments, the distributed stress quantization model is subjected to dynamic stress change simulation according to the time variable to obtain stress simulation results including:

[0228] The stress simulation results are plotted as radar charts and displayed;

[0229] And, mapping and storing the risk position information corresponding to the stress area exceeding the stress risk threshold with the risk stress value and the duration of the risk stress value, and calculating the reference service life of the risk position information. The stress risk threshold is obtained by converting the parameter information of the sewage interception and drainage measurement system;

[0230] The structural assessment report generated based on the stress simulation results includes:

[0231] The stress simulation results are input into the trained neural network model to obtain a structural assessment report.

[0232] In this embodiment, a simulation algorithm such as finite element analysis (FEA) or computational fluid dynamics (CFD) can be specifically used to dynamically simulate the stress distribution of the sewage interception and drainage measurement system in combination with the time variable. By recording and sorting the stress distribution results at different time nodes, the stress change of the system under time change can be obtained. The simulated stress results are visualized in the form of a radar chart. Data visualization technology, such as Python libraries such as Matplotlib and Plotly, can be used to draw the stress distribution data obtained by dynamic simulation into a radar chart. The stress values ​​at different positions of the system are mapped to the axis of the radar chart to form an intuitive stress distribution graph. By adjusting the scale range and color coding of the radar chart, high stress areas can be better highlighted, risk points can be easily identified, and stress distribution conditions in different areas can be conveniently and intuitively viewed; this graphical display method helps to quickly identify possible high-risk areas. A reasonable stress risk threshold is set, which can be calculated or selected based on information such as material properties and design standards. Traverse the stress distribution data obtained by dynamic simulation, mark the area that exceeds the risk threshold, and use fatigue life calculation formulas, such as Miner linear damage accumulation theory, to estimate the reference service life of the area based on the stress amplitude and duration that exceeds the risk threshold. You can use data structures such as dictionaries and DataFrames to effectively store and manage this information, associate the calculated reference service life with the location information, and form complete risk assessment data. Help evaluate the safety and durability of the structure under specific stress conditions.

[0233] Finally, the stress simulation results are input into the trained neural network model to generate a structural assessment report. Specifically, the neural network model can be constructed based on a convolutional neural network model to train and learn the aforementioned stress simulation results, risk information, etc. After the training is completed, the new stress simulation data is input into the trained model to obtain a comprehensive structural assessment report, which is convenient for structural adjustment and modification of the current sewage interception and drainage measurement system, and improves the design efficiency of the sewage interception and drainage measurement system.

[0234] See also Figure 3 and Figure 4 In the second aspect, the present embodiment further provides a distributed stress quantified sewage interception and drainage measurement system, which is applicable to the method described in the first aspect, and the system comprises: a main body 1, a first guide plate 2, a first sewage interception cylinder 3, a second sewage interception cylinder 4, a first dredging pipe and a first overflow pipe, the main body 1 comprises a first chamber, the first chamber is used for storing water, and a first opening is arranged above the main body 1; the first guide plate 2 is spirally distributed at the first opening from top to bottom, the first guide plate 2 has a first inner edge and a first outer edge, the height of the first inner edge in the same radial direction is higher than the first outer edge, the first guide plate 2 also comprises a first end and a second end, the first end is parallel to the horizontal height where the first opening is located, The second end is arranged lower than the first end; the first sewage interception cylinder 3 is arranged at the central axis of the first opening, the side wall of the first sewage interception cylinder 3 is provided with a first sewage interception grid 31, and the outer side wall of the first sewage interception cylinder 3 is connected to the first inner edge; the second sewage interception cylinder 4 is arranged at the second end, and the second sewage interception cylinder 4 is placed in the first chamber, and the side wall of the second sewage interception cylinder 4 is provided with a second sewage interception grid 41, and the first chamber and the first opening are separated by the first sewage interception cylinder 3, the second sewage interception cylinder 4 and the first guide plate 2; the first dredging pipe is connected to the first chamber; the first overflow pipe is connected to the first chamber, and the first dredging pipe and the first overflow pipe are arranged at different heights; the sensor 5, multiple sensors 5 are arranged on the side wall of the main body 1.

[0235] In this embodiment, when a rainstorm comes, rainwater enters the main body 1 from the first opening. The spiral distribution shape of the first guide plate 2 can guide rainwater to enter the main body 1 in a rotating flow state. During the water intake process, the interception grids on the first sewage interception barrel 3 and the second sewage interception barrel 4 will intercept large particles of pollutants in the water, playing a role in preliminary sewage interception. As the water level rises, when it reaches the setting height of the first dredging pipe, the water will be discharged through the first dredging pipe. If the water level continues to rise and exceeds the setting height of the first overflow pipe, the excess rainwater will be discharged through the first overflow pipe, thereby achieving regulation and control of the water volume.

[0236] This distributed sewage interception and drainage design can effectively improve the system's processing capacity and sewage interception efficiency. The spiral flow state of the first guide plate 2 and the arrangement of the two-stage sewage interception cylinder can make full use of the inertial force of water flow and enhance the ability to capture suspended matter. At the same time, the dredging pipes and overflow pipes with different setting heights can flexibly adjust the drainage speed according to actual conditions to meet the needs of heavy rains of different intensities. In addition, the main body 1 and each component of the device adopt a reasonable structural design, which can withstand large water flow impact and dynamic loads, ensure the long-term stable operation of the system, and maintain good safety performance under extreme weather conditions to reduce the risk of accidents. In short, this distributed stress quantification sewage interception and drainage measurement system integrates sewage interception, drainage, and regulation, and has important application value in improving the flood prevention and disaster reduction capabilities of urban and rural fringe areas and improving water environment quality.

[0237] Furthermore, the following example can be developed in combination with the above solution: in actual application, when it rains, road debris is discharged along with the runoff rainwater through the road diversion ditch to the rainwater concave collection area.

[0238] The device is provided with longitudinal and transverse slopes. When part of the rainwater overflows to the device, the rainwater and debris quickly flow to the lowest "debris collector" under the action of the hydrocyclone. The middle of the collector is designed with a first dirt interception barrel 3 with side wall openings. When the debris in the collector is full, the excess debris accumulates near the inner wall of the first guide plate 2 under the action of the scouring force of the rainwater. Part of the rainwater can be discharged from the side wall of the first dirt interception barrel 3, improving the rainwater discharge capacity and avoiding the obstruction of rainwater discharge due to the accumulation of debris.

[0239] In order to solve the problem of sediment staying in the rainwater concave collection area (permeable brick surface layer), the sewage interception and drainage measurement system proposed in this embodiment should be independently controllable, and the interception multiple of the sewage interception and drainage measurement system should be adjustable. Through the three-layer structure and multiple hydraulic cyclone / filtration collection system, the sewage interception and drainage measurement system should be set on the combined pipe. The sewage interception and drainage measurement system should be in an adjustable form, and the weir height suitable for the drainage system can be adjusted according to the water quality and water volume of multiple incoming water, taking into account the functions of sewage interception and drainage.

[0240] According to the interception flow of the sewage interception and drainage measurement system and the difference in interception flow of the sewage interception main pipe along the river, a buffer storage facility is set to buffer the impact of the intercepted water volume on the first sewage interception grid 31, which is consistent with the interception multiple of the sewage interception main pipe achieved by the device. The sewage interception and drainage measurement system provided by this embodiment is conducive to effectively solving the problem of total control of surface source pollutants in water conservancy runoff and purifying and filtering the initial rainwater in the water conservancy river. The interception of the source of debris (solid garbage such as leaves, sediment, stones, food peels, etc.) can effectively prevent the discharge into the pipe network from causing poor drainage and increasing sediment in the pipe. In addition, the device makes it simple and convenient to clean up debris, which is conducive to reducing urban maintenance costs.

[0241] In a third aspect, this embodiment further provides a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method described in the first aspect is implemented.

[0242] The computer program involved in this embodiment can be stored in a computer device readable storage medium, which includes but is not limited to a disk, a tape, a magnetic card, a floppy disk, a flash memory, an optical disk, an optical card, a read-only memory (ROM), a random access memory (RAM), an erasable programmable ROM (EPROM) and an electrically erasable programmable ROM (EEPROM), etc., and also includes other biological, physical or chemical structures that can achieve the same or equivalent functions as the storage media listed above, such as DNA, RNA, protein and other units with information storage capabilities. In a specific embodiment, the storage medium involved can be one of the above-mentioned media types, or a combination of the above-mentioned media types. In different embodiments, the computer program involved in the embodiment can be stored in a single medium in a centralized manner, or it can be stored in multiple media in a distributed manner. The memory containing the computer device readable storage medium can be a non-volatile memory or a random access memory. These computer device readable storage media can be built into the device, or they can be connected to the device involved in the embodiment as an external device or a part of an external device. In some embodiments, a memory with a computer device readable storage medium is deployed locally; in other embodiments, a solution of deploying the memory away from the processor may also be adopted, such as a network attached memory accessed via an RF circuit or an external port and a communication network, wherein the communication network may be the Internet, one or more intranets, a local area network (LAN), a wide area network (WLAN), a storage area network (SAN), etc., or a suitable combination thereof, as long as the computer device can access the memory. In addition, the computer program involved in the embodiment may be stored in plaintext / ciphertext form, or may be designed as training data, which may be integrated and reorganized through model training and implicitly stored in the parameter state of a deep neural network or other machine learning model.

[0243] In a fourth aspect, this embodiment further provides an electronic device, comprising a memory and a processor, wherein the memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method described in the first aspect.

[0244] The processor described in this embodiment can be implemented by hardware, firmware, software or a combination thereof, and can use a circuit, a single or multiple application-specific integrated circuits (Application Specific Integrated Circuit, ASIC), a digital signal processor (Digital Signal Processor, DSP), a digital signal processing device (Digital Signal Processing Device, DSPD), a programmable logic device (Programmable Logic Device, PLD), a field programmable gate array (Field Programmable Gate Array, FPGA), a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, at least one of a microprocessor, and also includes other physical, biological or chemical structures that can achieve similar or equivalent functions to the processors listed above, such as biological neurons, quantum computing units, DNA computing units, etc., so that the processor can execute some or all of the steps in the computer program or method involved in the various embodiments of the present application, or any combination of the steps mentioned therein.

[0245] The above technical solution can effectively extract the characteristics reflecting the stress state from complex sensor signals by collecting real-time sensor information from multiple sensors, including stress information and water flow velocity information, and using advanced data processing technologies such as empirical mode decomposition and blind source separation. At the same time, using the PSD algorithm and time difference positioning model, the stress frequency domain characteristics and stress source location information can be accurately obtained, providing reliable basic data for the establishment of the stress quantification model, and obtaining the stress distribution characteristics of various parts of the system. Combined with the structural parameter information of the system, a dynamic distributed stress quantification model is established, which can simulate the service life changes of the system under stress and provide a basis for structural evaluation and optimization. According to the stress simulation results, the structural risk information of various parts of the system can be grasped in time, and targeted structural improvement suggestions can be put forward to provide effective decision-making support for the safe operation and extended service life of the system. The comprehensive monitoring and dynamic evaluation of the stress state of the sewage interception and drainage measurement system is realized, and it is adapted to a variety of flood scenarios to achieve stress state simulation under different states, which is helpful to improve the sewage interception and drainage measurement system.

[0246] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0247] If the integrated unit is implemented in the form of 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, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0248] The above descriptions are only some embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any equivalent device or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for measuring sewage interception and flood drainage by distributed stress quantification, applicable to sewage interception and flood drainage measurement system, characterized in that: include: Acquire parameter information of the sewage interception and drainage measurement system, the parameter information includes material information and structural information of the sewage interception and drainage measurement system, acquire sensor information collected by multiple sensors in real time, the sensor information includes stress information and water flow velocity information, and map and store the sensor information with the position information of the sensor in the sewage interception and drainage measurement system; Preprocessing the sensing information one by one, wherein the preprocessing includes empirical mode decomposition and blind source separation, to obtain a first sensing feature data set, wherein the first sensing feature data set includes a plurality of first sensing features; Performing data fusion on the first sensor feature data set to obtain a first weight value of each of the first sensor features, and performing secondary value assignment on the first sensor features according to the first weight value to obtain a second sensor feature data set, wherein the second sensor feature data set includes a plurality of second sensor features at the same time node; Arrange the second sensor features in the plurality of second sensor feature data sets according to the position information and in time order to form a position feature sequence corresponding to each position information; Inputting the position feature sequences into the PSD algorithm model one by one to obtain stress frequency domain features; And, inputting the position feature sequences into the time difference positioning model one by one to obtain the stress source position information corresponding to each position feature sequence; Constructing a distributed stress quantization model of the current sewage interception and drainage measurement system according to the stress frequency domain characteristics, the stress source location information and the parameter information of the sewage interception and drainage measurement system; Performing dynamic stress change simulation on the distributed stress quantification model according to the time variable to obtain stress simulation results, wherein the stress simulation results include reference service life of different positions of the current sewage interception and drainage measurement system under stress impact; A structural assessment report is generated according to the stress simulation result, wherein the structural assessment report includes structural risk information of the current sewage interception and drainage measurement system under stress and suggestions for structural improvement.

2. The method for measuring sewage interception and flood drainage by distributed stress quantification according to claim 1, characterized in that: Preprocessing the sensing information one by one, wherein the preprocessing includes empirical mode decomposition and blind source separation, to obtain a first sensing feature data set, wherein the first sensing feature data set includes a plurality of first sensing features including: The sensor information is subjected to empirical mode decomposition one by one to obtain an IMF signal group and a residual value of each sensor information, which are expressed by formula (1). The formula (1) is as follows: ； In formula (1), For sensor information, For the The intrinsic mode functions, is the residual term corresponding to the sensor information, For time, , is a positive integer; Performing frequency domain conversion on each of the multi-layer IMF signals of the sensing information to obtain first frequency signals corresponding to each of the multi-layer IMF signals, wherein the frequency domain conversion is performed into Fourier transform or wavelet transform; The following steps are performed on each of the multiple layers of first frequency signals corresponding to the sensing information: Determine one by one whether the frequency range to which the first frequency signal belongs is within the range of a preset threshold, wherein the preset threshold includes a preset high frequency threshold and a preset low frequency threshold; When the first frequency signal is within a range of a preset high frequency threshold, the first frequency signal is removed; and, when the first frequency signal is within a range of a preset low-frequency threshold, removing the first frequency signal; The remaining IMF signals corresponding to the first frequency signals are summed and accumulated with the residual value to obtain the first processing feature. The first processing feature is the vector representation of the sensor information after empirical mode decomposition, which is expressed by formula (2). The formula (2) is as follows: ； In formula (2), is the first processing feature, is the set to which the first frequency signal within the range of the preset high frequency threshold belongs, is a set to which the first frequency signal placed within the range of a preset low-frequency threshold belongs; Performing blind source separation on the first processing features to obtain at least one first source feature contained in each of the first processing features, mapping and storing the first processing features and the at least one first source feature to obtain the first sensing feature; A plurality of first sensory features are aggregated into the first sensory feature data set.

3. The method for measuring sewage interception and flood drainage by distributed stress quantification according to claim 2 is characterized in that: Blind source separation is performed on the first processing features to obtain at least one first source feature contained in each of the first processing features, including: Initialize a first separation matrix for blind source separation, and construct a blind source separation function of the first processing feature and the current first source feature; Iteratively optimizing the first separation matrix by using a BSS algorithm until a preset convergence threshold is reached, wherein the BSS algorithm is an ICA algorithm or an NMF algorithm; Substitute the first processing feature into the first separation matrix after iteration to obtain the first source feature, which is expressed by formula (3). The formula (3) is as follows: ； In formula (3), is the first source feature of the current is the first separation matrix after iteration, is the first processing feature.

4. The method for measuring sewage interception and flood drainage by distributed stress quantification according to claim 1, characterized in that: Performing data fusion on the first sensor feature data set to obtain a first weight value of each of the first sensor features includes: Inputting the first sensor feature data set into the DS algorithm model for data fusion to obtain the confidence of each of the first sensor features under a preset feature category, wherein the preset feature categories include a first category, a second category, a third category, and a fourth category; The first sensing features are grouped once according to the confidence under the preset feature category to obtain a first preset feature group, a second preset feature group, a third preset feature group and a fourth preset feature group, and the confidence of each first sensing feature is converted into the first weight value corresponding thereto: The first sensing feature is revalued according to the first weight value to obtain a second sensing feature data set, where the second sensing feature data set includes a plurality of second sensing features including: According to the position information of the sensor in the sewage interception and drainage determination system, the first sensor feature is secondary grouped to obtain a plurality of regional feature groups, each of which contains at least one first sensor feature, each of which has a position feature, and each of which corresponds to one of the second sensor features; Calculating a second weight value of each of the first sensing features in the same regional feature group; Performing a weighted operation on the first sensing feature belonging to the first preset feature group in the same regional feature group according to the first weight value and the second weight value to obtain a first feature value corresponding to the first preset feature group in the same regional feature group; Performing a weighted operation on the first sensing feature belonging to the second preset feature group in the same regional feature group according to the first weight value and the second weight value to obtain a second feature value corresponding to the second preset feature group in the same regional feature group; Performing a weighted operation on the first sensing feature belonging to the third preset feature group in the same regional feature group according to the first weight value and the second weight value to obtain a third feature value corresponding to the third preset feature group in the same regional feature group; Performing a weighted operation on the first sensing feature belonging to the fourth preset feature group in the same regional feature group according to the first weight value and the second weight value to obtain a fourth feature value corresponding to the fourth preset feature group in the same regional feature group; splicing the first characteristic value of the first category, the second characteristic value of the second category, the third characteristic value of the third category and the fourth characteristic value of the fourth category of the same regional characteristic group into the second sensing characteristic; Repeat the above steps until a plurality of the regional feature groups have one second sensing feature; A plurality of the second sensing features and the position features are mapped and stored, and are collected into a second sensing feature data set.

5. The method for measuring sewage interception and flood drainage by distributed stress quantification according to claim 1, characterized in that: The position feature sequences are input into the PSD algorithm model one by one, and the stress frequency domain features obtained include: Dividing the position feature sequence into a plurality of overlapping segments, each of the overlapping segments having at least an overlapping area within a preset threshold range; A window function is applied to each of the overlapping segments, and an FFT transformation is performed on the overlapping segments after the window function is applied to obtain a power spectrum density corresponding to each overlapping segment and a frequency axis corresponding to each power spectrum density. The power spectrum density is expressed by formula (4), and the formula (4) is as follows: ； In formula (4), is the power spectral density, is the length of the overlapping segment, is the result of FFT transformation, is the frequency axis corresponding to the overlapping segment; The frequency axis is represented by formula (5), which is as follows: ； In formula (5), is the sampling time interval, is the index of the frequency component corresponding to the current overlapping segment; The stress frequency domain features are extracted from the power spectral density.

6. The method for measuring sewage interception and flood drainage by distributed stress quantification according to claim 1, characterized in that: Inputting the position feature sequences into the time difference positioning model one by one, obtaining the stress source position information corresponding to each position feature sequence includes: Each of the position feature sequences is represented by a matrix, which is expressed by formula (6). Formula (6) is as follows: ； In formula (6), For the Position feature sequences, is the first eigenvalue of the first time node, is the second eigenvalue of the first time node, is the third eigenvalue of the first time node, is the fourth eigenvalue of the first time node, For the The first eigenvalue of the time node, For the The second eigenvalue of the time node, For the The third eigenvalue of the time node, For the The fourth eigenvalue of the time node; The feature differences of the feature sequences of the adjacent positions are calculated to obtain feature difference values, wherein the feature difference values ​​include a first feature difference value, a second feature difference value, a third feature difference value and a fourth feature difference value, which are expressed by formula (7). The formula (7) is as follows: ； In formula (7), For the The time node The characteristic difference, ; The time difference between two adjacent position feature sequences is calculated according to the first feature difference, the second feature difference, the third feature difference, and the fourth feature difference, which is expressed by formula (8). Formula (8) is as follows: ； In formula (8), For the The first position feature sequence The feature difference is the time difference between the two position feature sequences, For the The rate of change of the preset feature category to which the feature value belongs; Assume the stress source location information is , the position information of the sensor corresponding to each position feature sequence is expressed as , the equation group of the stress source position information is obtained, as shown in formula (9), which is as follows: ； In formula (9), For the Position information of position feature sequences; The equation group shown in formula (9) is solved to obtain the stress source position information.

7. The method for measuring sewage interception and flood drainage by distributed stress quantification according to claim 1, characterized in that: The distributed stress quantification model is subjected to dynamic stress change simulation according to the time variable, and the stress simulation results obtained include: Plotting the stress simulation results into a radar chart and displaying it; And, mapping and storing the risk position information corresponding to the stress area exceeding the stress risk threshold with the risk stress value and the duration of the risk stress value, and calculating the reference service life of the risk position information, wherein the stress risk threshold is obtained by converting the parameter information of the sewage interception and drainage measurement system; The structural assessment report generated based on the stress simulation results includes: The stress simulation results are input into the trained neural network model to obtain the structural evaluation report.

8. A distributed stress quantification sewage interception and flood drainage measurement system, characterized in that: The method for measuring sewage interception and flood drainage according to any one of claims 1 to 7, wherein the system comprises: A body, the body comprising a first chamber, the first chamber is used to store water, and a first opening is provided on the top of the body; A first guide plate, the first guide plate is spirally distributed at the first opening from top to bottom, the first guide plate has a first inner edge and a first outer edge, the height of the first inner edge in the same radial direction is higher than the first outer edge, the first guide plate further includes a first end and a second end, the first end is parallel to the horizontal height of the first opening, and the second end is arranged lower than the first end; A first sewage interception cylinder is arranged at the central axis of the first opening, a first sewage interception grid is arranged on the side wall of the first sewage interception cylinder, and an outer side wall of the first sewage interception cylinder is connected to the first inner edge; A second sewage interception cylinder is arranged at the second end portion, and the second sewage interception cylinder is placed in the first chamber, a side wall of the second sewage interception cylinder is provided with a second sewage interception grid, and the first chamber and the first opening are separated by the first sewage interception cylinder, the second sewage interception cylinder and the first guide plate; A first dredging pipe, connected to the first chamber; The first overflow pipe is communicated with the first chamber, and the first dredging pipe and the first overflow pipe are arranged at different heights.

9. A computer-readable storage medium storing computer program instructions, characterized in that: The computer program instructions implement the method of any one of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method according to any one of claims 1 to 7.

Images