Intelligent sponge city construction comprehensive prediction system and prediction method

Abstract

The invention relates to the technical field of municipal management, and discloses an intelligent sponge city construction comprehensive prediction system and prediction method. Comprising the steps of identifying a sponge construction point from comprehensive construction points, drawing a sponge area in a to-be-detected area, collecting sponge construction data of the sponge area in a sponge period, inputting the collected sponge construction data into a trained machine learning model for predicting the sponge construction state, and predicting the sponge construction state of the to-be-detected area. Making a sponge construction grade of the to-be-detected region; compared with the prior art, the method can achieve the collection of a plurality of different-dimension data in the construction of the smart sponge city, avoids the limitation caused by the collection of single-dimension data, and predicts the high-low state of the construction quality of the smart sponge city in combination with the machine learning model. The problem of low efficiency caused by a prediction mode of analyzing and calculating different data one by one is avoided, and the prediction precision of the construction quality result of the smart sponge city is also improved.

Description

Technical Field

[0001] The present invention relates to the technical field of municipal management, and more specifically, to a comprehensive prediction system and method for the construction of a smart sponge city. Background Art

[0002] As a new generation of urban stormwater management concept, a sponge city aims to enable the city to be as flexible as a sponge in adapting to environmental changes and coping with natural disasters brought by rainwater. With the continuous development of the construction of modern smart sponge cities, more and more sponge facilities are widely used in the construction of smart sponge cities. In order to optimize and adjust the facilities for the construction of smart sponge cities in the future, it is necessary to predict the quality of the construction of smart sponge cities.

[0003] The patent application with the publication number CN116681335A discloses a sponge city construction evaluation method, device, computer device and storage medium, including: obtaining urban construction data and underlying surface data of the area to be evaluated, dividing the area to be evaluated into several drainage sub-areas based on the urban construction data and the underlying surface data, determining the urban construction data and the underlying surface data of each drainage sub-area, calculating the comprehensive runoff coefficient of the drainage sub-area based on the urban construction data and the underlying surface data of the drainage sub-area, calculating the annual runoff total control rate of the drainage sub-area based on the urban construction data, the underlying surface data and the comprehensive runoff coefficient of the drainage sub-area, screening the qualified drainage sub-areas that reach the target value of the annual runoff total control rate based on the annual runoff total control rate of the drainage sub-area, and calculating the area ratio of the qualified drainage sub-areas in the area to be evaluated to obtain the area ratio of the sponge city construction area in the area to be evaluated;

[0004] In the existing comprehensive prediction of sponge city construction, by dividing the sponge construction area into sub-areas, collecting relevant data of a single dimension within the sub-areas, and calculating and integrating the collected relevant data one by one, the effect of comprehensive prediction of sponge city construction is achieved. For example, in the above patent application, by dividing the area to be evaluated into drainage sub-areas and collecting the construction data and the underlying surface data of the drainage sub-areas, and then calculating the collected data one by one to obtain the area ratio of the sponge city construction area. This comprehensive prediction method of sponge city construction can only collect a large amount of the same type of data in a certain dimension in the area to be evaluated, resulting in the problem of single data dimension, making the prediction result of sponge city construction have limitations, and the method of calculating and analyzing the collected data one by one is also prone to calculation errors, resulting in the problem of low efficiency in the prediction of sponge city construction, reducing the accuracy of the comprehensive prediction of sponge city construction.

[0005] In view of this, the present invention proposes a comprehensive prediction system and method for the construction of a smart sponge city to solve the above problems. Summary of the Invention

[0006] To overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solution: An integrated prediction system for the construction of a smart sponge city, comprising:

[0007] A construction point identification module, configured to query one by one the construction attributes of the comprehensive construction points in the area to be measured, and identify the sponge construction points from the comprehensive construction points;

[0008] A region division module, configured to collect the point spacings of the sponge construction points, calibrate the effective spacings, and draw sponge regions in the area to be measured based on the region drawing criterion. The region drawing criterion is that the adjustment ranges of all sponge regions are the same, and there is no overlapping or intersecting region between any two sponge regions;

[0009] A data collection module, configured to construct a sponge cycle and collect sponge construction data of the sponge regions in the sponge cycle. The sponge construction data includes the water level rising rate, the runoff ratio, the total recovery value, and the ponding settlement rate;

[0010] A model prediction module, configured to input the collected sponge construction data into a trained machine learning model for predicting the sponge construction state, predict the sponge construction state of the area to be measured, and determine whether there is an abnormal region in the area to be measured;

[0011] A prediction report module, configured to identify the abnormal regions from the sponge regions and formulate the sponge construction grade of the area to be measured.

[0012] Further, the method for identifying the sponge construction points is as follows:

[0013] In the electronic map, mark one by one the locations of all sponge facilities in the area to be measured, denoted as comprehensive construction points;

[0014] Query one by one the attribute boxes of all comprehensive construction points through the database, and identify the attribute semantics of the construction attributes in the attribute boxes through natural language processing technology;

[0015] The construction attributes with attribute semantics of groundwater, runoff, utilization, and ponding are denoted as groundwater attributes, runoff attributes, utilization attributes, and ponding attributes, and the comprehensive construction points corresponding to the groundwater attributes, runoff attributes, utilization attributes, and ponding attributes are denoted as sponge construction points.

[0016] Further, the method for calibrating the effective spacing is as follows:

[0017] On the electronic map, measure the distance between any two sponge construction points one by one through the scale to obtain the point spacing;

[0018] Compare the sizes of all point spacings one by one, select the minimum value of the point spacings, and record one-third of the minimum value of the point spacings as the effective spacing.

[0019] Furthermore, the sponge area includes a water storage area, a pipe diameter area, a recycling area, and a water accumulation area;

[0020] The drawing methods of the water storage area, the pipe diameter area, the recycling area, and the water accumulation area are as follows:

[0021] S2.1: On the electronic map, draw circles with four sponge construction points as the centers and the effective spacing as the radius to obtain four initial areas;

[0022] S2.2: Using the preset adjustment amplitude as the adjustment standard, increase the radii of the four initial areas simultaneously, drive the boundary lines of the four initial areas to expand and move outward, and observe whether there is overlap or intersection between the boundary lines of any two initial areas;

[0023] S2.3: Repeat S2.2 until there is overlap or intersection between the boundary lines of two initial areas. At the same time, stop adjusting the radii of the four initial areas, and record the initial areas before the boundary lines of the two initial areas overlap or intersect as the drawn areas;

[0024] S2.4: Query the boundary coordinates of the four drawn areas one by one through the positioning system, record the area within the four boundary coordinates in the area to be measured as the sponge area, and record the sponge areas with the construction attributes of groundwater, runoff, utilization, and water accumulation at the sponge construction points as the water storage area, the pipe diameter area, the recycling area, and the water accumulation area respectively.

[0025] Furthermore, the acquisition method of the water level rise rate is as follows:

[0026] During the sponge cycle, query the precipitation start time and precipitation end time of B precipitation events in the water storage area one by one through the time stamp, and record the duration between the precipitation start time and the precipitation end time as the precipitation duration to obtain B precipitation durations;

[0027] Detect the water level heights of the underground rivers in the water storage area at the B precipitation start times and B precipitation end times one by one through the water level sensor to obtain B starting height values and B ending height values;

[0028] Subtract each of the B ending height values from each of the B starting height values, and compare the differences with the B precipitation durations to obtain B sub-rise rates;

[0029] The expression of the sub-rise rate is:

[0030]

[0031] In the formula, SSzb is the b-th sub-rise rate, where b = 1, 2... B, GD jsb is the b-th end height value, GD qsb is the b-th starting height value, SC jsb is the b-th precipitation duration;

[0032] Remove the maximum and minimum values of the sub-rise rates, accumulate the remaining B - 2 sub-rise rates and then take the average to obtain the water level rise rate;

[0033] The expression for the water level rise rate is:

[0034]

[0035] In the formula, SS sw is the water level rise rate, SS zc is the C-th sub-rise rate.

[0036] Furthermore, the acquisition method of the runoff ratio is:

[0037] During the sponge cycle, randomly mark D non-adjacent monitoring times between the B precipitation start times and the B precipitation end times;

[0038] Detect the water flow velocity in the drainage pipe within the pipe diameter area at the D monitoring times one by one through a flow velocity sensor to obtain D water flow velocities;

[0039] Query the cross-sectional area of the drainage pipe through the technical parameter table, and compare the cross-sectional area of the drainage pipe with the average value of the D water flow velocities to obtain the runoff ratio;

[0040] The expression for the runoff ratio is:

[0041]

[0042] In the formula, JL bz is the runoff ratio, JM mj is the cross-sectional area of the drainage pipe, SL lsd is the d-th water flow velocity.

[0043] Furthermore, the acquisition method of the ponding settlement rate is:

[0044] Real-time detect the ponding depth of the ponding on the ground road in the ponding area, and record the ponding with a ponding depth greater than the lower limit of ponding as effective ponding;

[0045] During the sponge cycle, mark the points on the ground road where effective ponding appears one by one to obtain E ponding points, and detect the ponding depth of the E ponding points one by one to obtain E initial depth values;

[0046] After the preset sedimentation time, the water depths of E waterlogging points are detected one by one to obtain E real-time depth values;

[0047] Subtract the E initial depth values from the corresponding E real-time depth values, and then add up the differences and average them to obtain the water accumulation sedimentation rate;

[0048] The expression of water sedimentation rate is:

[0049]

[0050] In the formula, JS cj is the water sedimentation rate, SD cse is the e-th initial depth value, SD sse is the e-th real-time depth value.

[0051] Furthermore, the sponge construction state includes a high construction state and a low construction state;

[0052] The training method of the machine learning model is:

[0053] Collect multiple sets of sponge construction data in the tested area under high construction and low construction status in advance;

[0054] Mark each group of sponge construction data as training features, convert the sponge construction status of each group of training features into digital labels, convert high construction status into 0, and convert low construction status into 1;

[0055] The labeled training features are divided into a training set and a test set. The training set is used to train the machine learning model, and the test set is used to test the machine learning model. The error threshold is preset. When the mean of the prediction errors of all training features in the test set is less than the error threshold, the machine learning model is obtained.

[0056] The method for determining whether there is an abnormal area is:

[0057] When the output of the machine learning model is 0, the sponge construction state of the tested area is a high construction state, and it is determined that there is no abnormal area in the tested area;

[0058] When the output of the machine learning model is 1, the sponge construction state of the tested area is in a low construction state, and it is determined that there are abnormal areas in the tested area.

[0059] Furthermore, the method for identifying abnormal areas is:

[0060] The abnormalities of water level rise rate, runoff ratio, total recovery value and ponding sedimentation rate were identified respectively;

[0061] When the water level rise rate is less than the water level calibration value, the storage area is recorded as an abnormal area;

[0062] When the runoff ratio is greater than the runoff calibration value, mark the pipe diameter area as an abnormal area;

[0063] When the total recovery value is less than the recovery calibration value, mark the recovery area as an abnormal area;

[0064] When the water accumulation settlement rate is less than the water accumulation calibration value, mark the water accumulation area as an abnormal area;

[0065] The sponge construction levels include the first-level construction level, the second-level construction level, and the third-level construction level. The methods for formulating the first-level construction level, the second-level construction level, and the third-level construction level are as follows:

[0066] Count the number of abnormal areas in the area to be measured, and record it as the abnormal value;

[0067] When the abnormal value is 0, formulate the first-level construction level;

[0068] When the abnormal value is 1, formulate the second-level construction level;

[0069] When the abnormal value is 2 or 3 or 4, formulate the third-level construction level.

[0070] A comprehensive prediction method for the construction of a smart sponge city is realized based on the above-mentioned comprehensive prediction system for the construction of a smart sponge city, and includes:

[0071] S1: Query one by one the construction attributes of the comprehensive construction points in the area to be measured, and identify the sponge construction points from the comprehensive construction points;

[0072] S2: Collect the point spacings of the sponge construction points, calibrate the effective spacings, and draw the sponge areas in the area to be measured based on the regional drawing criteria;

[0073] S3: Construct a sponge cycle, and collect the sponge construction data of the sponge areas in the sponge cycle. The sponge construction data includes the water level rising rate, the runoff ratio, the total recovery value, and the water accumulation settlement rate;

[0074] S4: Input the collected sponge construction data into the trained machine learning model for predicting the sponge construction status to predict the sponge construction status of the area to be measured, and determine whether there are abnormal areas in the area to be measured;

[0075] S5: If there are abnormal areas, identify the abnormal areas from the sponge areas and formulate the sponge construction level of the area to be measured.

[0076] The technical effects and advantages of the comprehensive prediction system and prediction method for the construction of a smart sponge city of the present invention:

[0077] The present invention can accurately identify, from numerous and complex comprehensive construction points, the objects that can predict the construction of a smart sponge city by querying one by one the construction attributes of the comprehensive construction points in the area to be measured and identifying the sponge construction points from the comprehensive construction points, thereby improving the accuracy of subsequent data collection. By collecting the point spacings of the sponge construction points, calibrating the effective spacings, and drawing sponge areas in the area to be measured based on the area drawing criteria, it can provide a reasonable time limit for the data collection in the prediction of the construction of a smart sponge city, ensuring the rationality and accuracy of the data collected within the sponge cycle. By collecting the sponge construction data of the sponge areas during the sponge cycle and inputting the collected sponge construction data into a trained machine learning model that predicts the sponge construction status, the collection effect of multiple different-dimensional data in the construction of a smart sponge city can be achieved, avoiding the limitations brought by collecting single-dimensional data. Combining with the machine learning model can quickly and accurately predict the high or low status of the construction quality of a smart sponge city, avoiding the inefficiency problems brought by the prediction method of analyzing and calculating different data one by one, and also improving the prediction accuracy of the construction quality result of a smart sponge city. At the same time, the sponge construction grade of the area to be measured can be formulated according to the abnormal area, so as to reasonably output the final prediction result of the construction of a smart sponge city and provide guiding opinions for the further development and optimization of the subsequent construction of a smart sponge city, realizing the efficient, accurate and reasonable prediction effect of the construction quality of a smart sponge city. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a schematic diagram of the modules of a comprehensive prediction system for the construction of a smart sponge city provided in Embodiment 1 of the present invention;

[0079] Figure 2 It is a schematic flowchart of a comprehensive prediction method for the construction of a smart sponge city provided in Embodiment 2 of the present invention;

[0080] Figure 3 It is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention;

[0081] Figure 4 It is a schematic diagram of the structure of a computer-readable storage medium provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0082] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0083] Embodiment 1: Please refer to Figure 1As shown in the figure, a comprehensive prediction system for the construction of a smart sponge city in this embodiment includes:

[0084] A construction point identification module that marks out the comprehensive construction points in the area to be measured one by one, queries the construction attributes of the comprehensive construction points, and identifies the sponge construction points from the comprehensive construction points according to the construction attributes;

[0085] The area to be measured refers to the area range in the smart sponge city that needs to comprehensively predict the quality of sponge construction and serves as the corresponding range for subsequent relevant data collection. The comprehensive construction point refers to the specific point where the sponge construction facilities are located in the area to be measured, so that each comprehensive construction point corresponds to a single sponge construction facility;

[0086] The construction attribute is used to represent the specific type of the sponge construction facility corresponding to the comprehensive construction point, that is, it can represent each different comprehensive construction point separately and accurately, and serves as the basis for further identification and analysis of the comprehensive construction point in the future.

[0087] The sponge construction point refers to the location where the specific facilities for the construction of the smart sponge city are located in the comprehensive construction point and serves as the corresponding area for subsequent relevant data collection. Since the comprehensive construction point not only includes the specific facilities for the construction of the smart sponge city, but also includes other specific facilities that do not belong to the construction of the smart sponge city, therefore, it is necessary to accurately identify the sponge construction point according to the construction attributes of the comprehensive construction point;

[0088] The identification method of the sponge construction point is as follows:

[0089] In the electronic map, mark out the locations of all sponge facilities in the area to be measured one by one, and record them as comprehensive construction points;

[0090] Query the attribute boxes of all comprehensive construction points one by one through the database, and identify the attribute semantics of the construction attributes in the attribute boxes one by one through natural language processing technology; the attribute box is used to describe the content box of the specific facilities for the sponge city construction corresponding to the comprehensive construction point, and the construction attribute is the corresponding note of the specific facilities for the sponge city construction stored in the attribute box and serves as the basis for judging the sponge construction point;

[0091] Record the construction attribute with the attribute semantics of groundwater as the groundwater attribute, and record the comprehensive construction point corresponding to the groundwater attribute as the sponge construction point;

[0092] Record the construction attribute with the attribute semantics of runoff as the runoff attribute, and record the comprehensive construction point corresponding to the runoff attribute as the sponge construction point;

[0093] Record the construction attribute with the attribute semantics of utilization as the utilization attribute, and record the comprehensive construction point corresponding to the utilization attribute as the sponge construction point;

[0094] The construction attribute with the semantic of ponding is denoted as the ponding attribute, and the comprehensive construction point corresponding to the ponding attribute is denoted as the sponge construction point.

[0095] It should be noted that the groundwater attribute means that the sponge construction facility corresponding to the comprehensive construction point is to introduce the precipitation in the city into the underground river. The runoff attribute means that the sponge construction facility corresponding to the comprehensive construction point is to introduce the precipitation in the city into the drainage pipe for discharge. The utilization attribute means that the sponge construction facility corresponding to the comprehensive construction point is to treat the precipitation in the city and then reuse it. The ponding attribute means that the sponge construction facility corresponding to the comprehensive construction point settles the road ponding after precipitation in the city.

[0096] The area division module collects the point spacing of the sponge construction points, calibrates the effective spacing, and based on the area drawing criterion, draws the sponge area in the area to be measured. The sponge area includes the water storage area, the pipe diameter area, the recycling area, and the ponding area;

[0097] The point spacing refers to the distance between any two sponge construction points, which can provide a reference basis for the length of the distance between any two sponge construction points, so that the point spacing can be used as the drawing radius basis for the subsequent drawing of the sponge area;

[0098] The effective spacing refers to the initial drawing radius of the sponge area, to ensure that each initial drawing area can be used as the object for the subsequent size adjustment of the sponge area, and thus provide a basis for the accurate drawing of the sponge area;

[0099] The calibration method of the effective spacing is as follows:

[0100] On the electronic map, measure the distance between any two sponge construction points one by one through the scale to obtain the point spacing;

[0101] Compare the sizes of all the point spacings one by one, select the minimum value of the point spacing, and record one-third of the minimum value of the point spacing as the effective spacing.

[0102] It should be noted that the size of the effective spacing is only used as the drawing radius during the initial drawing of the sponge area, and is not used as the radius of the final sponge area. Therefore, the initially drawn sponge area still needs to be adjusted in terms of radius size.

[0103] The sponge area refers to the area range where the specific sponge construction facilities are drawn with the sponge construction points as the center points in the area to be measured, so that each sponge area only contains one type of sponge construction facility and the corresponding data;

[0104] Since each sponge area corresponds to a specific sponge construction facility, any two sponge areas need to be kept independent. To ensure that there are no overlapping or intersecting areas between any two sponge areas, it is necessary to adjust the size of the sponge areas under the restriction of the area drawing criteria;

[0105] The sponge areas include a water storage area, a pipe diameter area, a recycling area, and a water accumulation area; the location of the water storage area refers to the area where the facilities for diverting water to the underground river in sponge city construction are located, the location of the pipe diameter area refers to the area where the facilities for the drain pipe diameter in sponge city construction are located, the location of the recycling area refers to the area where the facilities for recycling and utilization in sponge city construction are located, and the location of the water accumulation area refers to the area where urban road water accumulation occurs in sponge city construction.

[0106] The area drawing criteria are as follows: the adjustment ranges of all sponge areas are the same, and there are no overlapping or intersecting areas between any two sponge areas; this can ensure that the initially drawn sponge areas can maintain the consistency of the adjustment range during subsequent adjustments, and at the same time ensure the independence of any two adjusted sponge areas, thus ensuring the singularity of the data types within each sponge area and avoiding the mutual interference between different types of data;

[0107] The drawing methods for the water storage area, the pipe diameter area, the recycling area, and the water accumulation area are as follows:

[0108] S2.1: On the electronic map, draw circles with four sponge construction points as the centers and the effective spacing as the radius respectively to obtain four initial areas;

[0109] S2.2: Using the preset adjustment range as the adjustment standard, increase the radii of the four initial areas simultaneously, drive the boundary lines of the four initial areas to expand and move outwards, and observe whether there are any overlaps or intersections between the boundary lines of any two initial areas; the preset adjustment range is the minimum length unit used to increase the radius of the initial area, which can ensure the synchronization and rationality of the radius increase of the initial area and avoid the phenomenon of the radius of the initial area increasing erratically; specifically, the preset adjustment range is usually much smaller than the effective spacing. Exemplarily, the preset adjustment range is one-tenth of the effective spacing;

[0110] S2.3: Repeat S2.2 until there are overlaps or intersections between the boundary lines of two initial areas. At the same time, stop adjusting the radii of the four initial areas, and record the initial areas before the boundary lines of the two initial areas overlap or intersect as the drawn areas;

[0111] S2.4: Query the boundary coordinates of the four drawing areas one by one through the positioning system, mark the area within the four boundary coordinates in the area to be measured as the sponge area, and mark the sponge areas with the construction attributes of groundwater, runoff, utilization, and water accumulation of the sponge construction points as the water storage area, pipe diameter area, recycling area, and water accumulation area respectively.

[0112] It should be noted that the identified water storage area, pipe diameter area, recycling area, and water accumulation area are used to represent sponge city construction data of different dimensions and types respectively, and meet the requirements of relative independence between data of different dimensions and types.

[0113] The data acquisition module constructs a sponge cycle and acquires sponge construction data of the sponge area in the sponge cycle. The sponge construction data includes the water level rise rate, runoff ratio, total recycling value, and water accumulation settlement rate.

[0114] Sponge construction data refers to the comprehensive data used to represent the quality and achievements of sponge city construction in different sponge areas, which can provide a calculation basis for the quality of subsequent sponge city construction. Since the number of sponge construction data corresponding to each sponge area is not unique, in order to ensure the uniqueness and accuracy of sponge construction data, it is necessary to ensure the accurate and reasonable acquisition of sponge construction data in each sponge area under the limitation of the sponge cycle.

[0115] Since the sponge cycle is used to represent the start time to the end time of the acquisition time of sponge construction data, when constructing the sponge cycle, the current time is used as the end time of the cycle, the end time of the previous sponge city construction prediction is used as the start time of the cycle, and the time period between the end time of the cycle and the start time of the cycle is recorded as the sponge cycle.

[0116] The sponge construction data includes the water level rise rate, runoff ratio, total recycling value, and water accumulation settlement rate.

[0117] The water level rise rate refers to the rising rate of the groundwater level in the water storage area per unit time, which can represent the precipitation drainage construction effect in sponge city construction. The greater the water level rise rate, the better the precipitation drainage construction effect in sponge city construction.

[0118] The acquisition method of the water level rise rate is as follows:

[0119] During the sponge cycle, query the precipitation start time and precipitation end time of B precipitation events in the water storage area one by one through the time stamp, and record the duration between the precipitation start time and the precipitation end time as the precipitation duration to obtain B precipitation durations. A precipitation event refers to an event that can represent the comprehensive data of the precipitation process in the water storage area, providing a basis for querying and calculating the precipitation time.

[0120] The water levels of the underground rivers in the water storage area are detected one by one through a water level sensor at B precipitation start times and B precipitation end times, obtaining B start height values and B end height values;

[0121] After subtracting each of the B end height values from the B start height values one by one, and comparing the differences with the B precipitation durations, B sub-rise rates are obtained;

[0122] The expression for the sub-rise rate is:

[0123]

[0124] In the formula, SS zb is the b-th sub-rise rate, b = 1, 2... B, GD jsb is the b-th end height value, GD qsb is the b-th start height value, SC jsb is the b-th precipitation duration;

[0125] Remove the maximum and minimum values of the sub-rise rates, accumulate the remaining B - 2 sub-rise rates and then take the average to obtain the water level rise rate;

[0126] The expression for the water level rise rate is:

[0127]

[0128] In the formula, SS sw is the water level rise rate, SS zc is the C-th sub-rise rate.

[0129] The runoff ratio refers to the ratio of the drainage pipe in the pipe diameter area to the water flow velocity in the pipe, which can represent the drainage performance of the pipe diameter in the construction of a sponge city. When the runoff ratio is larger, it means that the drainage performance of the pipe diameter in the construction of a sponge city is poorer;

[0130] The acquisition method of the runoff ratio is:

[0131] During the sponge cycle, randomly mark D non-adjacent monitoring times between B precipitation start times and B precipitation end times; the non-adjacent monitoring times can ensure the independence of the data collected at each monitoring time, thus avoiding the phenomenon of mutual adhesion between the data collected at two adjacent monitoring times;

[0132] The water flow velocities in the drainage pipe in the pipe diameter area are detected one by one through a flow velocity sensor at D monitoring times, obtaining D water flow velocities;

[0133] Query the cross-sectional area of the drainage pipe through the technical parameter table, and compare the cross-sectional area of the drainage pipe with the average value of the D water flow velocities to obtain the runoff ratio;

[0134] The expression for the runoff ratio is as follows:

[0135]

[0136] In the formula, JL bz is the runoff ratio, JM mj is the cross-sectional area of the drainage pipe, and SL lsd is the flow velocity of the d-th water flow.

[0137] The total recovery value refers to the total amount of precipitation recycled and reused in the recovery area, which can represent the precipitation reuse performance in the construction of a sponge city. When the total recovery value is larger, it indicates that the precipitation reuse performance in the construction of a sponge city is stronger; the total recovery value is obtained by querying the increase in the water capacity of the storage reservoir in the recovery area during the sponge cycle.

[0138] The ponding settlement rate refers to the settlement rate of the ponding on the ground road in the ponding area per unit time, which can represent the road drainage performance in the construction of a sponge city. When the ponding settlement rate is larger, it indicates that the road drainage performance in the construction of a sponge city is stronger;

[0139] The collection method of the ponding settlement rate is as follows:

[0140] Real-time detect the ponding depth of the ponding on the ground road in the ponding area, and record the ponding with a ponding depth greater than the lower limit of ponding as effective ponding; the lower limit of ponding refers to the minimum ponding depth at which the ponding on the ground road will have a negative impact on traffic, so as to provide a comparison basis for the identification of effective ponding;

[0141] During the sponge cycle, mark the positions of the ground roads where effective ponding appears one by one to obtain E ponding positions, and detect the ponding depth of the E ponding positions one by one to obtain E initial depth values;

[0142] After a preset settlement duration, detect the ponding depth of the E ponding positions one by one to obtain E real-time depth values; the preset settlement duration refers to the minimum duration when the ponding depth of the effective ponding changes significantly, so as to ensure that the data of the ponding depth can be accurately collected after the preset settlement duration;

[0143] Subtract the E initial depth values from the corresponding E real-time depth values respectively, accumulate the differences and then take the average to obtain the ponding settlement rate;

[0144] The expression for the ponding settlement rate is as follows:

[0145]

[0146] In the formula, JS cj is the ponding settlement rate, and SD cseis the e-th initial depth value, SD sse is the e-th real-time depth value.

[0147] The model prediction module inputs the collected sponge construction data into a trained machine learning model that predicts the sponge construction status, predicts the sponge construction status of the area to be measured, and determines whether there are abnormal areas in the area to be measured;

[0148] The machine learning model is an artificial intelligence model used to predict the quality status of sponge city construction based on the input sponge construction data, enabling the machine learning model to predict the sponge construction status corresponding to the sponge construction data based on the sponge construction data, and judging the quality of sponge city construction in the area to be measured through the sponge construction status;

[0149] The sponge construction status includes a high construction status and a low construction status; the high construction status means that the quality of sponge city construction in the area to be measured is relatively high, and the low construction status means that the quality of sponge city construction in the area to be measured is relatively low; the high construction status and the low construction status are obtained after collecting a large number of historical sponge construction statuses corresponding to the water level rise rate, runoff ratio, total recovery value, and ponding settlement rate.

[0150] The training method of the machine learning model is as follows:

[0151] Pre-collect multiple groups of sponge construction data in the area to be measured under the high construction status and the low construction status;

[0152] Mark each group of sponge construction data as training features, label the sponge construction status of each group of training features, the labels include the high construction status and the low construction status, and convert the high construction status and the low construction status into digital labels respectively. Exemplarily, convert the high construction status to 0 and the low construction status to 1;

[0153] Divide the labeled training features into a training set and a test set, use 70% of the training features as the training set, use 30% of the training features as the test set, use the training set to train the machine learning model, use the test set to test the machine learning model, and preset an error threshold. When the mean of the prediction errors of all training features in the test set is less than the error threshold, the machine learning model is obtained.

[0154] Exemplarily, the machine learning model adopts any one of the support vector machine model or the random forest model; the preset error threshold is set in advance according to the actual required accuracy of the machine learning model.

[0155] The abnormal area refers to the sponge area corresponding to the specific sponge construction data when the area to be measured is in a low construction state. By inputting the collected sponge construction data into the trained machine learning model, the sponge construction state corresponding to the sponge construction data can be predicted.

[0156] The method for determining whether there is an abnormal area is as follows:

[0157] When the output of the machine learning model is 0, the sponge construction state of the area to be measured is in a high construction state, and it is determined that there is no abnormal area in the area to be measured.

[0158] When the output of the machine learning model is 1, the sponge construction state of the area to be measured is in a low construction state, and it is determined that there is an abnormal area in the area to be measured.

[0159] The prediction report module identifies the abnormal area from the sponge area and formulates the sponge construction level of the area to be measured according to the abnormal area.

[0160] When it is determined that there is an abnormal area in the area to be measured, it is necessary to identify the specific sponge area corresponding to the abnormal area, which can be used as the basis for subsequent analysis of the sponge city construction results.

[0161] The method for identifying the abnormal area is as follows:

[0162] Perform abnormality identification on the water level rise rate, runoff ratio, total recovery value, and ponding settlement rate respectively. Abnormality identification is used to compare whether the magnitudes of the water level rise rate, runoff ratio, total recovery value, and ponding settlement rate are abnormal, which can lay a foundation for the identification of the abnormal area.

[0163] When the water level rise rate is less than the water level calibration value, it indicates that there is a low-quality phenomenon in the water storage area of the area to be measured, and the water storage area is recorded as an abnormal area. The water level calibration value refers to the minimum value of the water level rise rate when the water storage area is not marked as an abnormal area, which can provide a numerical basis for the abnormal identification of the water level rise rate.

[0164] When the runoff ratio is greater than the runoff calibration value, it indicates that there is a low-quality phenomenon in the pipe diameter area of the area to be measured, and the pipe diameter area is recorded as an abnormal area. The runoff calibration value refers to the maximum value of the runoff ratio when the pipe diameter area is not marked as an abnormal area, which can provide a numerical basis for the abnormal identification of the runoff ratio.

[0165] When the total recovery value is less than the recovery calibration value, it indicates that there is a low-quality phenomenon in the recovery area of the area to be measured, and the recovery area is recorded as an abnormal area. The recovery calibration value refers to the minimum value of the total recovery value when the recovery area is not marked as an abnormal area, which can provide a numerical basis for the abnormal identification of the total recovery value.

[0166] When the water accumulation settlement rate is less than the water accumulation calibration value, it indicates that there is a low-quality phenomenon in the water accumulation area of the area to be measured, and the water accumulation area is recorded as an abnormal area. The water accumulation calibration value refers to the minimum value of the water accumulation settlement rate when the water accumulation area is not marked as an abnormal area, so as to provide a numerical basis for the abnormal identification of the water accumulation settlement rate.

[0167] The sponge construction level is used to represent the level range corresponding to the quality of the sponge city construction in the area to be measured, so as to be able to display the final result of the sponge city construction in the area to be measured;

[0168] The sponge construction level includes the first-level construction level, the second-level construction level, and the third-level construction level; specifically, the sponge city construction quality of the area to be measured corresponding to the first-level construction level is the highest, the sponge city construction quality of the area to be measured corresponding to the second-level construction level is medium, and the sponge city construction quality of the area to be measured corresponding to the third-level construction level is the lowest;

[0169] The formulation methods of the first-level construction level, the second-level construction level, and the third-level construction level are as follows:

[0170] Count the number of abnormal areas in the area to be measured and record it as the abnormal value;

[0171] When the abnormal value is 0, the sponge city construction quality of the area to be measured is the highest at this time, and the first-level construction level is formulated;

[0172] When the abnormal value is 1, the sponge city construction quality of the area to be measured is medium at this time, and the second-level construction level is formulated;

[0173] When the abnormal value is 2 or 3 or 4, the sponge city construction quality of the area to be measured is low at this time, and the third-level construction level is formulated.

[0174] In this embodiment, by querying the construction attributes of comprehensive construction points in the area to be measured one by one and identifying sponge construction points from the comprehensive construction points, it is possible to accurately identify the objects that can predict the construction of a smart sponge city from numerous and complex comprehensive construction points, improving the accuracy of subsequent data collection. By collecting the point spacings of sponge construction points, calibrating the effective spacings, and drawing sponge areas in the area to be measured based on the area drawing criteria, it is possible to provide a reasonable time limit for the data collection of the prediction of smart sponge city construction, ensuring the rationality and accuracy of the data collected within the sponge cycle. By collecting the sponge construction data of the sponge area during the sponge cycle and inputting the collected sponge construction data into a trained machine learning model that predicts the sponge construction status, the data collection effect of multiple different dimensions in the construction of a smart sponge city can be achieved, avoiding the limitations brought by collecting single-dimensional data. Combining with the machine learning model can quickly and accurately predict the high or low status of the construction quality of a smart sponge city, avoiding the inefficiency problem brought by the prediction method of analyzing and calculating different data one by one, and also improving the prediction accuracy of the construction quality result of a smart sponge city. At the same time, by formulating the sponge construction grade of the area to be measured according to the abnormal area, the final prediction result of the construction of a smart sponge city can be reasonably output, and guiding opinions can be provided for the further development and optimization of the subsequent construction of a smart sponge city, realizing the efficient, accurate and reasonable prediction effect of the construction quality of a smart sponge city.

[0175] Embodiment 2: Please refer to Figure 2 As shown, for the parts not described in detail in this embodiment, refer to the description content of Embodiment 1. A comprehensive prediction method for the construction of a smart sponge city is provided, which is implemented based on a comprehensive prediction system for the construction of a smart sponge city and includes:

[0176] S1: Query the construction attributes of comprehensive construction points in the area to be measured one by one, and identify sponge construction points from the comprehensive construction points;

[0177] S2: Collect the point spacings of sponge construction points, calibrate the effective spacings, and draw sponge areas in the area to be measured based on the area drawing criteria;

[0178] S3: Construct a sponge cycle, and collect the sponge construction data of the sponge area during the sponge cycle. The sponge construction data includes the water level rise rate, runoff ratio, total recovery value, and ponding settlement rate;

[0179] S4: Input the collected sponge construction data into a trained machine learning model that predicts the sponge construction status, predict the sponge construction status of the area to be measured, and determine whether there is an abnormal area in the area to be measured;

[0180] S5: If there is an abnormal area, identify the abnormal area from the sponge area and formulate the sponge construction grade of the area to be measured.

[0181] Embodiment 3: Please refer to Figure 3 As shown, this embodiment discloses an electronic device, including a processor and a memory;

[0182] Among them, a computer program that can be called by the processor is stored in the memory;

[0183] The processor executes to implement the described comprehensive prediction method for the construction of a smart sponge city by calling the computer program stored in the memory.

[0184] Since the electronic device introduced in this embodiment is the electronic device used to implement the comprehensive prediction method for the construction of a smart sponge city in Embodiment 2 of the present application, based on the comprehensive prediction method for the construction of a smart sponge city introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation manners and various variations of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the method in the embodiments of the present application will not be described in detail here. As long as those skilled in the art implement the electronic device used in the comprehensive prediction method for the construction of a smart sponge city in the embodiments of the present application, it falls within the scope of protection of the present application.

[0185] Embodiment 4: Please refer to Figure 4 As shown, this embodiment discloses a computer-readable storage medium, on which a rewritable computer program is stored;

[0186] When the computer program is run, it executes to implement the described comprehensive prediction method for the construction of a smart sponge city.

[0187] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, and all should be covered by the protection scope of the present invention.

Claims

1. An integrated prediction system for the construction of a smart sponge city, characterized in that, Including: A construction point identification module, which is used to query the construction attributes of comprehensive construction points in the area to be measured one by one, and identify sponge construction points from the comprehensive construction points; A region division module, which is used to collect the point spacings of sponge construction points, calibrate the effective spacings, and draw sponge regions in the area to be measured based on the region drawing criterion. The region drawing criterion is that the adjustment amplitudes of all sponge regions are the same, and there are no overlapping or intersecting regions between any two sponge regions; A data collection module, which is used to construct a sponge cycle and collect sponge construction data of the sponge regions in the sponge cycle. The sponge construction data includes the water level rise rate, runoff ratio, total recovery value, and ponding settlement rate; A model prediction module, which is used to input the collected sponge construction data into a trained machine learning model for predicting the sponge construction state, predict the sponge construction state of the area to be measured, and determine whether there are abnormal regions in the area to be measured; A prediction report module, which is used to identify abnormal regions from the sponge regions and formulate the sponge construction grade of the area to be measured.

2. The integrated prediction system for the construction of a smart sponge city according to claim 1, characterized in that, The identification method of sponge construction points is as follows: In the electronic map, mark the locations of all sponge facilities in the area to be measured one by one, which are recorded as comprehensive construction points; Query the attribute frames of all comprehensive construction points one by one through the database, and identify the attribute semantics of the construction attributes in the attribute frames one by one through natural language processing technology; The construction attributes with attribute semantics of groundwater, runoff, utilization, and ponding are recorded as groundwater attributes, runoff attributes, utilization attributes, and ponding attributes, respectively, and the comprehensive construction points corresponding to the groundwater attributes, runoff attributes, utilization attributes, and ponding attributes are recorded as sponge construction points.

3. The integrated prediction system for the construction of a smart sponge city according to claim 2, wherein, The calibration method of the effective spacing is as follows: On the electronic map, measure the distance between any two sponge construction points one by one through the scale to obtain the point spacing; Compare the sizes of all the point spacings one by one, select the minimum value of the point spacings, and record one-third of the minimum value of the point spacings as the effective spacing.

4. The integrated prediction system for the construction of a smart sponge city according to claim 3, characterized in that, The sponge region includes a water storage region, a pipe diameter region, a recovery region, and a ponding region; The drawing methods of the water storage region, the pipe diameter region, the recovery region, and the ponding region are as follows: S2.1: On the electronic map, draw circles with four sponge construction points as the centers and the effective spacing as the radius respectively to obtain four initial regions; S2.2: Taking the preset adjustment amplitude as the adjustment standard, increase the radii of the four initial regions at the same time, drive the boundary lines of the four initial regions to expand and move outward, and observe whether there are any overlaps or intersections between the boundary lines of any two initial regions; S2.3: Repeat S2.2 until there are overlaps or intersections between the boundary lines of two initial regions, and then stop adjusting the radii of the four initial regions at the same time, and record the initial regions before the boundary lines of the two initial regions have overlaps or intersections as the drawn regions; S2.4: Query the boundary coordinates of the four drawn regions one by one through the positioning system, record the regions in the area to be measured within the four boundary coordinates as sponge regions, and record the sponge regions with the construction attributes of groundwater, runoff, utilization, and ponding of the sponge construction points as the water storage region, the pipe diameter region, the recovery region, and the ponding region, respectively.

5. The integrated prediction system for the construction of a smart sponge city according to claim 4, characterized in that, The method for collecting the water level rising rate is as follows: During the sponge cycle, query the precipitation start time and precipitation end time of B precipitation events in the water storage area one by one through timestamps, and record the duration between the precipitation start time and the precipitation end time as the precipitation duration, obtaining B precipitation durations; Detect the water levels of the underground rivers in the water storage area at the B precipitation start times and B precipitation end times one by one through a water level sensor, obtaining B starting height values and B ending height values; Subtract each of the B ending height values from the corresponding B starting height values one by one, and compare the differences with the B precipitation durations to obtain B sub-rising rates; The expression of the sub-rising rate is: where SS zb is the b-th sub-rise rate, b = 1, 2... B, GD jsb is the b-th end height value, GD qsb is the b-th starting height value, SC jsb is the b-th precipitation duration; Remove the maximum and minimum values of the sub-rising rates, accumulate the remaining B - 2 sub-rising rates and then take the average to obtain the water level rising rate; The expression of the water level rising rate is: Wherein, SS sw is the water level rising rate, and SS zc is the rising rate of the c-th sub 6. The integrated prediction system for the construction of a smart sponge city according to claim 5, characterized in that, The method for collecting the runoff ratio is as follows: During the sponge cycle, randomly mark D non-adjacent monitoring times between the B precipitation start times and B precipitation end times; Detect the water flow velocities of the water in the drainage pipe in the pipe diameter area at the D monitoring times one by one through a flow velocity sensor, obtaining D water flow velocities; Query the cross-sectional area of the drainage pipe through a technical parameter table, and compare the cross-sectional area of the drainage pipe with the average value of the D water flow velocities to obtain the runoff ratio; The expression of the runoff ratio is: Where, JL bz is the runoff ratio, JM mj is the cross-sectional area of the drainage pipe, SL lsd is the flow velocity of the d-th water flow.

7. The integrated prediction system for the construction of a smart sponge city according to claim 6, characterized in that, The method for collecting the ponding settlement rate is as follows: Real-time detect the ponding depth of the ponding on the ground road in the ponding area, and record the ponding with a ponding depth greater than the lower limit of ponding as effective ponding; During the sponge cycle, mark the positions of the effective ponding on the ground road one by one, obtaining E ponding positions, and detect the ponding depths of the E ponding positions one by one, obtaining E initial depth values; After a preset settlement duration, detect the ponding depths of the E ponding positions one by one, obtaining E real-time depth values; Subtract each of the E initial depth values from the corresponding E real-time depth values respectively, accumulate the differences and then take the average to obtain the ponding settlement rate; The expression of the ponding settlement rate is: Where, JS cj is the ponding settlement rate, SD cse is the e-th initial depth value, SD sse is the e-th real-time depth value.

8. The integrated prediction system for the construction of a smart sponge city according to claim 7, wherein The sponge construction state includes a high construction state and a low construction state; The training method of the machine learning model is as follows: Pre-collect multiple groups of sponge construction data of the area to be measured in the high construction state and the low construction state; Mark each group of sponge construction data as training features, convert the sponge construction state of each group of training features into digital annotations, convert the high construction state into 0, and convert the low construction state into 1; Divide the annotated training features into a training set and a test set, use the training set to train the machine learning model, use the test set to test the machine learning model, preset an error threshold, and when the mean value of the prediction errors of all training features in the test set is less than the error threshold, obtain the machine learning model; The determination method for whether there is an abnormal area is as follows: When the output of the machine learning model is 0, the sponge construction state of the area to be measured is in the high construction state, and it is determined that there is no abnormal area in the area to be measured; When the output of the machine learning model is 1, the sponge construction state of the area to be measured is in the low construction state, and it is determined that there is an abnormal area in the area to be measured.

9. The integrated prediction system for the construction of a smart sponge city according to claim 8, wherein, The identification method for the abnormal area is: Anomaly identification is performed on the water level rise rate, runoff ratio, total recovery value, and ponding settlement rate respectively; When the water level rise rate is less than the water level calibration value, the water storage area is recorded as an abnormal area; When the runoff ratio is greater than the runoff calibration value, the pipe diameter area is recorded as an abnormal area; When the total recovery value is less than the recovery calibration value, the recovery area is recorded as an abnormal area; When the ponding settlement rate is less than the ponding calibration value, the ponding area is recorded as an abnormal area; The sponge construction levels include the first-level construction level, the second-level construction level, and the third-level construction level. The formulation methods for the first-level construction level, the second-level construction level, and the third-level construction level are as follows: Count the number of abnormal areas in the area to be measured, and record it as the abnormal value; When the abnormal value is 0, the first-level construction level is formulated; When the abnormal value is 1, the second-level construction level is formulated; When the abnormal value is 2 or 3 or 4, the third-level construction level is formulated.

10. A comprehensive prediction method for the construction of a smart sponge city, implemented based on the comprehensive prediction system for the construction of a smart sponge city according to any one of claims 1-9, characterized in that, Including: S1: Query the construction attributes of the comprehensive construction points in the area to be measured one by one, and identify the sponge construction points from the comprehensive construction points; S2: Collect the point spacings of the sponge construction points, calibrate the effective spacings, and draw the sponge areas in the area to be measured based on the regional drawing criteria; S3: Construct a sponge cycle, and collect the sponge construction data of the sponge areas in the sponge cycle. The sponge construction data includes the water level rise rate, runoff ratio, total recovery value, and ponding settlement rate; S4: Input the collected sponge construction data into the trained machine learning model that predicts the sponge construction status, predict the sponge construction status of the area to be measured, and determine whether there are abnormal areas in the area to be measured; S5: If there are abnormal areas, identify the abnormal areas from the sponge areas and formulate the sponge construction level of the area to be measured.

Images