Drainage device of rainwater infiltration purification system and automatic regulation and control method

Abstract

The invention discloses a drainage device of a rainwater infiltration purification system and an automatic regulation and control method. The drainage device comprises an inner pipe and an outer pipe which are arranged in a nested mode. The inner pipe is rotationally arranged in the outer pipe, and the inner pipe is connected with a rotating device; the rotating device is used for receiving an external rotating control instruction and controlling the inner pipe to rotate to a set position relative to the outer pipe; the inner pipe is provided with a plurality of through holes with set aperture sizes; the outer pipe comprises a semi-surrounded pipe body section; the semi-surrounded pipe body section only comprises a pipe body part with a set area range at the bottom, and the rest part of the semi-surrounded pipe body section is opened and exposes the inner pipe; and constructing a CART decision tree, and controlling the rotation of the inner pipe of the perforated drain pipe of the rainwater infiltration purification system according to rainfall conditions and facility operation conditions, so that the perforated holes with corresponding hole diameters are exposed out of the top of the outer pipe. According to the invention, the adaptability of the rainwater infiltration purification system can be obviously improved, and rapid discharge under large flow and low pollution load in the middle and later periods of rainfall and effective retention purification under small flow and high pollution load in the early period of rainfall are realized.

Description

Technical Field

[0001] The invention relates to the technical field of rainwater infiltration purification system control, and in particular to a drainage device and an automatic control method of a rainwater infiltration purification system. Background Art

[0002] With climate change and accelerated urbanization, the frequency of extreme rainfall and the rate of ground hardening have increased significantly, and the problem of waterlogging has become increasingly prominent, seriously threatening urban safety and economic development. As typical low-impact development measures, rainwater infiltration purification systems such as biological retention ponds and rain gardens are widely used in sponge city construction, and have multiple functions such as controlling the total amount of runoff, peak flow, and runoff pollution. However, it is difficult for traditional rainwater infiltration purification systems to achieve a balance between runoff volume and runoff pollution control under complex and changeable rainfall conditions. As an important design parameter, the emptying time is mainly determined based on specific needs or local historical rainfall data and soil medium layer conditions. The initial design is not adaptable enough and cannot take into account the rapid discharge under large flow and low pollution load in the middle and late stages of rainfall and the retention and purification under small flow and high pollution load in the early stages of rainfall.

[0003] The drainage device of the rainwater infiltration purification system, especially the perforated drain pipe, has an aperture that is a key factor affecting the emptying time of the facility. Therefore, how to provide a perforated drain pipe and automatic control method for the rainwater infiltration purification system that can adapt to actual dynamic rainfall is a technical problem that needs to be urgently solved by technical personnel in this field. Summary of the invention

[0004] In view of the above research status and existing problems, the present invention provides a drainage device and an automatic control method for a rainwater infiltration purification system. The present invention adjusts the aperture size of the perforated drainage pipe in real time according to rainfall conditions and facility operation conditions, thereby ensuring the targeted operation of the rainwater infiltration purification system.

[0005] The present invention provides a perforated drainage pipe for a rainwater infiltration purification system, comprising: an inner pipe and an outer pipe which are nested; the inner pipe is rotatably arranged in the outer pipe, and the inner pipe is connected to a rotating device;

[0006] The rotating device is used to receive an external rotation control instruction and control the inner tube to rotate to a set position relative to the outer tube;

[0007] The inner tube is provided with a plurality of perforations with set aperture sizes;

[0008] The outer tube includes a semi-enclosed tube body section; the semi-enclosed tube body section only includes a tube body portion within a set bottom area, and the rest of the semi-enclosed tube body section is open to expose the inner tube.

[0009] Preferably, the inner tube has a plurality of through holes with different apertures; the through holes with different apertures are distributed along the circumference of the inner tube, and the through holes with the same aperture are arranged and distributed along the axial direction of the inner tube.

[0010] Preferably, the semi-enclosed tubular body section only includes the bottom 1 / 2 of the tubular body portion, and the remaining 1 / 2 of the semi-enclosed tubular body section is open to expose the inner tube.

[0011] Preferably, the rotating device includes a rotation controller, an electric shaft, a connecting rod and a pipe clamp; the rotation controller is connected to the electric shaft, and is used to provide a rotational torque and receive external braking control instructions to control the electric shaft to rotate a set angle, and the electric shaft is vertically connected to the midpoint of the connecting rod; both ends of the connecting rod are connected to the pipe clamp, and the connecting rod is installed along the diameter of the inner tube, and is clamped to one side end of the inner tube through the pipe clamp.

[0012] The present invention also provides a method for automatically controlling a perforated drainage pipe of a rainwater infiltration purification system according to a drainage device of the rainwater infiltration purification system, comprising the following steps:

[0013] S1. Construct a CART decision tree, wherein the input of the CART decision tree is urban rainfall data and rainwater infiltration purification system water quality data, and the output is a classification result of whether flooding occurs;

[0014] S2. Real-time collection of urban rainfall data and water quality data of the rainwater infiltration purification system, input into the CART decision tree, and send a rotation control instruction to the rotation device according to the output result of the CART decision tree, so as to control the rotation of the inner tube of the perforated drainage pipe of the rainwater infiltration purification system, so that the perforations of corresponding aperture size are exposed at the top of the outer tube.

[0015] Preferably, S1 includes a CART decision tree training step:

[0016] S11: obtaining input features and corresponding classification results of the CART decision tree, and dividing them into a training set, a validation set, and a test set, wherein the input features include: historical rainfall data of the city and its corresponding water quality data of the rainwater infiltration purification system;

[0017] S12: Calculate the characteristic Gini index corresponding to the training set, and determine the nodes and node characteristic thresholds of the CART decision tree according to the characteristic Gini index, so as to construct a primary structure of the CART decision tree;

[0018] S13: Calculate the loss function of the subtree corresponding to the input feature in the verification set before and after pruning the non-leaf nodes in the primary structure of the CART decision tree, determine the non-leaf nodes to be pruned with the goal of minimizing the loss function, and repeat multiple rounds of pruning operations to obtain multiple subtree structures;

[0019] S14: cross-validate the multiple subtree structures using the input features of the test set, and determine the optimal tree according to the accuracy of the classification result, which is the final CART decision tree model.

[0020] Preferably, the input features in S11 are a data set, including: rainfall intensity, rainfall duration, air pollution index, pre-rain dry period, rainwater infiltration purification system operation time and comprehensive pollution index.

[0021] Preferably, the S12 comprises the following steps:

[0022] S121: Using the feature value corresponding to the minimum value of the feature Gini index in the training set as the threshold of the root node of the classification decision tree;

[0023] S122: Using the threshold of the root node as a standard, the training set is divided into two parts, D1 and D2, and then the eigenvalue corresponding to the minimum value of the corresponding characteristic Gini index is repeatedly calculated for D1 and D2 to determine the threshold of the secondary node;

[0024] S123: Using the threshold of the secondary node as a standard, repeatedly perform the calculation operation of the eigenvalue corresponding to the minimum value of the characteristic Gini index until no eigenvalue can be assigned, and construct the primary structure of the CART decision tree.

[0025] Preferably, the S13 comprises the following steps:

[0026] S131: Construct the loss function of the subtree corresponding to the input feature in the validation set before and after non-leaf node pruning:

[0027] The loss function formula for the unpruned subtree of the cotyledon of the node with eigenvalue t is:

[0028] C α (Tt) = C(Tt) + α|Tt|;

[0029] After the cotyledons of the node with eigenvalue t are pruned, |Tt|=1, and the subtree loss function formula is:

[0030] C α (t) = C (t) + α;

[0031] Where T is any subtree, C(*) is the error cost of the validation set, |*| is the number of nodes in the subtree, and α is the regularization parameter;

[0032] S132: Make the subtree loss function equal before and after pruning, and get the expression of α:

[0033]

[0034] Calculate the α values ​​of all non-leaf nodes, compare and obtain the non-leaf nodes corresponding to the minimum α value, and perform pruning operations on them to obtain the first subtree structure;

[0035] S133: Repeatedly calculate the corresponding α values ​​for the non-leaf nodes of the pruned decision tree, compare and obtain the non-leaf nodes corresponding to the minimum α value, and perform pruning operations on them to obtain a second subtree structure, until there are no non-leaf nodes in the pruned decision tree that can perform pruning operations, and obtain multiple subtree structures.

[0036] The present invention also provides an automatic control system for drainage devices of a rainwater infiltration purification system, comprising: a drainage pipe, a controller, a memory, and a computer program stored in the memory and executable on the controller; wherein:

[0037] The drainage pipe structure adopts the perforated drainage pipe of the rainwater infiltration purification system;

[0038] When the controller executes the computer program, the steps of the method for automatically controlling the perforated drain pipe of a rainwater infiltration purification system are implemented.

[0039] The drainage device and automatic control method of the rainwater infiltration purification system proposed in the present invention are intended to solve the problem that the existing rainwater infiltration purification system cannot be adjusted in time to fully exert its function of intercepting runoff and improving water quality when facing a complex and changeable climate environment. Compared with the existing technology, it has the following beneficial effects:

[0040] The present invention is based on the CART (Classification and Regression Tree) model, which can respond to climate changes in real time and automatically adjust the aperture size of the perforated drainage pipe according to real-time rainfall and water quality data. By adjusting the aperture, the permeability of the rainwater infiltration purification system can be increased when the rainfall is large, and the peak runoff flow can be reduced; when the rainfall is small, the emptying time can be extended and the denitrification time can be increased, thereby effectively reducing initial rainwater pollution. The present invention not only significantly improves the synchronous and efficient control of the runoff volume and runoff pollution of the rainwater infiltration purification system, but also has important reference value for the drainage structure design of various source rainwater control facilities such as rainwater wetlands and green roofs. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] 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 prior art descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without creative work.

[0042] Figure 1 Schematic diagram of the cross-sectional structure of a perforated drain pipe of a rainwater infiltration purification system provided by an embodiment of the present invention;

[0043] Figure 2 It is a schematic diagram of the inner tube structure of the perforated drainage pipe of the rainwater infiltration purification system provided by an embodiment of the present invention;

[0044] Figure 3 It is a schematic structural diagram of a rotating device of a perforated drain pipe of a rainwater infiltration purification system provided by an embodiment of the present invention;

[0045] Figure 4 It is a flow chart of an automatic control method of a perforated drainage pipe of a rainwater infiltration purification system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] like Figure 1 As shown, a perforated drain pipe of a rainwater infiltration purification system provided in the first aspect of an embodiment of the present invention comprises: an inner pipe 1 and an outer pipe 2 which are nested; the inner pipe 1 is rotatably arranged in the outer pipe 2, and a rotating device 4 is arranged between the inner pipe 1 and the outer pipe 2; the rotating device 4 is used to receive external rotation control instructions and control the inner pipe 1 to rotate to a set position relative to the outer pipe 2; a plurality of perforations 5 with set aperture sizes are provided on the pipe wall of the inner pipe 1; the outer pipe 2 comprises a semi-enclosed pipe body section; the semi-enclosed pipe body section only includes the pipe body portion within the set area range at the bottom, and the rest of the semi-enclosed pipe body section is open and exposes the inner pipe 1.

[0048] It should be noted that the inner tube 1 is used to classify the apertures of the perforations 5. The outer tube 2 is kept in a certain ratio, wherein the lower portion of the outer tube 2 remains intact, while the upper portion of the outer tube 2 is partially removed, thereby exposing the inner tube 1. The lower portion and the upper portion mentioned here are upper and lower relationships distinguished according to the direction of gravity.

[0049] In one embodiment, Figure 2 As shown, the inner tube 1 has a plurality of through holes 5 with different apertures; the through holes 5 with different apertures are distributed along the circumference of the inner tube 1 , and the through holes 5 with the same aperture are arranged and distributed along the axial direction of the inner tube 1 .

[0050] It is understandable that the apertures of the perforations 5 in the same row along the axis of the pipe wall of the inner pipe 1 are the same, and the apertures of the perforations 5 in different rows along the axis may be different, so as to ensure that the apertures of the rows of perforations 5 leaking out of the top of the outer pipe by rotating the inner pipe meet the drainage requirements of the rainwater infiltration purification system.

[0051] The aperture in this embodiment can be divided into large aperture and small aperture. By controlling the rotation of the inner tube 1, the large aperture or small aperture perforation 5 is exposed at the top of the outer tube 2, thereby achieving real-time and effective response to different climate changes.

[0052] In one embodiment, the semi-enclosed tubular body section only includes the bottom 1 / 2 of the tubular body portion, and the remaining 1 / 2 of the semi-enclosed tubular body section is open to expose the inner tube 1 .

[0053] In one embodiment, the rotating device includes a rotating controller 6, an electric shaft 7, a connecting rod 8 and a pipe clamp 3; the rotating controller 6 is connected to the electric shaft 7, and is used to provide a torque and receive an external braking control command to control the electric shaft 7 to rotate a set angle, and the electric shaft 7 is vertically connected to the midpoint of the connecting rod 8; both ends of the connecting rod 8 are connected to the pipe clamp 3, and the connecting rod 8 is installed along the diameter of the inner pipe, and is clamped to one side end of the inner pipe through the pipe clamp 3.

[0054] In this embodiment, the rotation controller 6 has a built-in rotor and a signal receiver, which can be used to receive external information and drive the electric shaft 7 to rotate. A stepper motor can also be used.

[0055] The pipe clamp 3 adopts a hinge structure, which includes a pin and two hinges; the pin is fixedly connected to the end of the connecting rod 8, and the fixing direction is perpendicular to the rod body direction of the connecting rod 8. The two hinges are respectively inserted into the inner wall and outer wall of the inner tube, and the inner tube is fixedly clamped by the expansion screws of the round hole, so that it rotates with the rotation of the electric shaft 7, thereby realizing the change of the perforation diameter exposed at the top of the outer tube.

[0056] In this embodiment, a support column 4 is provided in the bottom gap between the inner tube and the outer tube. The support column 4 can rotate flexibly relative to the inner tube and the outer tube. The central axis of the support column 4 is parallel to the central axis of the inner tube and the outer tube, which can support the inner tube and facilitate the rotation of the inner tube in the outer tube.

[0057] It should also be noted that the gap between the inner tube and the outer tube is much smaller than the minimum perforation diameter.

[0058] The second aspect of the embodiment of the present invention further provides a method for automatically controlling a rainwater infiltration purification system perforated drainage pipe of a drainage device of a rainwater infiltration purification system according to the first aspect of the embodiment, such as Figure 3 As shown, the following steps are included:

[0059] S1. Construct a CART decision tree. The input of the CART decision tree is urban rainfall data and rainwater infiltration purification system water quality data, and the output is the classification result of whether flooding occurs;

[0060] S2. Collect urban rainfall data and rainwater infiltration purification system water quality data in real time, input them into the CART decision tree, send a rotation control instruction to the rotation device 4 according to the output result of the CART decision tree, control the rotation of the inner tube 1 of the perforated drainage pipe of the rainwater infiltration purification system, so that the perforations 5 of the corresponding aperture size are exposed at the top of the outer tube 2.

[0061] During the specific execution, the output results are traversed. If the result is that flooding will occur, that is, y1, the inner tube of the perforated drainage pipe is controlled to have a large aperture and turned upward, which reduces the emptying time and increases the permeability of the rainwater infiltration purification system, thereby reducing the peak flow of runoff. If the output is that flooding will not occur, that is, y2, the inner tube of the perforated drainage pipe is controlled to have a small aperture and turned upward, which increases the emptying time of the facility and the denitrification time, thereby reducing the initial rainwater pollution.

[0062] In one embodiment, S1 includes a CART decision tree training step, and the water quantity and water quality analysis of the collected real-time data is performed based on the trained model. If the rainfall is large, the rainwater infiltration purification system plays a role in reducing the total amount of runoff peak, and vice versa, it plays a role in purifying the initial rainfall pollution. The process is achieved by adjusting the aperture of the perforated drainage pipe. The training step includes:

[0063] S11: Obtain the input features and corresponding classification results of the CART decision tree, and divide them into a training set, a validation set, and a test set. The input features include: historical rainfall data of the city and its corresponding water quality data of the rainwater infiltration purification system;

[0064] S12: Calculate the characteristic Gini index corresponding to the training set, and determine the nodes and node characteristic thresholds of the CART decision tree according to the characteristic Gini index, and construct the primary structure of the CART decision tree;

[0065] S13: Calculate the loss function of the subtree corresponding to the input feature in the verification set before and after pruning the non-leaf nodes in the primary structure of the CART decision tree, determine the non-leaf nodes to be pruned with the goal of minimizing the loss function, and repeat multiple rounds of pruning operations to obtain multiple subtree structures;

[0066] S14: Use the input features of the test set to cross-validate multiple subtree structures, and determine the optimal tree based on the accuracy of the classification results, which is the final CART decision tree model.

[0067] In this embodiment, the model iteration step is also included: in view of the current complex climate, each rainfall situation is compared with the judgment result of the CART decision tree model, and new data is continuously added to the original data set. At the same time, the accuracy of the model results and the actual flood situation is determined. If the accuracy of the CART decision tree model is less than 70%, the newly added real-time input feature data used for the prediction of the CART decision tree model is added to the training data set, and the above training steps are repeated. Through model iteration, repeated training and pruning using updated data sets, the system can continuously improve the prediction accuracy and ensure the efficient operation of the rainwater infiltration purification system.

[0068] In one embodiment, the input features in S11 are a data set including: rainfall intensity, rainfall duration, air pollution index, dry period before rain, operation time of rainwater infiltration purification system and comprehensive pollution index.

[0069] In the specific implementation, the urban rainfall and rainwater infiltration purification system water quality data are collected, and the code "1" indicates the occurrence of floods, and the code "0" indicates the absence of floods. The water quality data includes ammonia nitrogen, nitrate nitrogen, total phosphorus and cod content, and the comprehensive pollution index method is used to quantify the water quality.

[0070] The calculation formula of comprehensive pollution index is as follows:

[0071]

[0072] Where K is the comprehensive pollution index, c i is the concentration of pollutants such as ammonia nitrogen and COD, c oi It is the highest standard for various pollutants discharged into water bodies. k It is the unified maximum allowable standard for all pollutants discharged into water.

[0073] The data input format is {X1, X2, X3, X4, X5, X6, Y} = {rainfall intensity, rainfall duration, air pollution index, dry period before rain, operation time of rainwater infiltration purification system, comprehensive pollution index, flood situation}.

[0074] In one embodiment, S12 includes the following steps:

[0075] S121: taking the feature value corresponding to the minimum value of the feature Gini index in the training set as the threshold of the root node of the classification decision tree;

[0076] S122: Using the threshold of the root node as a standard, the training set is divided into two parts, D1 and D2, and the eigenvalue corresponding to the minimum value of the corresponding characteristic Gini index is repeatedly calculated for D1 and D2 to determine the threshold of the secondary node;

[0077] S123: Using the threshold of the secondary node as a standard, repeatedly perform the calculation operation of the eigenvalue corresponding to the minimum value of the characteristic Gini index until no eigenvalue can be assigned, and construct the primary structure of the CART decision tree.

[0078] In the specific implementation, the Gini index of the training set is first initialized to evaluate the training set, which is also required for later calculations. The Gini index Gini(D) and feature Gini index Gini(D, a) of the initialized training set are as follows:

[0079] The formula for Gini(D) is:

[0080] Gini(D)=1-∑ i p i 2 ;

[0081] This formula represents the purity of the training set. The smaller Gini(D), the higher the purity of the training set. The data pi represents the probability of flood conditions accounting for all possible target features. In particular, when there is only one possible target feature, the Gini(D) of the training set is 0.

[0082] The formula of characteristic Gini index Gini(D, a) is:

[0083]

[0084] Gini(D, a) represents the uncertainty of set D after segmentation by feature a;

[0085] a represents a certain characteristic value of the training set. For each characteristic (such as rainfall intensity, rainfall duration, etc.), the corresponding characteristic Gini index is calculated;

[0086] D v represents the set of values ​​v under feature a;

[0087] It represents the ratio of v values ​​with eigenvalue a to all possible values;

[0088] Gini (D v ) represents the Gini index of the target feature with the value v of the feature value a.

[0089] The smaller Gini(D, a), the higher the certainty. Find the feature with the smallest Gini index and its corresponding value among all the features. This will be the best split point of the decision tree. Finally, the root node of the classification decision tree is selected with the smallest Gini(D, a) value.

[0090] The characteristic Gini index Gini (D, a) formula is used to calculate the characteristic Gini index of flooding conditions for each input feature {rainfall intensity, rainfall duration, air pollution index, dry period before rain, rainwater infiltration purification system operation time, comprehensive pollution index}, and the input features are sorted, including:

[0091] After the root node is determined, the threshold of the root node is used as the standard to divide the data set into two parts, D1 and D2. The characteristic Gini index of D1 and D2 is then calculated repeatedly to continue selecting nodes. The above operation is repeated until no characteristic value can be assigned. At this point, the decision tree is established.

[0092] In one embodiment, in order to improve the generalization ability and reliability of the model, the decision tree model is pruned using the validation set to further optimize the arrangement of nodes. S13 includes the following steps:

[0093] S131: Construct the loss function of the subtree corresponding to the input feature in the validation set before and after non-leaf node pruning:

[0094] During the pruning process, for any subtree T, the loss function formula of the subtree whose cotyledons of the node with eigenvalue t are not pruned is:

[0095] C α (Tt) = C(Tt) + α|Tt|;

[0096] After the cotyledons of the node with eigenvalue t are pruned, they become leaf nodes, |Tt|=1, and the subtree loss function formula is:

[0097] C α (t) = C (t) + α;

[0098] Where T is an arbitrary subtree, C(*) is the error cost of the validation set, |*| is the number of nodes in the subtree, and α is the regularization parameter, which is used to weigh the degree of fit of the training data and the complexity of the model.

[0099] S132: The purpose of pruning is to make C α (T) is as small as possible to improve the prediction effect of the model on the validation data set, but C α (T) is constrained by C(T) and α|T|. There is a contradiction between the two during the pruning process. The smaller C(T), the more complex the decision classification tree is, the more nodes there are, and more nodes lead to a larger value of |T|. Therefore, the key to the problem of pruning is to find a balance between the two, that is, to find a suitable α value so that within a certain α

[0100] C α (T) is the smallest. In the actual operation process, we first find the critical point of the two, that is, we use the equality of the right side of the unpruned and pruned equations to find α at this time. The formula is as follows:

[0101] C(Tt)+α|Tt|=C(t)+α;

[0102] Make the subtree loss function equal before and after pruning, and get the expression of α:

[0103]

[0104] Let g(t) = α, the formula becomes:

[0105]

[0106] Calculate the α values ​​of all non-leaf nodes, compare and obtain the non-leaf nodes corresponding to the minimum α value, and perform pruning operations on them to obtain the first subtree structure;

[0107] S133: Repeatedly calculate the corresponding α values ​​for the non-leaf nodes of the pruned decision tree, compare and obtain the non-leaf nodes corresponding to the minimum α value, and perform pruning operations on them to obtain a second subtree structure, until there are no non-leaf nodes in the pruned decision tree that can perform pruning operations, and obtain multiple subtree structures.

[0108] In the specific implementation, for a decision tree with three nodes, let the nodes be t0, t1 and t2, and the unpruned decision tree be T0.

[0109] Calculate g(t0), g(t1) and g(t2) for the T0 tree respectively and compare their minimum values. Let g(t0) be the minimum and prune the t0 node, that is, α0=g(t0).

[0110] The pruned decision tree is T1, and the remaining nodes are t1 and t2. Repeat the above calculation to generate T2 and T3, and finally get four mutually nested subtrees of T0, T1, T2, and T3. Use the test set to cross-validate the four subtrees, and determine the optimal tree based on the final accuracy. This decision tree is the final CART decision tree model.

[0111] For a CART decision tree whose final nodes are arranged as {X1, X2, X3}, the corresponding threshold is {x1, x2, x3}, and the target feature is {y1, y2}, the final output result is:

[0112] {'X1':{'≤x1':'y1','>x1':{'X2':{'≤x2':'y1','>x2':{'X3':{'≤x3':'y1','>x3':'y2'}}}}}}.

[0113] The third aspect of the embodiment of the present invention further provides a rainwater infiltration purification system perforated drain pipe automatic control system, comprising: a drain pipe, a controller, a memory, and a computer program stored in the memory and executable on the controller; wherein:

[0114] The drainage pipe structure adopts the rainwater infiltration purification system perforated drainage pipe of the first aspect of the embodiment;

[0115] When the controller executes the computer program, the steps of the method for automatically controlling the perforated drain pipe of a rainwater infiltration purification system according to the second aspect of the embodiment are implemented.

[0116] In one embodiment, a rain gauge is pre-installed on the top of the rainwater infiltration purification system to monitor the urban rainfall in real time. At the same time, a water quality collection device is installed at the end of the perforated drainage pipe of the rainwater infiltration purification system to collect water quality data of the rainwater infiltration purification system in real time.

[0117] In one embodiment, a control terminal is also included: the control terminal is used to visualize rainfall data and the pipe wall and perforation aperture of the perforated drainage pipe of the rainwater infiltration purification system. At the same time, manual control can be performed on the control terminal to handle emergency situations, including manually sending rotation control instructions and braking control instructions.

[0118] The above is a detailed introduction to the perforated drain pipe and automatic control method and system of the rainwater infiltration purification system provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0119] In this article, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

Claims

1. A drainage device for a rainwater infiltration purification system, characterized in that: include: An inner tube and an outer tube are nested; the inner tube is rotatably arranged inside the outer tube, and the inner tube is connected to a rotating device; The rotating device is used to receive an external rotation control instruction and control the inner tube to rotate to a set position relative to the outer tube; The inner tube is provided with a plurality of perforations with set aperture sizes; The outer tube includes a semi-enclosed tube body section; the semi-enclosed tube body section only includes a tube body portion within a set bottom area, and the rest of the semi-enclosed tube body section is open to expose the inner tube.

2. A drainage device for a rainwater infiltration purification system according to claim 1, characterized in that: The inner tube is provided with a plurality of through holes with different apertures; the through holes with different apertures are distributed along the circumference of the inner tube, and the through holes with the same aperture are arranged and distributed along the axial direction of the inner tube.

3. A drainage device for a rainwater infiltration purification system according to claim 1, characterized in that: The semi-enclosed tubular body section only includes the bottom 1 / 2 of the tubular body portion, and the remaining 1 / 2 of the semi-enclosed tubular body section is open to expose the inner tube.

4. A drainage device for a rainwater infiltration purification system according to claim 1, characterized in that: The rotating device includes a rotation controller, an electric shaft, a connecting rod and a pipe clamp; the rotation controller is connected to the electric shaft, and is used to provide a rotational torque and receive an external braking control instruction to control the electric shaft to rotate a set angle, and the electric shaft is vertically connected to the midpoint of the connecting rod; both ends of the connecting rod are connected to the pipe clamp, and the connecting rod is installed along the diameter of the inner tube and is clamped to one side end of the inner tube through the pipe clamp.

5. A method for automatically controlling a drainage device of a rainwater infiltration purification system according to any one of claims 1 to 4, characterized in that: The steps include: S1. Construct a CART decision tree, wherein the input of the CART decision tree is urban rainfall data and rainwater infiltration purification system water quality data, and the output is a classification result of whether flooding occurs; S2. Collect urban rainfall data and water quality data of the rainwater infiltration purification system in real time, input them into the CART decision tree, send a rotation control instruction to the rotating device according to the output result of the CART decision tree, control the rotation of the inner tube of the perforated drainage pipe of the rainwater infiltration purification system, so that the perforations of corresponding aperture sizes are exposed at the top of the outer tube.

6. The automatic control method of the drainage device of the rainwater infiltration purification system according to claim 5 is characterized in that: S1 includes the training steps of the CART decision tree: S11: obtaining input features and corresponding classification results of the CART decision tree, and dividing them into a training set, a validation set, and a test set, wherein the input features include: historical rainfall data of the city and its corresponding water quality data of the rainwater infiltration purification system; S12: Calculate the characteristic Gini index corresponding to the training set, and determine the nodes and node characteristic thresholds of the CART decision tree according to the characteristic Gini index, so as to construct a primary structure of the CART decision tree; S13: Calculate the loss function of the subtree corresponding to the input feature in the verification set before and after pruning the non-leaf nodes in the primary structure of the CART decision tree, determine the non-leaf nodes to be pruned with the goal of minimizing the loss function, and repeat multiple rounds of pruning operations to obtain multiple subtree structures; S14: cross-validate the multiple subtree structures using the input features of the test set, and determine the optimal tree according to the accuracy of the classification result, which is the final CART decision tree model.

7. The automatic control method of the drainage device of the rainwater infiltration purification system according to claim 5 is characterized in that: The input features in S11 are a data set, including: rainfall intensity, rainfall duration, air pollution index, dry period before rain, operation time of rainwater infiltration purification system and comprehensive pollution index.

8. The automatic control method of the drainage device of the rainwater infiltration purification system according to claim 5 is characterized in that: The S12 comprises the following steps: S121: Using the feature value corresponding to the minimum value of the feature Gini index in the training set as the threshold of the root node of the classification decision tree; S122: Using the threshold of the root node as a standard, the training set is divided into two parts, D1 and D2, and the eigenvalue corresponding to the minimum value of the corresponding characteristic Gini index is repeatedly calculated for D1 and D2 to determine the threshold of the secondary node; S123: Using the threshold of the secondary node as a standard, repeatedly perform the calculation operation of the eigenvalue corresponding to the minimum value of the characteristic Gini index until no eigenvalue can be assigned, and construct the primary structure of the CART decision tree.

9. The automatic control method of the drainage device of the rainwater infiltration purification system according to claim 5 is characterized in that: The S13 comprises the following steps: S131: Construct the loss function of the subtree corresponding to the input feature in the validation set before and after non-leaf node pruning: The loss function formula for the unpruned subtree of the cotyledon of the node with eigenvalue t is: C a (Tt)=C(Tt)+α|Tt|; After the cotyledons of the node with eigenvalue t are pruned, |Tt|=1, and the subtree loss function formula is: C α (t)=C(t)+α; Where T is any subtree, C(*) is the error cost of the validation set, |*| is the number of nodes in the subtree, and α is the regularization parameter; S132: Make the subtree loss function equal before and after pruning, and get the expression of α: Calculate the α values ​​of all non-leaf nodes, compare and obtain the non-leaf nodes corresponding to the minimum α value, and perform pruning operations on them to obtain the first subtree structure; S133: Repeatedly calculate the corresponding α values ​​for the non-leaf nodes of the pruned decision tree, compare and obtain the non-leaf nodes corresponding to the minimum α value, and perform pruning operations on them to obtain a second subtree structure, until there are no non-leaf nodes in the pruned decision tree that can perform pruning operations, and obtain multiple subtree structures.

10. An automatic control system for drainage device of rainwater infiltration purification system, characterized in that: include: A drain pipe, a controller, a memory, and a computer program stored in the memory and executable on the controller; wherein, The drainage pipe structure adopts the rainwater infiltration purification system drainage device described in any one of claims 1 to 4; When the controller executes the computer program, the steps of the automatic control method of the drainage device of the rainwater infiltration purification system as described in any one of claims 5-9 are implemented.

Images