Sponge city rainwater treatment system based on big data

Abstract

The invention discloses a sponge city rainwater treatment system based on big data, which comprises a data acquisition module, and relates to the technical field of sponge city rainwater treatment.In a rainwater purification module, an automatic purifying agent adding program is started, so that the rainwater purification efficiency is improved, and rainwater is accelerated to flow into the rainwater purification module; a water storage space is vacated for a rainwater regulation and storage facility; a drainage valve of the drainage pipe is in a semi-open state, rainwater overflow in the rainwater regulation and storage facility caused by the fact that a large amount of initial rainwater rapidly flows into the rainwater regulation and storage facility is avoided, and the rainwater is dredged in order; when the time is within the overflow risk time period and reaches the dynamic early warning threshold value, the rainwater is further dredged through the fact that the drainage valve of the drainage pipe is in a full-open state and the operation power of the pump station is increased; at the moment, the rainwater regulation and storage facility conducts intermittent drainage, the time needed for reducing the water stored in the rainwater regulation and storage facility to be below the early warning threshold value is shortened, and the urban waterlogging risk is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sponge city rainwater treatment, and particularly relates to a sponge city rainwater treatment system based on big data. BACKGROUND

[0002] A sponge city is a city that can adapt to environmental changes and cope with natural disasters, etc. in terms of good "elasticity". When it rains, it can absorb, store, infiltrate, and purify water; when needed, it can "release" the stored water for use, such as for urban greening irrigation, road washing, and landscape water replenishment. The stored rainwater can be replenished to groundwater through natural infiltration or artificial recharge. During rainfall, rainwater storage facilities are used to collect rainwater, allowing sufficient time and space for rainwater to infiltrate underground to replenish groundwater resources, which helps to maintain the stability of the groundwater level and prevent geological problems such as land subsidence.

[0003] Publication No. CN111115907B relates to a sponge city rainwater treatment method, which separates the treatment of initial rainwater and later rainwater, uses different purification processes according to the different impurity contents of rainwater at different times, has better purification effect, lower treatment cost, and improves economic benefit;

[0004] However, the above-mentioned application still has the following problems: conventional rainwater treatment cannot effectively alleviate the storage and pressure when waterlogging occurs, and exacerbates the risk of urban waterlogging, such as the efficiency improvement treatment of rainwater purification and the opening of the flood discharge channel. When the rainwater storage facilities cannot store the overflow or the rainwater storage facilities store the overflow warning, the rainwater storage facilities store the overflow warning. At this time, due to the large amount of water entering the rainwater storage facilities, due to the lack of pretreatment scheme, it takes longer to reduce the water storage in the rainwater storage facilities to below the warning line, thus exacerbating the risk of urban waterlogging. SUMMARY

[0005] To solve the technical problems in the background art, the present application provides a sponge city rainwater treatment system based on big data.

[0006] The sponge city rainwater treatment system based on big data provided by the present application comprises:

[0007] The data acquisition module acquires rainwater-related data by setting sensors and monitors;

[0008] The data transmission module transmits the collected rainwater-related data to the data storage module using wired or wireless communication technology;

[0009] The data storage module stores rainwater-related data using a distributed storage architecture;

[0010] Data processing module: set a period, divide a period into multiple time periods, standardize the rainwater related data in the data storage module in the history X periods;

[0011] Rainwater storage facility module: store rainwater through rainwater storage facilities, which are used to transmit rainwater to the rainwater purification module or the flood discharge module;

[0012] Rainwater purification module: used to receive rainwater transmitted by the rainwater storage facility module and purify the rainwater;

[0013] Flood discharge module: used to discharge rainwater through drainage pipes and pump stations;

[0014] Rainwater storage facility overflow prediction module: analyze the rainwater related data standardized in the data processing module to predict the probability of rainwater storage facility overflow in each time period in the future period; when the probability is greater than the set threshold, the time period is determined as an overflow risk time period;

[0015] Dynamic early warning threshold module: when the time comes to the predicted overflow risk time period, the early warning threshold for rainwater storage facility overflow warning at this time adopts a dynamic early warning threshold, which dynamically adjusts the early warning threshold;

[0016] Purification and drainage control module: when the time comes to the predicted overflow risk time period and rainfall occurs:

[0017] In the rainwater purification module, start the automatic addition program of the purification agent, and increase the amount of purification agent put into the rainwater purification per unit time;

[0018] In the flood discharge module, the drainage valve of the drainage pipe is in a semi-open state, and the operating power of the pump station remains unchanged;

[0019] When the time comes to the predicted overflow risk time period and reaches the dynamic early warning threshold:

[0020] In the flood discharge module, the drainage valve of the drainage pipe is in a fully open state, and the operating power of the pump station is increased;

[0021] At this time, the rainwater storage facility enters the drainage mode.

[0022] Preferably, the rainwater related data includes rainfall, rainfall intensity, rainfall duration, water level of the rainwater storage facility, and rainwater flow in the pipeline of the rainwater storage facility.

[0023] Preferably, in the data processing module, 3≤X, and X is a positive integer.

[0024] Preferably, in the dynamic early warning threshold module, the dynamic early warning threshold is set as follows: assuming that the early warning threshold of the rainwater storage facility overflow early warning is S The dynamic early warning threshold is S w ;

[0025] The dynamic early warning threshold S w is set as follows: assuming that the total capacity of the rainwater storage facility is C t , the current remaining capacity is C s , the remaining capacity ratio C r is:

[0026]

[0027] Assuming that the degree of unobstructedness of the rainwater storage facility inlet pipeline is D c , the actual flow in the inlet pipeline is Q, and the theoretical maximum flow in the inlet pipeline is Q max , then

[0028] Assuming that the adjustment coefficient of the dynamic early warning threshold is K, w4 is the weight coefficient of the remaining capacity ratio C r , and w5 is the weight coefficient of the degree of unobstructedness D c of the inlet pipeline, then K is:

[0029] K = w4 × (1 - C r ) + w5 × (1 - D c );

[0030] Then

[0031] Preferably, the value range of w4 is 0.6-0.9, and the value range of w5 is 0.1-0.4.

[0032] Preferably, in the rainwater storage facility overflow prediction module, a machine learning algorithm is used to analyze the normalized data in the past X periods to predict the probability p1 of rainwater storage facility overflow in each time period in the future one period;

[0033] A time series analysis algorithm is used to analyze the normalized data in the past X periods to predict the probability p2 of rainwater storage facility overflow in each time period in the future one period;

[0034] A deep learning algorithm is used to analyze the normalized data in the past X periods to predict the probability p3 of rainwater storage facility overflow in each time period in the future one period;

[0035] Assuming that the weights of the machine learning algorithm, the time series analysis algorithm, and the deep learning algorithm are w1, w2, and w3 respectively, then the fused probability p is:

[0036] P = w1 * p1 + w2 * p2 + w3 * p3

[0037] When the derived probability P is greater than a set threshold, the time period is determined as a risk of overflow time period.

[0038] Preferably, in the purification and drainage control module, the drainage mode is:

[0039] When the time comes to the predicted risk of overflow time period and reaches the dynamic early warning threshold, and the remaining capacity ratio in the rainwater storage facility does not reach the set value, the rainwater in the rainwater storage facility is discharged by the flood discharge drainage module by a unit amount of rainwater, and after an interval of a unit time, the rainwater is discharged by the flood discharge drainage module again by a unit amount of rainwater.

[0040] When the remaining capacity ratio in the rainwater storage facility reaches the set value, the rainwater storage facility no longer receives rainwater until the remaining capacity ratio in the rainwater storage facility decreases to below the set value, and the rainwater storage facility continues to receive rainwater.

[0041] A big data-based sponge city rainwater treatment method, comprising the following steps:

[0042] S1, collecting water level data of the rainwater storage facility, rainfall, rainfall intensity and rainfall duration data of the corresponding period through sensors and monitors;

[0043] S2, setting a period, dividing the period into multiple time periods, and performing standardization processing on the data in the historical X periods;

[0044] 3≤X, and X is a positive integer;

[0045] S3, using a machine learning algorithm to analyze the standardized data in the historical X periods, and predicting the probability p1 of overflow of the rainwater storage facility in each time period in the future period;

[0046] Using a time series analysis algorithm to analyze the standardized data in the historical X periods, and predicting the probability p2 of overflow of the rainwater storage facility in each time period in the future period;

[0047] Using a deep learning algorithm to analyze the standardized data in the historical X periods, and predicting the probability p3 of overflow of the rainwater storage facility in each time period in the future period;

[0048] Let the weights of the machine learning algorithm, the time series analysis algorithm and the deep learning algorithm be w1, w2 and w3 respectively, then the fused probability p is:

[0049] P = w1 * p1 + w2 * p2 + w3 * p3

[0050] When the derived probability P is greater than a set threshold value, the time period is determined as an overflow risk time period;

[0051] S4, when the time comes to the predicted overflow risk time period, the overflow warning threshold of the rainwater storage facility at this time adopts a dynamic warning threshold;

[0052] S5, when the time comes to the predicted overflow risk time period, and when rainfall occurs:

[0053] In the rainwater purification module, the automatic addition program of the purification agent is started, and the purification agent for rainwater purification in a unit time is increased;

[0054] In the flood drainage module, the drainage valve of the drainage pipe is in a semi-open state, and the operating power of the pump station is unchanged;

[0055] S6, when the time comes to the predicted overflow risk time period, and when the dynamic warning threshold is reached:

[0056] In the flood drainage module, the drainage valve of the drainage pipe is in a fully open state, and the operating power of the pump station is increased;

[0057] At this time, the rainwater storage facility enters the drainage mode;

[0058] The drainage mode is:

[0059] When the time comes to the predicted overflow risk time period, and when the dynamic warning threshold is reached, and when the remaining capacity ratio of the rainwater storage facility does not reach the set value, the rainwater in the rainwater storage facility is discharged by the flood drainage module by a unit amount of rainwater, and after an interval of a unit time, the rainwater is discharged by the flood drainage module by a unit amount of rainwater again;

[0060] When the remaining capacity ratio of the rainwater storage facility reaches the set value, the rainwater storage facility no longer receives rainwater, until the remaining capacity ratio of the rainwater storage facility decreases to below the set value, and the rainwater storage facility continues to receive rainwater.

[0061] In the present application, the proposed big data-based sponge city rainwater treatment system has the following beneficial technical effects:

[0062] 1. The setting of the purification and drainage control module, when the time comes to the predicted overflow risk period and rainfall occurs: in the rainwater purification module, the automatic addition program of the purification agent is started, the amount of the purification agent for rainwater purification per unit time is increased, the efficiency of rainwater purification is improved, thereby speeding up the inflow of rainwater in the rainwater storage facility into the rainwater purification module, and freeing up the water storage space of the rainwater storage facility; in the flood drainage module, the drainage valve of the drainage pipe is in a semi-open state, and the operating power of the pump station remains unchanged, to avoid the rapid inflow of initial rainwater into the rainwater storage facility, causing the rainwater in the rainwater storage facility to overflow, and orderly draining rainwater; and when the time comes to the predicted overflow risk period and reaches the dynamic early warning threshold: in the flood drainage module, the drainage valve of the drainage pipe is in a fully open state, and the operating power of the pump station is increased, to further drain rainwater; at this time, the rainwater storage facility enters the drainage mode, and the rainwater in the rainwater storage facility is discharged through the flood drainage module by a unit amount of rainwater, and after an interval of a unit time, a unit amount of rainwater is discharged again through the flood drainage module, thereby shortening the time required to reduce the water storage in the rainwater storage facility to below the early warning threshold, fully utilizing the water storage capacity of the rainwater storage facility, and reducing the risk of urban waterlogging.

[0063] 2. The present application predicts the probability of overflow of the rainwater storage facility by analyzing historical data and using machine learning algorithms, time series analysis algorithms and deep learning algorithms for algorithm fusion, overcomes the limitations of single algorithms through algorithm fusion mode, greatly improves the prediction accuracy, and when the probability obtained in the prediction period is greater than the set threshold, the time period is determined as an overflow risk period, and when the time is located in the overflow risk period, the early warning threshold of the rainwater storage facility overflow early warning is dynamically adjusted, the early warning limit is flexibly adjusted, the false alarm or missed alarm caused by the fixed threshold is greatly reduced, and the timeliness and accuracy of early warning are improved.

[0064] Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following and / or can be learned by practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0065] Figure 1 The principle block diagram of the system of the present application;

[0066] Figure 2 The flowchart of the method of the present application. DETAILED DESCRIPTION

[0067] Embodiments of the present application are described below in the accompanying drawings, which are indicative of like or similar elements by like or similar reference numerals throughout the several views. The embodiments described below are exemplary only with the intent and purpose to convey aspects of the application, but not to limit the application.

[0068] As Figure 1 indicated in one of the sponge city rainwater treatment systems based on big data, comprising:

[0069] Data acquisition module: collect rainwater related data through setting sensors and monitors;

[0070] Rainwater related data includes rainfall, rainfall intensity, rainfall duration, water level of rainwater storage facilities, rainwater flow in the pipeline of rainwater storage facilities;

[0071] Sensor types include rain sensor, water level sensor, water quality monitoring sensor;

[0072] Data transmission module: use wired or wireless communication technology to transmit the collected rainwater related data to the data storage module;

[0073] Data storage module: for receiving and storing rainwater related data transmitted by data transmission module, using distributed storage architecture to store rainwater related data;

[0074] Data processing module: set a period, divide the period into multiple time periods, and standardize the rainwater related data in the data storage module in the past X periods;

[0075] 3≤X, and X is a positive integer;

[0076] The data is standardized by using Z-score standardization method;

[0077] Rainwater storage facility module: store rainwater through rainwater storage facilities, rainwater storage facilities are used to transmit rainwater to rainwater purification module or flood discharge module;

[0078] Rainwater purification module: for receiving rainwater transmitted by rainwater storage facility module, and purifying rainwater;

[0079] Flood discharge module: for discharging rainwater through drainage pipe and pump station;

[0080] As an illustration, the flood drainage module is usually used for flood relief to alleviate the risk of urban waterlogging; rainwater is usually stored in rainwater storage facilities in sponge cities and naturally lowered as groundwater. The rainwater stored in rainwater storage facilities can be released for use after purification; usually, when the rainfall is large, the rainwater storage facilities overflow or overflow, and the flood drainage is carried out.

[0081] The rainwater storage facility overflow prediction module analyzes the standardized rainwater related data in the data processing module to predict the probability of rainwater storage facility overflow in each time period in the future period; when the probability is greater than the set threshold, the time period is determined as an overflow risk period;

[0082] The mechanical learning algorithm is used to analyze the standardized data in the past X periods to predict the probability p1 of rainwater storage facility overflow in each time period in the future period;

[0083] The mechanical learning algorithm uses the standardized historical data for training, takes the standardized characteristic values of different time periods as input and the rainwater storage facility overflow as output label, trains the model to learn the inherent law of the data, and then predicts the probability p1 of rainwater storage facility overflow in the future time period;

[0084] The time series analysis algorithm is used to analyze the standardized data in the past X periods to predict the probability p2 of rainwater storage facility overflow in each time period in the future period;

[0085] The time series analysis algorithm mines the trend of standardized data changing over time, fully considers the correlation of rainwater data in the time dimension in the historical period, and predicts the probability p2 of rainwater storage facility overflow in the future time period;

[0086] The deep learning algorithm is used to analyze the standardized data in the past X periods to predict the probability p3 of rainwater storage facility overflow in each time period in the future period;

[0087] The deep learning algorithm has strong non-linear modeling capability, can deeply capture the complex characteristics of rainwater data, and can output the probability p3 of rainwater storage facility overflow in the future time period by training a large amount of standardized historical data;

[0088] Let the weights of the mechanical learning algorithm, the time series analysis algorithm and the deep learning algorithm be w1, w2 and w3 respectively, then the fused probability p is:

[0089] P=w1*p1+w2*p2+w3*p3

[0090] When the derived probability P is greater than a set threshold, the time period is determined as an overflow risk time period;

[0091] The weights of the mechanical learning algorithm, the time series analysis algorithm, and the deep learning algorithm can be determined based on performance evaluation:

[0092] For example, the F1-score of the mechanical learning algorithm, the time series analysis algorithm, and the deep learning algorithm can be verified, and the weights of the mechanical learning algorithm, the time series analysis algorithm, and the deep learning algorithm can be determined according to the F1-score;

[0093] The F1-score is an important indicator for measuring the performance of a classification model, which combines the concepts of precision and recall, and evaluates the performance of the model by calculating their harmonic mean. The maximum value of the F1-score is 1, and the minimum value is 0. The higher the value, the better the performance of the model.

[0094] In an optional embodiment, the algorithm with the highest F1-score value among the three algorithms of mechanical learning algorithm, time series analysis algorithm, and deep learning algorithm can be set to have a weight ratio of 0.5, the algorithm with the F1-score value in the middle can be set to have a weight ratio of 0.3, and the algorithm with the lowest F1-score value can be set to have a weight ratio of 0.2; wherein, if the highest F1-score value minus the lowest F1-score value is greater than 0.1, the algorithm with the highest F1-score value among the three algorithms of mechanical learning algorithm, time series analysis algorithm, and deep learning algorithm can be set to have a weight ratio of 0.6, the algorithm with the F1-score value in the middle can be set to have a weight ratio of 0.3, and the algorithm with the lowest F1-score value can be set to have a weight ratio of 0.1;

[0095] Dynamic early warning threshold module: when the time comes to the predicted overflow risk time period, the early warning threshold for overflow early warning of the rainwater storage facility adopts a dynamic early warning threshold, and the early warning threshold is dynamically adjusted;

[0096] Let the early warning threshold for overflow early warning of the rainwater storage facility be S The dynamic early warning threshold is S w ;

[0097] The dynamic early warning threshold S w is set as follows: let the total capacity of the rainwater storage facility be C t , the current remaining capacity be C s , then the remaining capacity ratio C r is:

[0098]

[0099] Let the degree of unobstructedness of the water inlet pipe of the rainwater storage facility be D c, the actual flow in the inlet pipeline is Q, and the theoretical maximum flow in the inlet pipeline is Q max , then

[0100] Let the adjustment coefficient of the dynamic early warning threshold be K, w4 be the proportion of the remaining capacity C r , and w5 be the weight coefficient of the unobstructed degree D c of the inlet pipeline, then K is:

[0101] K = w4 × (1 - C r ) + w5 × (1 - D c );

[0102] The value range of w4 is 0.6-0.9, and the value range of w5 is 0.1-0.4;

[0103] In an optional embodiment, w1 = 0.6, and w2 = 0.4;

[0104] , then

[0105] In this way, when the remaining capacity of the system rainwater storage facility is small or the unobstructed degree of the inlet pipeline is poor, the early warning threshold will be lowered accordingly, so that the early warning is easier to trigger, and the ability of the sponge city rainwater treatment system to cope with potential risks is improved;

[0106] As an illustration, when it has not rained yet, the actual flow Q in the inlet pipeline is particularly low, resulting in a decrease in the dynamic early warning threshold S w , but at this time, the value of C r is high, and the high value of C r causes the dynamic early warning threshold S w to rise, so that the dynamic early warning threshold S w will remain within a certain range and will not be easily triggered;

[0107] Flexibly adjusting the early warning threshold avoids false positives or omissions caused by a fixed threshold, ensuring the timeliness and accuracy of early warning.

[0108] Purification and drainage control module: when the time comes to the predicted risk of overflow period, and rainfall occurs:

[0109] In the rainwater purification module, the automatic addition program of the purification agent is started, and the amount of the purification agent for rainwater purification per unit time is increased; the efficiency of rainwater purification is improved,

[0110] In an optional embodiment, the amount of the purification agent for rainwater purification per unit time is increased by 30%-50%;

[0111] In an optional embodiment, when the automatic addition program for the purifying agent is started in the rainwater purification module, multiple stirring devices are simultaneously deployed and started to accelerate the fusion reaction between the purifying agent and the rainwater. At the same time, the sedimentation, filtration, and adsorption processes are activated to quickly reduce the content of suspended particles, heavy metals, and organic matter in the rainwater, improve the purification efficiency, and accelerate the flow of rainwater into the rainwater purification module from the rainwater storage facility, thus freeing up water storage space for the rainwater storage facility.

[0112] In the flood discharge and drainage module, the drainage valve of the drainage pipe is in a semi-open state, and the operating power of the pump station remains unchanged;

[0113] A drain valve being in a partially open state means that the drain valve is only half open.

[0114] In the flood discharge and drainage module, a distributed control system is used to control the valves and pumping stations in the drainage network. The distributed control system is used to adjust the opening status of the valves and the operating power of the pumping stations.

[0115] When the predicted overflow risk period arrives and the dynamic early warning threshold is reached:

[0116] In the flood discharge and drainage module, the drainage valve of the drainage pipe is fully open, and the operating power of the pump station is increased;

[0117] This ensures a steady increase in flood discharge, preventing sudden changes in flow from impacting the drainage system.

[0118] At this time, the rainwater storage facility enters drainage mode.

[0119] The drainage mode is:

[0120] When the predicted overflow risk period arrives and the dynamic warning threshold is reached, and the remaining capacity ratio in the rainwater storage facility does not reach the set value, the rainwater in the rainwater storage facility will discharge one unit of rainwater through the flood discharge module, and after an interval of one unit of time, discharge another unit of rainwater through the flood discharge module.

[0121] When the remaining capacity ratio in the rainwater storage facility reaches the set value, the rainwater storage facility will stop receiving rainwater until the remaining capacity ratio in the rainwater storage facility drops below the set value, at which point the rainwater storage facility will continue to receive rainwater.

[0122] Discharging a unit amount of rainwater through the flood discharge and drainage module refers to the flow of water from the rainwater storage facility into the flood discharge and drainage pipe.

[0123] like Figure 2 The illustrated sponge city rainwater treatment method includes the following steps:

[0124] S1, collecting water level data of rainwater storage facilities, rainfall in corresponding period, rainfall intensity and rainfall duration data through sensors and monitors;

[0125] S2, setting a period, dividing the period into multiple time periods, and performing standardization processing on data in X historical periods;

[0126] 3≤X, and X is a positive integer;

[0127] S3, using a machine learning algorithm to analyze the standardized data in X historical periods, and predicting the probability p1 of overflow of rainwater storage facilities in each time period in the future period;

[0128] Using a time series analysis algorithm to analyze the standardized data in X historical periods, and predicting the probability p2 of overflow of rainwater storage facilities in each time period in the future period;

[0129] Using a deep learning algorithm to analyze the standardized data in X historical periods, and predicting the probability p3 of overflow of rainwater storage facilities in each time period in the future period;

[0130] Let the weights of the machine learning algorithm, the time series analysis algorithm and the deep learning algorithm be w1, w2 and w3 respectively, then the fused probability p is:

[0131] P=w1*p1+w2*p2+w3*p3;

[0132] When the probability P is greater than the set threshold, the time period is determined as an overflow risk time period;

[0133] S4, when the time comes to the predicted overflow risk time period, the warning threshold of the rainwater storage facility overflow warning at this time adopts a dynamic warning threshold;

[0134] S5, when the time comes to the predicted overflow risk time period and rainfall occurs:

[0135] In the rainwater purification module, the automatic addition program of the purification agent is started, and the amount of purification agent for rainwater purification per unit time is increased;

[0136] In the flood drainage module, the drainage valve of the drainage pipe is in a semi-open state, and the operating power of the pump station remains unchanged;

[0137] S6, when the time comes to the predicted overflow risk time period and reaches the dynamic warning threshold:

[0138] In the flood drainage module, the drainage valve of the drainage pipe is in a fully open state, and the operating power of the pump station is increased;

[0139] At this time, the rainwater storage facility enters the drainage mode;

[0140] The drainage mode is:

[0141] When the time comes to the predicted overflow risk period and the dynamic early warning threshold is reached, and the remaining capacity ratio in the rainwater storage facility does not reach the set value, the rainwater in the rainwater storage facility is discharged by one unit of rainwater through the flood discharge drainage module, and after an interval of one unit of time, one unit of rainwater is discharged again through the flood discharge drainage module.

[0142] When the remaining capacity ratio in the rainwater storage facility reaches the set value, the rainwater storage facility no longer receives rainwater until the remaining capacity ratio in the rainwater storage facility decreases to below the set value, and the rainwater storage facility continues to receive rainwater.

[0143] In summary, the present application analyzes historical data, uses machine learning algorithms, time series analysis algorithms, and deep learning algorithms for algorithm fusion to predict the probability of overflow of the rainwater storage facility. The algorithm fusion mode overcomes the limitations of a single algorithm, greatly improves the prediction accuracy, and when the probability obtained during the prediction period is greater than the set threshold, the time period is determined to be an overflow risk period. When the time is in the overflow risk period, the early warning threshold for rainwater storage facility overflow is dynamically adjusted, the early warning limit is flexibly adjusted, the occurrence of false positives or false negatives caused by a fixed threshold is greatly reduced, and the timeliness and accuracy of early warning are improved.

[0144] When the time comes to the predicted overflow risk period and rainfall occurs, the purification and drainage control module is set: in the rainwater purification module, the automatic addition program of the purifying agent is started, the amount of purifying agent added for rainwater purification per unit time is increased, the efficiency of rainwater purification is improved, thereby accelerating the flow of rainwater in the rainwater storage facility into the rainwater purification module, freeing up storage space for the rainwater storage facility; in the flood discharge drainage module, the drainage valve of the drainage pipe is in a semi-open state, and the operating power of the pump station remains unchanged, to avoid the rapid influx of initial rainwater into the rainwater storage facility, causing the rainwater in the rainwater storage facility to overflow, and to orderly drain the rainwater; when the time comes to the predicted overflow risk period and the dynamic early warning threshold is reached: in the flood discharge drainage module, the drainage valve of the drainage pipe is in a fully open state, and the operating power of the pump station is increased, to further drain the rainwater; at this time, the rainwater storage facility enters the drainage mode, and intermittent drainage is performed, thereby shortening the time required to reduce the water storage in the rainwater storage facility to below the early warning threshold, fully utilizing the water storage capacity of the rainwater storage facility, and reducing the risk of urban waterlogging.

[0145] Meanwhile, the contents not described in detail in the specification are all existing technologies known to those skilled in the art.

[0146] In the embodiments of the present application, it should be understood that the disclosed system or method can be implemented in other ways. For example, the embodiments of the application described above are merely schematic, and the division of the modules is merely a logical function division. There can be another division during actual implementation.

[0147] The modules described as separated components can or can not be physically separated, and the components displayed as modules can or can not be physical modules, and can be located in one place or distributed on multiple network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0148] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of hardware plus software functional module.

[0149] For those skilled in the art, it is obvious that the present application is not limited to the details of the above exemplary embodiments, and the present application can be implemented in other specific forms without departing from the essential characteristics of the present application.

[0150] The above description is only the preferred embodiment of the present application, but the protection scope of the present application is not limited thereto, and any skilled person in the art can make equivalent replacements or changes to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A big data-based sponge city rainwater treatment system, characterized in that, Comprise: Data acquisition module: collect rainwater related data through setting sensors and monitors; Data transmission module: transmit the collected rainwater related data to the data storage module through wired or wireless communication technology; Data storage module: for storing rainwater related data; Data processing module: set a period, divide the period into multiple time periods, and standardize the rainwater related data in the data storage module in the past X periods; Rainwater storage facility module: for storing rainwater through rainwater storage facilities; Rainwater purification module: for receiving rainwater transmitted by the rainwater storage facility module and purifying the rainwater; Flood drainage module: for receiving rainwater transmitted by the rainwater storage facility module and discharging the rainwater through drainage pipes and pump stations; Rainwater storage facility overflow prediction module: for predicting the probability of rainwater storage facility overflow in the future; when the probability is greater than the set threshold, the time period is judged as an overflow risk time period; Dynamic early warning threshold module: for dynamically adjusting the early warning threshold of rainwater storage facility overflow warning; when the time is in the overflow risk time period, the dynamic early warning threshold in the dynamic early warning threshold module is used; Purification and drainage control module: when the time is in the overflow risk time period and rainfall occurs, the amount of purifying agent added per unit time in the rainwater purification module increases, and the drainage valve of the drainage pipe in the flood drainage module is in a semi-open state; when the time is in the overflow risk time period and reaches the dynamic early warning threshold: the drainage valve of the drainage pipe in the flood drainage module is in a fully open state, and the operating power of the pump station increases, at which time the rainwater storage facility module enters the drainage mode. 2.The big data-based sponge city rainwater treatment system according to claim 1, wherein, Rainwater related data includes rainfall, rainfall intensity, rainfall duration, water level of rainwater storage facilities, and rainwater flow in pipes of rainwater storage facilities. 3.The big data-based sponge city rainwater processing system according to claim 1, wherein, In the data processing module, 3≤X, and X is a positive integer. 4.The big data-based sponge city rainwater processing system according to claim 1, wherein, In the dynamic early warning threshold module, the setting of the dynamic early warning threshold is as follows: setting the early warning threshold of the rainwater storage facility overflow early warning as The dynamic early warning threshold is S w ; Dynamic pre-alarm threshold S w The following is set: In the formula, K is the adjustment coefficient of the dynamic early warning threshold; Let the total capacity of the rainwater storage facility be C t , and the current remaining capacity be C s , then the remaining capacity ratio C r is: Degree of unobstructedness of the inlet pipe of the rainwater storage facility is D c Actual flow in the inlet pipe is Q max Then w4 is the weight coefficient of the remaining capacity ratio C r w5 is the weight coefficient of the water inlet pipe patency D c K is: K = w4 x (1 - C r ) + w5 x (1 - D c ). 5.The big data-based sponge city rainwater processing system according to claim 4, characterized in that, The value range of w4 is 0.6-0.9, and the value range of w5 is 0.1-0.

4. 6.The big data-based sponge city rainwater processing system according to claim 1, wherein, In the rainwater storage facility overflow prediction module, a mechanical learning algorithm is used to analyze the standardized data in the past X periods to predict the probability p1 of rainwater storage facility overflow in each time period in the future period; A time series analysis algorithm is used to analyze the standardized data in the past X periods to predict the probability p2 of rainwater storage facility overflow in each time period in the future period; A deep learning algorithm is used to analyze the standardized data in the past X periods to predict the probability p3 of rainwater storage facility overflow in each time period in the future period; Set the weights of the mechanical learning algorithm, time series analysis algorithm, and deep learning algorithm as w1, w2, and w3, respectively, then the fused probability p is: P=w1*p1+w2*p2+w3*p3; When the derived probability P is greater than the set threshold, the time period is judged as an overflow risk time period. 7.The big data-based sponge city rainwater processing system according to claim 1, wherein, In the purification and drainage control module, the drainage mode is: When the time comes to the predicted overflow risk period and reaches the dynamic warning threshold, and the remaining capacity ratio in the rainwater storage facility does not reach the set value, the rainwater in the rainwater storage facility is discharged by one unit of rainwater through the flood discharge module, and after an interval of one unit of time, one unit of rainwater is discharged again through the flood discharge module; When the remaining capacity ratio in the rainwater storage facility reaches the set value, the rainwater storage facility no longer receives rainwater until the remaining capacity ratio in the rainwater storage facility decreases to below the set value, and the rainwater storage facility continues to receive rainwater.

8. The big data-based sponge city rainwater treatment method according to any one of claims 1-7, characterized in that, The method comprises the following steps: S1, collecting water level data of the rainwater storage facility, rainfall in the corresponding period, rainfall intensity and rainfall duration data through sensors and monitors; S2, setting a period, dividing the period into multiple time periods, and performing standardization processing on data in X historical periods; 3≤X, and X is a positive integer; S3, using a machine learning algorithm to analyze the standardized data in the X historical periods to predict the probability p1 of overflow of the rainwater storage facility in each time period in the future period; using a time series analysis algorithm to analyze the standardized data in the X historical periods to predict the probability p2 of overflow of the rainwater storage facility in each time period in the future period; using a deep learning algorithm to analyze the standardized data in the X historical periods to predict the probability p3 of overflow of the rainwater storage facility in each time period in the future period; the weights of the machine learning algorithm, the time series analysis algorithm and the deep learning algorithm are w1, w2 and w3 respectively, and the fused probability p is: P=w1*p1+w2*p2+w3*p3 When the probability P is greater than the set threshold, the time period is determined as an overflow risk period; S4, when the time comes to the predicted overflow risk period, the warning threshold of the rainwater storage facility overflow warning is a dynamic warning threshold; S5, when the time comes to the predicted overflow risk period and rainfall occurs: in the rainwater purification module, the automatic addition program of the purification agent is started, and the amount of the purification agent for rainwater purification in a unit of time is increased; in the flood discharge module, the drainage valve of the drainage pipe is in a half-open state, and the operating power of the pump station remains unchanged; S6, when the time comes to the predicted overflow risk period and reaches the dynamic warning threshold: in the flood discharge module, the drainage valve of the drainage pipe is in a fully open state, and the operating power of the pump station is increased; at this time, the rainwater storage facility enters the drainage mode; the drainage mode is: when the time comes to the predicted overflow risk period and reaches the dynamic warning threshold, and the remaining capacity ratio in the rainwater storage facility does not reach the set value, the rainwater in the rainwater storage facility is discharged by one unit of rainwater through the flood discharge module, and after an interval of one unit of time, one unit of rainwater is discharged again through the flood discharge module; When the proportion of the remaining capacity in the rainwater storage facility reaches the set value, the rainwater storage facility no longer receives rainwater until the proportion of the remaining capacity in the rainwater storage facility decreases to below the set value, and the rainwater storage facility continues to receive rainwater.

Images