An information-based intelligent drainage management method and system

Abstract

The present invention provides an information-based intelligent drainage management method and system, which relates to the technical field of drainage management. The method includes: obtaining water level data, first flow data and pump station flow control data, and obtaining a water level change function; determining the predicted rainwater inflow, the predicted water inflow into the water pipe and determining the predicted water inflow data; determining the predicted pump station flow control data according to the predicted water inflow data and the water level change function, and adjusting the pump station of the drainage pipe at the beginning of the next monitoring cycle. According to the present invention, a water level change function that can describe the relationship between the water level data, the first flow data and the pump station flow control data can be determined, and then the predicted pump station flow control data of the pump station of the drainage pipe of each water storage area can be determined, the drainage flow of each water storage area can be optimized, and the drainage flow of each area can be coordinated so that the water level of each water storage area is within a reasonable range, reducing the risk of overflow of the water storage tank in each area.

Description

Technical Field

[0001] The present invention relates to the technical field of drainage management, and in particular to an information-based intelligent drainage management method and system. Background Art

[0002] CN116596166B discloses an intelligent drainage management system for urban floods, which relates to the field of drainage management technology and includes a main control center, wherein the main control center is communicatively connected with a rainfall prediction module, a sewage monitoring module, a drainage treatment module, an equipment management module, and an information feedback module; the rainfall prediction module is used to construct a rainfall prediction model and obtain a predicted rainfall amount according to the rainfall prediction model; the sewage monitoring module is used to obtain a daily sewage amount and obtain a predicted sewage amount according to the daily sewage amount; the drainage treatment module is used to obtain a drainage state and generate alarm information and process it through drainage equipment; the equipment management module is used to determine whether the drainage equipment is a faulty equipment and generate corresponding fault information; the information feedback module is used to feed back the alarm information and the fault information to urban residents or staff respectively; through this scheme, the impact of floods can be effectively reduced.

[0003] CN118607746A discloses a visual command and dispatch management system for urban drainage, which includes: the system includes: infrastructure layer, IoT sensing layer, platform support layer and business application layer; the infrastructure layer is used to provide basic hardware resources for the system; the IoT sensing layer is connected to the infrastructure layer, and is used to obtain the monitoring data and video information of the urban drainage network in the target area in real time; the platform support layer is connected to the IoT sensing layer, and is used to collect and process the monitoring data of the urban drainage network, and build a visual management model for urban drainage; the business application layer is connected to the platform support layer, and is used to establish a comprehensive drainage display system according to user needs, analyze, judge and alarm the monitoring data of the urban drainage network, so as to assist in the command and dispatch management of urban drainage. The IAAS architecture adopted by the scheme saves the investment of personnel and technical costs, and at the same time, it has strong anti-interference ability, which effectively improves the efficiency and response speed of visual command and dispatch management of urban drainage.

[0004] Therefore, the relevant technology can manage the drainage equipment in various areas of the city through the main control center, which is conducive to improving the efficiency and accuracy of drainage management and reducing the possibility of flood disasters. However, the relevant technology cannot determine the optimal control data of the pumping stations of the drainage pipes in each area. When the drainage capacity of the main pipeline is certain, it is difficult to coordinate the drainage flow of each area, which may cause risks such as reservoir overflow in some areas.

[0005] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] The present invention provides an information-based intelligent drainage management method and system, which can solve the technical problem that related technologies coordinate the drainage flow of various areas, which may cause risks such as overflow of reservoirs in some areas.

[0007] According to a first aspect of the present invention, there is provided an information-based intelligent drainage management method, comprising:

[0008] At multiple moments in the current monitoring period, obtaining water level data of multiple water storage areas and first flow data of drainage pipes;

[0009] Obtain the pump station flow control data of the drainage pipes in each water storage area during the current monitoring period;

[0010] Determine a water level variation function of each water storage area according to the water level data, the first flow data and the pump station flow control data, wherein the water level variation function is used to describe the relationship between the water level data, the first flow data and the pump station flow control data;

[0011] Determine the predicted rainwater inflows for multiple water storage areas during the next monitoring period based on the weather forecast data for the next monitoring period;

[0012] Determine the predicted inflow volume of the water inlet pipe at each moment of the next monitoring cycle according to the start time of the next monitoring cycle and the historical flow data of the water inlet pipe in each historical monitoring cycle within a plurality of historical dates;

[0013] Determine the predicted water inflow data at each moment of the next monitoring cycle according to the predicted rainwater inflow and the predicted water inflow into the water pipe;

[0014] Determine the predicted pump station flow control data of the drainage pipes in each water storage area in the next monitoring cycle according to the predicted water inflow flow data at each moment of the next monitoring cycle and the water level change function;

[0015] At the beginning of the next monitoring cycle, the pump station of the drainage pipeline in the water storage area is adjusted according to the predicted pump station flow control data.

[0016] According to a second aspect of the present invention, there is provided an information-based intelligent drainage management system, comprising:

[0017] A first acquisition module is used to acquire water level data of multiple water storage areas and first flow data of the drainage pipe at multiple moments in the current monitoring period;

[0018] The second acquisition module is used to obtain the pump station flow control data of the drainage pipes of each water storage area in the current monitoring period;

[0019] A water level variation function module, used to determine the water level variation function of each water storage area according to the water level data, the first flow data and the pump station flow control data, wherein the water level variation function is used to describe the relationship between the water level data, the first flow data and the pump station flow control data;

[0020] A rainwater inflow prediction module is used to determine the predicted rainwater inflow of multiple water storage areas in the next monitoring period according to the weather forecast data of the next monitoring period;

[0021] The water inflow prediction module is used to determine the predicted water inflow at each moment of the next monitoring cycle according to the start time of the next monitoring cycle and the historical flow data of the water inflow at each historical monitoring cycle within multiple historical dates;

[0022] A predicted water inflow data module, used to determine the predicted water inflow data at each moment of the next monitoring cycle according to the predicted rainwater inflow and the predicted water inflow pipe inflow;

[0023] A pump station flow control data prediction module is used to determine the predicted pump station flow control data of the drainage pipes in each water storage area in the next monitoring period according to the predicted water inflow flow data at each moment of the next monitoring period and the water level change function;

[0024] The regulating module is used to regulate the pump station of the drainage pipeline in the water storage area according to the predicted pump station flow control data at the beginning of the next monitoring cycle.

[0025] Technical effect: According to the present invention, the water level change function can be determined based on the water level data of each water storage area, the first flow data of the drainage pipe and the pump station flow control data, and the inflow flow data of each water storage area in the next monitoring period can be predicted, so as to determine the predicted pump station flow control data of the pump station of the drainage pipe of each water storage area based on the water level change function and the inflow flow data, so as to optimize the drainage flow of each water storage area, and coordinate the drainage flow of each area under the condition of a certain total drainage flow, so that the water level of each water storage area is within a reasonable range, and the risk of overflow of the reservoir in each area is reduced. When determining the water level change function, the undetermined coefficient can be set by the undetermined coefficient equation, so as to describe the positive correlation between the first flow data and the water level data, and the positive correlation between the first flow data and the pump station flow control data, and the nonlinear relationship between the first flow data and the pump station control data can be described by exponential setting, so as to accurately describe the relationship between the water level data, the first flow data and the pump station flow control data of the water storage area by solving the undetermined coefficient. When training the flow prediction model, the reference value of the data can be described by the cosine similarity between the first target historical flow vector and the first flow vector and the sequence number of the target historical monitoring period in multiple candidate historical monitoring periods, thereby giving higher weights to data with greater reference value, and thus making the training more targeted, improving the training efficiency and training effect. When determining the predicted pump station flow control data, the relationship between the pending value of the predicted pump station flow control data and the predicted water level data at each moment can be determined by setting the recursive initial conditions and the recursive relationship, and then the constraints can be obtained, and the objective function can be set according to the global optimization goal, and then the optimal solution of the predicted pump station flow control data of each pump station can be obtained through the optimization solution process, so as to achieve global optimization to the greatest extent and reduce the risk of reservoir overflows in some water storage areas.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work.

[0028] Figure 1 A schematic diagram of a flow chart of an information-based intelligent drainage management method according to an embodiment of the present invention is exemplarily shown;

[0029] Figure 2 A block diagram of an information-based intelligent drainage management system according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0031] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0032] Figure 1 The following is a flow chart of an information-based intelligent drainage management method according to an embodiment of the present invention, wherein the method comprises:

[0033] Step S101, obtaining water level data of multiple water storage areas and first flow data of drainage pipes at multiple moments in the current monitoring cycle;

[0034] Step S102, obtaining pump station flow control data of drainage pipes in each water storage area in the current monitoring period;

[0035] Step S103, determining a water level variation function of each water storage area according to the water level data, the first flow data and the pump station flow control data, wherein the water level variation function is used to describe the relationship between the water level data, the first flow data and the pump station flow control data;

[0036] Step S104, determining the predicted rainwater inflow of multiple water storage areas in the next monitoring period according to the weather forecast data of the next monitoring period;

[0037] Step S105, determining the predicted inflow volume of the water inlet pipe at each moment of the next monitoring cycle according to the start time of the next monitoring cycle and the historical flow data of the water inlet pipe in each historical monitoring cycle within a plurality of historical dates;

[0038] Step S106, determining predicted water inflow data at each moment of the next monitoring cycle according to the predicted rainwater inflow and the predicted water inflow into the water pipe;

[0039] Step S107, determining the predicted pump station flow control data of the drainage pipes in each water storage area in the next monitoring period according to the predicted water inflow flow data at each moment in the next monitoring period and the water level change function;

[0040] Step S108, at the beginning of the next monitoring cycle, the pump station of the drainage pipe in the water storage area is adjusted according to the predicted pump station flow control data.

[0041] According to the information-based intelligent drainage management method of the embodiment of the present invention, the water level change function can be determined based on the water level data of each water storage area, the first flow data of the drainage pipe and the pump station flow control data, and the inflow flow data of each water storage area in the next monitoring period can be predicted, so as to determine the predicted pump station flow control data of the drainage pipe of each water storage area based on the water level change function and the inflow flow data, so as to optimize the drainage flow of each water storage area. When the total drainage flow is constant, the drainage flow of each area is coordinated so that the water level of each water storage area is within a reasonable range, thereby reducing the risk of overflow of the reservoir in each area.

[0042] According to one embodiment of the present invention, the information-based intelligent drainage management method can be executed by a background system. The background system may include a display system for displaying water level data of each water storage area, flow data of each pipeline, pump station flow control data of the pump station in the pipeline system, etc., and can manually input control instructions to control the pump station flow control data, or the processor can automatically determine the optimal pump station flow control data based on the above data, and then generate control instructions to transmit to each pump station to control the flow of the pump station. The flow of the pump station may be related to the number of water pumps started in the pump station and the power of each water pump. The power of the started water pump The greater the number, the greater the power, and the greater the flow of the pump station. In particular, the pump station flow control data can be used to indicate the total power of all water pumps in the pump station. After the pump station receives the pump station flow control data, it can reasonably classify the number of water pumps started and the power of each water pump based on the total power it indicates. For example, the range of the number of water pumps to be started can be preset based on the total power, and when the total power belongs to range 1, one water pump is started, and when the total power belongs to range 2, two water pumps are started... and the total power is evenly distributed through the multiple water pumps that are started. The present invention does not limit the specific meaning of the pump station flow control data and the control method of the water pump.

[0043] According to one embodiment of the present invention, each water storage area may be provided with a water level detection device, each pipeline may be provided with a flow meter, and the water level data and flow data may be summarized through the Internet of Things system for analysis by the background system. In addition, the background system may record the topological relationship between each water storage area and the pipeline, so as to determine the connection relationship between the pipelines and the main pipeline into which the drainage pipelines of multiple water storage areas converge according to the topological relationship, and thus determine the appropriate pump station control data based on the above connection relationship and the main pipeline, so as to coordinate the flow of the drainage pipelines of the water storage areas that converge into the same main pipeline, so that the water level of these water storage areas is maintained at a relatively ideal level as a whole, and reduce the possibility of risks such as overflow of the reservoir from a global perspective.

[0044] According to one embodiment of the present invention, the steps of obtaining flow data and pump station flow control data of the information-based intelligent drainage management method can be always executed, so that various types of data can be displayed in the background system. A preset early warning rule can be used to determine whether to execute the step of optimizing the pump station flow control data, for example, steps S103-S108. For example, when the water level data of at least one water storage area reaches the set water level, the step of optimizing the pump station flow control data is started, and the pump station flow control data of each pump station can be uniformly planned and managed.

[0045] According to one embodiment of the present invention, in step S101, the duration of each monitoring cycle can be 15 minutes, 30 minutes, one hour, etc., and the time interval between adjacent moments can be 30 seconds, 1 minute, 3 minutes, etc. The present invention does not limit the duration of the monitoring cycle and the time interval between adjacent moments. The water storage area may include a regulating reservoir, a water storage tank of a sewage treatment plant, a reservoir, etc., and the present invention does not limit the type of the water storage area.

[0046] According to one embodiment of the present invention, in step S102, the pump station flow control data can be adjusted before each monitoring cycle begins, and after the adjustment is completed, the pump station flow control data can be maintained for one monitoring cycle. The pump station flow control data of the drainage pipes of each water storage area in the current monitoring cycle can be obtained.

[0047] According to one embodiment of the present invention, in step S103, both the water level data and the pump station flow control data have a certain relationship with the first flow data of the drainage pipe. For example, if the drainage pipe is located below the water storage area, the higher the water level in the water storage area, the greater the water pressure in the drainage pipe, resulting in a greater flow in the drainage pipe. On the other hand, the greater the pump station control data, the greater the flow in the drainage pipe. Therefore, the water level change function of each water storage area can be determined based on the measured water level data, the first flow data and the pump station flow control data to describe the relationship between the water level data, the first flow data and the pump station flow control data. The water level data, the first flow data and the pump station flow control data in the current monitoring cycle can be used, and the water level data, the first flow data and the pump station flow control data in other historical monitoring cycles can also be used to increase the data volume and improve the accuracy of the water level change function.

[0048] According to one embodiment of the present invention, determining the water level variation function of each water storage area according to the water level data, the first flow data and the pump station flow control data includes: determining the undetermined coefficient equation of the water level variation function of the i-th water storage area according to formula (1),

[0049] (1)

[0050] in, is the first flow data of the i-th water storage area at the j-th moment, is the water level data of the i-th water storage area at the j-th moment, is the pump station flow control data of the drainage pipe in the i-th water storage area, , and is an undetermined coefficient; according to the water level data and the first flow data of the ith water storage area at multiple times and the pump station flow control data of the ith water storage area, the undetermined coefficient is solved to obtain the solution value of the undetermined coefficient; according to the solution value of the undetermined coefficient and the undetermined coefficient equation of the water level change function of the ith water storage area, the water level change function of the ith water storage area is obtained.

[0051] According to one embodiment of the present invention, the higher the water level, the deeper the position of the drainage pipe relative to the liquid surface, the greater the water pressure, resulting in a larger first flow data in the drainage pipe. On the other hand, the control data of the pump station is also positively correlated with the first flow data, and the two are not linearly related. Therefore, the undetermined coefficient is set As an index of the pump station flow control data, and set the coefficient to be determined Multiplier, set the coefficient to be determined As the residual term, the equation of the undetermined coefficient is obtained. Furthermore, the water level data, the first flow data and the pump station flow control data at multiple moments of the i-th water storage area can be substituted into the above-mentioned equation of the undetermined coefficient, and each undetermined coefficient can be solved to obtain the solution value of the undetermined coefficient. Then, the solution value of the undetermined coefficient can be substituted into the equation of the undetermined coefficient to obtain the water level change function of the i-th water storage area. Since the design parameters of the drainage pipes and the design parameters of the pump stations in each water storage area may be different from each other, the solution values ​​of the undetermined coefficients corresponding to each water storage area are also different. Therefore, the above solution can be performed for each water storage area, thereby obtaining the water level change function of each water storage area, thereby describing the relationship between the water level data, the first flow data and the pump station flow control data of each water storage area.

[0052] In this way, the undetermined coefficients can be set through the undetermined coefficient equations to describe the positive correlation between the first flow data and the water level data, and the positive correlation between the first flow data and the pump station flow control data, and the nonlinear relationship between the first flow data and the pump station control data can be described through exponential setting, so as to accurately describe the relationship between the water level data, the first flow data and the pump station flow control data in the water storage area by solving the undetermined coefficients.

[0053] According to one embodiment of the present invention, in step S104, the predicted rainwater inflow of multiple water storage areas in the next monitoring period can be determined based on the weather forecast data of the next monitoring period. For example, the predicted rainfall in the next monitoring period can be determined by weather forecast data, and the total amount of rainwater flowing into the water storage area in the next monitoring period can be obtained by multiplying the predicted rainfall by the area of ​​the water storage area, and then the ratio of the total amount of rainwater to the duration of the monitoring period is solved to obtain the predicted inflow into the water pipe at each moment of the next monitoring period. On rainy days, water storage areas are more prone to overflow risks, so the drainage of multiple water storage areas can be reasonably scheduled, and the pumping stations can be reasonably controlled to reduce the risk of overflow as a whole.

[0054] According to one embodiment of the present invention, in step S105, the flow rate of the water inlet pipe of the water storage area may fluctuate greatly, but there may be certain regularities. For example, the water storage area is a water storage tank of a sewage treatment plant, and its water inlet pipe may flow into domestic sewage, industrial sewage, etc. There may be certain regularities. For example, during working hours around 15:00 in the afternoon, the flow rate of industrial sewage may increase, and during dinner time around 18:00 in the afternoon, the flow rate of domestic sewage may increase, and the flow rate of industrial sewage may decrease. Therefore, the flow rate regularity of the water inlet pipe can be found based on the start time of the next monitoring cycle and the historical flow rate data of the water inlet pipe of the historical monitoring cycle in the historical date, and the flow rate of the water inlet pipe of the next monitoring cycle can be predicted, that is, the inflow of the water inlet pipe can be predicted.

[0055] According to one embodiment of the present invention, the predicted inflow volume of the water inlet pipe at each moment of the next monitoring cycle is determined according to the start time of the next monitoring cycle and the historical flow data of the water inlet pipe at each monitoring cycle within multiple historical dates, including: selecting a candidate historical monitoring cycle according to the start time of the next monitoring cycle among multiple historical dates; obtaining a first historical flow vector according to the first historical flow data of the water inlet pipe at multiple moments of the previous monitoring cycle of the candidate historical monitoring cycle; obtaining a first flow vector according to the water inlet flow data of the water inlet pipe at multiple moments of the current monitoring cycle; determining the cosine similarity between the first historical flow vector corresponding to each candidate historical monitoring cycle and the first flow vector; screening the candidate historical monitoring cycles according to the cosine similarity to obtain a target historical monitoring cycle; training a flow prediction model according to a first target historical flow vector composed of the target historical flow data at multiple moments of the previous monitoring cycle of the target historical monitoring cycle, the second target historical flow data at multiple moments of the target historical monitoring cycle and the cosine similarity to obtain a trained flow prediction model; processing the first flow vector according to the trained flow prediction model to obtain the predicted inflow volume of the water inlet pipe at each moment of the next monitoring cycle.

[0056] According to one embodiment of the present invention, the multiple historical dates may include historical dates one month before the current date, historical dates one quarter before the current date, or historical dates one year before the current date, etc. The present invention does not limit the number of historical dates. Further, the historical monitoring cycle in the historical date may be selected based on the start time of the next monitoring cycle, that is, the historical monitoring cycle whose start time in the historical date is the same as the start time of the next monitoring cycle is selected as the candidate historical monitoring cycle. For example, if the start time of the next monitoring cycle is 10:00 am, the historical monitoring cycle whose start time is 10:00 am is selected from the historical date as the candidate historical monitoring cycle.

[0057] According to one embodiment of the present invention, the first historical flow data of multiple moments in the previous monitoring period of the candidate historical monitoring period can be used as components of a vector to form a first historical flow vector. Similarly, the water inflow flow data of multiple moments in the current monitoring period can be used as components of a vector to form a first flow vector. Then, the cosine similarity between each first historical flow vector and the first flow vector can be calculated respectively, and the target historical monitoring period can be screened out based on the cosine similarity. For example, the first historical flow vectors with a cosine similarity greater than or equal to a preset similarity threshold can be screened out, and the candidate historical monitoring periods corresponding to these first historical flow vectors are used as target historical monitoring periods. That is, the cosine similarity can be used to screen out data with low similarity, and screen out data with high similarity and greater reference value, so as to improve the subsequent training efficiency and training targeting. Since the cosine similarity is greater than or equal to the preset similarity threshold, the first historical flow data at multiple moments of the previous monitoring cycle of the target historical monitoring cycle and the first flow data of the current monitoring cycle have a high degree of similarity, and the inflow flow of the water storage area has a certain regularity. Therefore, the change law between the target historical flow data of the previous monitoring cycle of the target historical monitoring cycle and the second target historical flow data of the target historical monitoring cycle is similar to the change law between the inflow flow data of the current monitoring cycle and the predicted inflow into the water pipe of the next monitoring cycle. Therefore, the law can be found through the flow prediction model, that is, the target historical flow data of the previous monitoring cycle of the target historical monitoring cycle and the second target historical flow data at multiple moments of the target historical monitoring cycle are used to train the flow prediction model to find the law, and the trained flow prediction model is used to process the first flow vector of the current monitoring cycle, so as to obtain the predicted inflow into the water pipe of the next monitoring cycle based on the law and the inflow flow data of the current monitoring cycle.

[0058] According to one embodiment of the present invention, the traffic prediction model is a deep learning neural network model. According to the first target historical traffic vector composed of the target historical traffic data at multiple moments of the previous monitoring period of the target historical monitoring period, the second target historical traffic data at multiple moments of the target historical monitoring period and the cosine similarity, the traffic prediction model is trained to obtain the trained traffic prediction model, including: inputting the first target historical traffic vector into the traffic prediction model to obtain the training traffic data at multiple moments of the target historical monitoring period; determining the loss function of the traffic prediction model according to the training traffic data, the second target historical traffic data and the cosine similarity; training the traffic prediction model according to the loss function to obtain the trained traffic prediction model.

[0059] According to one embodiment of the present invention, the loss function of the traffic prediction model is determined according to the training traffic data, the second target historical traffic data and the cosine similarity, including: determining the loss function of the traffic prediction model according to formula (2): ,

[0060] (2)

[0061] in, is the training flow data of the water inlet pipe of the i-th water storage area at the j-th moment in the k-th target historical monitoring period, is the second target historical flow data of the water inlet pipe of the i-th water storage area at the j-th moment in the k-th target historical monitoring cycle, is the cosine similarity between the first target historical flow vector and the first flow vector corresponding to the kth target historical monitoring period of the water inlet pipe of the i-th water storage area, N is the number of historical dates, is the sequence number of the kth target historical monitoring period in multiple candidate historical monitoring periods, is the number of target historical monitoring cycles screened based on the first flow vector of the water inlet pipe of the i-th water storage area, m is the number of water storage areas, n is the number of moments in the current monitoring cycle, i, j, k, n, N, and m are both positive integers.

[0062] According to one embodiment of the present invention, the first target historical flow vector of the water inlet pipe of the i-th water storage area in the previous monitoring period of the k-th target historical monitoring period is input into the flow prediction model to obtain the training flow data of the water inlet pipe of the i-th water storage area at multiple moments in the k-th target historical monitoring period. It is the difference between the training flow data of the water inlet pipe of the i-th water storage area at the j-th moment in the k-th target historical monitoring cycle and the measured second target historical flow data. That is, the total gap between the training flow data of the water inlet pipe of the i-th water storage area at multiple moments in the k-th target historical monitoring period and the measured second target historical flow data. This total gap, as part of the loss function, can reduce the loss function during the training process, thereby reducing the above total gap and improving the prediction accuracy of the flow prediction model.

[0063] According to one embodiment of the present invention, the cosine similarity between the first target historical flow vector and the first flow vector can be used as the weight of the above-mentioned total gap. The higher the cosine similarity, the higher the similarity between the target historical flow data of the previous period of the target historical monitoring period and the water inflow data of the current monitoring period, the greater the reference value of the second target historical flow data of the target historical monitoring period, the greater the reference value of the change law between the target historical flow data and the second target historical flow data, and the more important the reference significance for predicting the predicted water inflow volume of the next monitoring period based on the water inflow data of the current monitoring period. Therefore, the cosine similarity can be used as the weight of the above-mentioned total gap. On the other hand, It can also be used as the weight of the above total gap. The larger the serial number of the kth target historical monitoring period in multiple candidate historical monitoring periods, the closer the historical date corresponding to the kth target historical monitoring period is to the current date, and the greater its reference value. The weights of the above two aspects can be multiplied to obtain the comprehensive weight of the above total gap. The above comprehensive weight can be used to perform a weighted summation of the above total gaps corresponding to each water storage area in each target historical monitoring period to obtain the loss function of the flow prediction model, so that the weight of data with higher reference value can be increased during the training process, thereby making the training more targeted and improving the training efficiency and training effect.

[0064] According to one embodiment of the present invention, the flow prediction model can be trained by using the above loss function in a back-propagation manner to obtain a trained flow prediction model. The first flow vector of the current monitoring period is processed by the trained flow prediction model to obtain the predicted inflow volume of the water pipe at each moment of the next monitoring period.

[0065] In this way, the reference value of the data can be described by the cosine similarity between the first target historical traffic vector and the first traffic vector and the sequence number of the target historical monitoring period in multiple candidate historical monitoring periods, thereby giving higher weights to data with greater reference value, thereby making the training more targeted and improving the training efficiency and training effect during the training process.

[0066] According to one embodiment of the present invention, in step S106, the predicted rainwater inflow and the predicted water inflow into the water pipe at each moment may be added together to determine the predicted water inflow data at each moment.

[0067] According to one embodiment of the present invention, in step S107, a pump station control optimization model can be used to determine the predicted pump station flow control data. The pump station control optimization model can be an optimization model such as a nonlinear programming model, a genetic algorithm model, etc., and constraints and objective functions can be set. Under the constraints, the optimal solution that can satisfy the goal described by the objective function to the greatest extent is solved. The optimal solution is the predicted pump station flow control data.

[0068] According to one embodiment of the present invention, based on the predicted water inflow flow data at each moment of the next monitoring period and the water level change function, the predicted pump station flow control data of the drainage pipes of each water storage area in the next monitoring period is determined, including: determining the constraints of the pump station control optimization model based on the predicted water inflow flow, the water level change function and the water level data; determining the objective function of the pump station control optimization model based on the predicted water inflow flow, the water level change function and the target water level data of each water storage area; solving the pump station control optimization model according to the constraints and the objective function to obtain the predicted pump station flow control data.

[0069] According to one embodiment of the present invention, the constraint conditions of the pump station control optimization model are determined according to the predicted water inflow, the water level change function and the water level data, including: determining the constraint conditions of the pump station control optimization model according to formula (3), formula (4), formula (5), formula (6), formula (7), formula (8) and formula (9),

[0070] (3)

[0071] (4)

[0072] (5)

[0073] (6)

[0074] (7)

[0075] (8)

[0076] (9)

[0077] in, is the predicted inflow of the i-th water storage area at the j-th moment in the next monitoring period, is the predicted drainage flow of the drainage pipe in the i-th water storage area at the j-th moment in the next monitoring cycle, is the area of ​​the ith water storage area, is the water level change rate of the i-th water storage area at the j-th moment in the next monitoring cycle, is the predicted water level data of the i-th water storage area at the j+1th moment in the next monitoring cycle, is the predicted water level data of the i-th water storage area at the j-th moment in the next monitoring cycle, is the time interval between adjacent moments, is the predicted water level data of the ith water storage area at the first moment of the next monitoring cycle, is the water level data of the ith water storage area at the nth moment of the current monitoring cycle, where n is the number of moments in the current monitoring cycle. is the undetermined value of the predicted pump station flow control data for the i-th water storage area in the next monitoring cycle, for The solution value of for The solution value of for The solution value of is the maximum flow of the trunk pipeline, m is the number of water storage areas, is the maximum flow of the drainage pipe in the i-th water storage area, is the maximum control data of the pumping station of the drainage pipe in the i-th water storage area.

[0078] According to one embodiment of the present invention, in formula (3), the difference between the predicted inflow flow and the predicted drainage flow is the net inflow flow of the water storage area, and the ratio of the net inflow flow to the area of ​​the water storage area is the water level change rate of the water storage area. In formula (4), the difference in predicted water level data between adjacent moments is equal to the product of the water level change rate and the time interval between adjacent moments. In formula (5), the predicted water level data at the first moment of the next monitoring cycle is equal to the water level data at the last moment of the current monitoring cycle, which can be used as the starting condition for the recursion of the predicted water level data at multiple moments of the next monitoring cycle. In formula (6), the predicted drainage flow can be calculated based on the water level change function obtained above, as well as the pending value of the predicted pump station flow control data and the predicted water level data. In the recursive process, formula (5) can be used as the starting condition, and the pending value of the predicted pump station flow control data is set, so that the predicted water level data at the first moment of the next monitoring period and the pending value of the predicted pump station flow control data are substituted into formula (6) to obtain the predicted drainage flow at the first moment of the next monitoring period, and then the predicted drainage flow at the first moment of the next monitoring period and the predicted drainage flow determined based on the flow prediction model are substituted into formula (3) to obtain the water level change rate at the first moment of the next monitoring period, and then the water level change rate at the first moment of the next monitoring period and the predicted water level data are substituted into formula (4) to obtain the predicted water level at the second moment of the next monitoring period. According to the above process, the subsequent recursion can be started. For example, the predicted water level data at the second moment and the undetermined value of the predicted pump station flow control data are substituted into formula (6) to obtain the predicted drainage flow at the second moment of the next monitoring cycle. Then, the predicted drainage flow at the second moment of the next monitoring cycle and the predicted drainage flow determined based on the flow prediction model are substituted into formula (3) to obtain the water level change rate at the second moment of the next monitoring cycle. Then, the water level change rate at the second moment of the next monitoring cycle and the predicted water level data are substituted into formula (4) to obtain the predicted water level data at the third moment of the next monitoring cycle... The above process is iterated to obtain the predicted water level data at each moment of the next monitoring cycle. Moreover, it can be seen from the above recursive process that different undetermined values ​​of the predicted pump station flow control data can be set to obtain different predicted water level data. Therefore, the optimal predicted pump station flow control data for each water storage area can be solved so that the predicted water level data for each water storage area meets the requirements.

[0079] According to one embodiment of the present invention, formula (7) indicates that the sum of the predicted drainage flow rates of the drainage pipes in the water storage area that flow into the same trunk pipe at any time does not exceed the maximum flow rate of the trunk pipe, and formula (8) indicates that the predicted drainage flow rates of the drainage pipes at any time do not exceed the maximum flow rates of the drainage pipes. Formula (9) indicates that the undetermined value of the predicted pump station flow control data of each pump station does not exceed the maximum control data.

[0080] According to one embodiment of the present invention, determining the objective function of the pump station control optimization model according to the predicted water inflow, the water level variation function and the target water level data of each water storage area includes: determining the objective function of the pump station control optimization model according to formula (10),

[0081] (10)

[0082] in, is the target water level data of the ith storage area, and min is the minimization function.

[0083] According to one embodiment of the present invention, is the difference between the predicted water level data and the target water level data of the ith water storage area at the end of the next monitoring cycle. Formula (10) means minimizing the total difference between the predicted water level data and the target water level data of all water storage areas at the end of the next monitoring cycle, that is, setting reasonable target water level data. For example, the water storage area used for peak shaving and valley filling (e.g., regulating reservoir) is opened and set to a reasonable water level, and the sewage pool of the sewage treatment plant is set to a lower water level, so that the predicted water level data of all water storage areas can reach the target water level data as much as possible, thereby achieving global optimization, and reasonably utilizing the peak shaving and valley filling effect of regulating reservoir, and reducing the risk of reservoir overflow in some water storage areas.

[0084] According to one embodiment of the present invention, as described above, by setting different pending values ​​of predicted pump station flow control data, different predicted water level data can be obtained. Therefore, under the constraints of the above-mentioned constraints, the pending values ​​of the predicted pump station flow control data of each pump station can be continuously adjusted so that the corresponding predicted water level data can achieve the goal described by the objective function to the greatest extent, thereby obtaining the optimal solution for the predicted pump station flow control data.

[0085] In this way, by setting the initial conditions of the recursion and the recursive relationship, the relationship between the undetermined value of the predicted pump station flow control data and the predicted water level data at each moment can be determined, and then the constraint conditions can be obtained. The objective function is set according to the goal of global optimization, and then the optimal solution of the predicted pump station flow control data of each pump station can be obtained through the optimization solution process, thereby achieving global optimization to the greatest extent and reducing the risk of reservoir overflows in some water storage areas.

[0086] According to one embodiment of the present invention, in step S108, the optimal solution of the above-mentioned predicted pump station flow control data can be used as the predicted pump station flow control data of each pump station in the next monitoring cycle, and the control data of each pump station can be adjusted based on this.

[0087] According to the information-based intelligent drainage management method of the embodiment of the present invention, the water level change function can be determined based on the water level data of each water storage area, the first flow data of the drainage pipe and the pump station flow control data, and the inflow flow data of each water storage area in the next monitoring period can be predicted, so as to determine the predicted pump station flow control data of the pump station of the drainage pipe of each water storage area based on the water level change function and the inflow flow data, so as to optimize the drainage flow of each water storage area, and coordinate the drainage flow of each area under the condition of a certain total drainage flow, so that the water level of each water storage area is within a reasonable range, and the risk of overflow of the reservoir in each area is reduced. When determining the water level change function, the undetermined coefficient can be set by the undetermined coefficient equation, so as to describe the positive correlation between the first flow data and the water level data, and the positive correlation between the first flow data and the pump station flow control data, and the nonlinear relationship between the first flow data and the pump station control data can be described by the index setting, so as to accurately describe the relationship between the water level data, the first flow data and the pump station flow control data of the water storage area by solving the undetermined coefficient. When training the flow prediction model, the reference value of the data can be described by the cosine similarity between the first target historical flow vector and the first flow vector and the sequence number of the target historical monitoring period in multiple candidate historical monitoring periods, thereby giving higher weights to data with greater reference value, and thus making the training more targeted, improving the training efficiency and training effect. When determining the predicted pump station flow control data, the relationship between the pending value of the predicted pump station flow control data and the predicted water level data at each moment can be determined by setting the recursive initial conditions and the recursive relationship, and then the constraints can be obtained, and the objective function can be set according to the global optimization goal, and then the optimal solution of the predicted pump station flow control data of each pump station can be obtained through the optimization solution process, so as to achieve global optimization to the greatest extent and reduce the risk of reservoir overflows in some water storage areas.

[0088] Figure 2 A block diagram of an information-based intelligent drainage management system according to an embodiment of the present invention is exemplarily shown, and the system includes:

[0089] A first acquisition module is used to acquire water level data of multiple water storage areas and first flow data of the drainage pipe at multiple moments in the current monitoring period;

[0090] The second acquisition module is used to obtain the pump station flow control data of the drainage pipes of each water storage area in the current monitoring period;

[0091] A water level variation function module, used to determine the water level variation function of each water storage area according to the water level data, the first flow data and the pump station flow control data, wherein the water level variation function is used to describe the relationship between the water level data, the first flow data and the pump station flow control data;

[0092] A rainwater inflow prediction module is used to determine the predicted rainwater inflow of multiple water storage areas in the next monitoring period according to the weather forecast data of the next monitoring period;

[0093] The water inflow prediction module is used to determine the predicted water inflow at each moment of the next monitoring cycle according to the start time of the next monitoring cycle and the historical flow data of the water inflow at each historical monitoring cycle within multiple historical dates;

[0094] A predicted water inflow data module, used to determine the predicted water inflow data at each moment of the next monitoring cycle according to the predicted rainwater inflow and the predicted water inflow pipe inflow;

[0095] A pump station flow control data prediction module is used to determine the predicted pump station flow control data of the drainage pipes in each water storage area in the next monitoring period according to the predicted water inflow flow data at each moment of the next monitoring period and the water level change function;

[0096] The regulating module is used to regulate the pump station of the drainage pipeline in the water storage area according to the predicted pump station flow control data at the beginning of the next monitoring cycle.

[0097] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An information-based intelligent drainage management method, characterized in that: include: At multiple moments in the current monitoring period, obtaining water level data of multiple water storage areas and first flow data of drainage pipes; Obtain the pump station flow control data of the drainage pipes in each water storage area during the current monitoring period; Determine a water level variation function of each water storage area according to the water level data, the first flow data and the pump station flow control data, wherein the water level variation function is used to describe the relationship between the water level data, the first flow data and the pump station flow control data; Determine the predicted rainwater inflows for multiple water storage areas during the next monitoring period based on the weather forecast data for the next monitoring period; Determine the predicted inflow volume of the water inlet pipe at each moment of the next monitoring cycle according to the start time of the next monitoring cycle and the historical flow data of the water inlet pipe in each historical monitoring cycle within a plurality of historical dates; Determine the predicted water inflow data at each moment of the next monitoring cycle according to the predicted rainwater inflow and the predicted water inflow into the water pipe; Determine the predicted pump station flow control data of the drainage pipes in each water storage area in the next monitoring period according to the predicted water inflow flow data at each moment of the next monitoring period and the water level change function; At the beginning of the next monitoring cycle, adjusting the pumping station of the drainage pipeline in the water storage area according to the predicted pumping station flow control data; Determining a water level variation function of each water storage area according to the water level data, the first flow data and the pump station flow control data includes: According to the formula Determine the unknown coefficient equation of the water level change function of the i-th water storage area, where: is the first flow data of the i-th water storage area at the j-th moment, is the water level data of the i-th water storage area at the j-th moment, is the pump station flow control data of the drainage pipe in the i-th water storage area, , and is the coefficient to be determined; Solving the undetermined coefficients according to the water level data and the first flow data of the i-th water storage area at multiple times and the flow control data of the pump station of the i-th water storage area to obtain the solution values ​​of the undetermined coefficients; The water level variation function of the ith water storage area is obtained according to the solved value of the undetermined coefficient and the undetermined coefficient equation of the water level variation function of the ith water storage area.

2. The information-based intelligent drainage management method according to claim 1 is characterized in that: According to the start time of the next monitoring cycle and the historical flow data of the water inlet pipe in each monitoring cycle within multiple historical dates, the predicted water inlet pipe inflow at each time of the next monitoring cycle is determined, including: Selecting a candidate historical monitoring period from among multiple historical dates according to the start time of the next monitoring period; Obtaining a first historical flow vector according to first historical flow data of the water inlet pipeline at multiple moments in a monitoring period before the candidate historical monitoring period; Obtaining a first flow vector according to water inlet flow data of the water inlet pipeline at multiple moments in the current monitoring period; Determine the cosine similarity between the first historical traffic vector corresponding to each candidate historical monitoring period and the first traffic vector; According to the cosine similarity, candidate historical monitoring periods are screened to obtain a target historical monitoring period; Training a traffic prediction model based on a first target historical traffic vector composed of target historical traffic data at multiple moments of a previous monitoring period of a target historical monitoring period, second target historical traffic data at multiple moments of a target historical monitoring period, and the cosine similarity to obtain a trained traffic prediction model; The first flow vector is processed according to the trained flow prediction model to obtain the predicted water pipe inflow at each moment in the next monitoring cycle.

3. The information-based intelligent drainage management method according to claim 2 is characterized in that: According to the first target historical traffic vector composed of the target historical traffic data at multiple moments of the previous monitoring period of the target historical monitoring period, the second target historical traffic data at multiple moments of the target historical monitoring period and the cosine similarity, the traffic prediction model is trained to obtain the trained traffic prediction model, including: Inputting the first target historical flow vector into the flow prediction model to obtain training flow data at multiple moments of the target historical monitoring period; Determine a loss function of a traffic prediction model according to the training traffic data, the second target historical traffic data and the cosine similarity; The traffic prediction model is trained according to the loss function to obtain a trained traffic prediction model.

4. The information-based intelligent drainage management method according to claim 3 is characterized in that: Determining a loss function of a traffic prediction model according to the training traffic data, the second target historical traffic data, and the cosine similarity includes: According to the formula Determine the loss function of the traffic prediction model ,in, is the training flow data of the water inlet pipe of the i-th water storage area at the j-th moment in the k-th target historical monitoring period, is the second target historical flow data of the water inlet pipe of the i-th water storage area at the j-th moment in the k-th target historical monitoring cycle, is the cosine similarity between the first target historical flow vector and the first flow vector corresponding to the kth target historical monitoring period of the water inlet pipe of the i-th water storage area, N is the number of historical dates, is the sequence number of the kth target historical monitoring period in multiple candidate historical monitoring periods, is the number of target historical monitoring cycles screened based on the first flow vector of the water inlet pipe of the i-th water storage area, m is the number of water storage areas, n is the number of moments in the current monitoring cycle, i, j, k, n, N, and m are both positive integers.

5. The information-based intelligent drainage management method according to claim 1 is characterized in that: According to the predicted water inflow flow data at each moment of the next monitoring cycle and the water level change function, the predicted pump station flow control data of the drainage pipes in each water storage area in the next monitoring cycle is determined, including: Determining the constraint conditions of the pump station control optimization model according to the predicted inflow flow, the water level variation function and the water level data; Determining the objective function of the pump station control optimization model according to the predicted inflow flow, the water level variation function and the target water level data of each water storage area; The pump station control optimization model is solved according to the constraint conditions and the objective function to obtain predicted pump station flow control data.

6. The information-based intelligent drainage management method according to claim 5 is characterized in that: According to the predicted inflow flow, the water level variation function and the water level data, the constraint conditions of the pump station control optimization model are determined, including: According to the formula Determine the constraints of the pump station control optimization model, where: is the predicted inflow of the i-th water storage area at the j-th moment in the next monitoring period, is the predicted drainage flow of the drainage pipe in the i-th water storage area at the j-th moment in the next monitoring cycle, is the area of ​​the ith water storage area, is the water level change rate of the i-th water storage area at the j-th moment in the next monitoring cycle, is the predicted water level data of the i-th water storage area at the j+1th moment in the next monitoring cycle, is the predicted water level data of the i-th water storage area at the j-th moment in the next monitoring cycle, is the time interval between adjacent moments, is the predicted water level data of the ith water storage area at the first moment of the next monitoring cycle, is the water level data of the ith water storage area at the nth moment of the current monitoring cycle, where n is the number of moments in the current monitoring cycle. is the undetermined value of the predicted pump station flow control data for the i-th water storage area in the next monitoring cycle, for The solution value of for The solution value of for The solution value of is the maximum flow of the trunk pipeline, m is the number of water storage areas, is the maximum flow of the drainage pipe in the i-th water storage area, is the maximum control data of the pumping station of the drainage pipe in the i-th water storage area.

7. The information-based intelligent drainage management method according to claim 6 is characterized in that: According to the predicted inflow flow, the water level variation function and the target water level data of each water storage area, the objective function of the pump station control optimization model is determined, including: According to the formula Determine the objective function of the pump station control optimization model, where is the target water level data of the ith water storage area, min is the minimization function, It is the predicted water level data of the ith water storage area at the end of the next monitoring period.

Images