A coordinated dispatching method for sewage pumping stations based on predicted liquid level in pipe network

Abstract

The present invention discloses a sewage pump station coordinated dispatching method based on pipe network predicted liquid level, and specifically relates to the technical field of sewage pump station management. The present invention provides a method for constructing a pump pool dynamic liquid level prediction model, which is beneficial to improving the accuracy of pump pool dynamic liquid level prediction, and provides data support for subsequent dispatching and analysis processes. A method for constructing a pump station optimization dispatching model is provided, and the objective functions are as follows: under the constraint of storage capacity, the pump station flow rate fluctuation is minimized, the flow rate is constant, the energy consumption of the water pump unit is minimized, and the number of starts and stops of the water pump unit is minimized. This is beneficial to improving the reliability and stability of the pump station operation, and is beneficial to maximizing the life of the water pump and reducing maintenance costs, meeting the requirements of economic benefits and operation quality. At the same time, constraint conditions and optimization algorithms are set, and the optimization dispatching scheme solved by the pump station optimization dispatching model is finally obtained, which is more beneficial to the operation of the pump station.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage pump station management, and more specifically, to a sewage pump station coordinated scheduling method based on pipe network predicted liquid level. Background Art

[0002] In the existing sewage treatment technology, the scheduling method of sewage pumping stations is mainly based on manual operation or simple automatic control systems. These methods often have problems such as low efficiency, high energy consumption, and large equipment loss when treating large amounts of complex sewage. In addition, due to the lack of effective coordinated scheduling, the working conditions of various pumping stations cannot be optimized, resulting in poor overall treatment effects.

[0003] The existing coordinated scheduling method of sewage pumping stations first collects the working data of each pumping station, and then predicts the optimal working parameters of each pumping station through data analysis and machine learning algorithms. Finally, based on the prediction results, the working parameters of each pumping station are automatically adjusted, realizing the automatic control of the pumping station and improving the control accuracy and speed to a certain extent.

[0004] However, there are still some problems with the above system: the above system does not predict the liquid level in the pump station and its front-end drainage network, cannot guarantee the accuracy of the subsequent pump station coordinated scheduling data analysis, does not specifically explain the principle of automatic control, and cannot guarantee that the results of automatic control will definitely achieve high-efficiency and low-energy operation of the sewage pump station, and cannot ensure that the results of automatic control meet the economic benefit requirements. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a sewage pump station coordinated scheduling method based on the predicted liquid level of the pipeline network to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a sewage pump station coordinated scheduling method based on pipe network predicted liquid level, comprising the following steps:

[0007] S1: Input the static information, dynamic information of each pumping station and the monitoring site information set in the drainage network in front of the pumping station;

[0008] S2: A screen or mesh is installed at the pump station pump pool as a pre-treatment device to remove particulate impurities of a preset size in the sewage entering the pump;

[0009] S3: Obtain real-time liquid level and flow data at different monitoring locations through sensors, and simultaneously obtain real-time dynamic information of the pump station and input it into the pump pool dynamic liquid level prediction model to predict the pump pool liquid level of the pump station;

[0010] S4: inferring the predicted pump pool liquid level to calculate the future time required for the pump pool liquid level to reach the maximum pump pool liquid level in the current state;

[0011] S5: When the calculated theoretical required future duration is less than or equal to the preset scheduling analysis duration, the current monitoring data and the real-time dynamic information of the pumping station are input into the pumping station optimization scheduling model to generate a scheduling configuration plan;

[0012] S6: Perform scheduling based on the generated scheduling configuration plan, and update the scheduling strategy based on real-time data;

[0013] S7: Collect the actual equipment operation energy consumption data during the coordinated dispatching process of the sewage pumping station, the number of pumps that can be normally activated and the number of pumps that need to be activated in the pumping station, the response time of each device in the pumping station after the dispatching configuration plan is generated, the volume of target particulate impurities remaining in each cubic meter of sewage after pretreatment, the number of starts and stops of each pump used, and the number of sewage overflows;

[0014] S8: Set the pump station operation data summary analysis cycle, and process the above data to calculate the energy consumption control capacity coefficient, particle impurity removal accuracy, average water pump availability, average equipment response timeliness rate, average water pump start-stop rate, and sewage overflow control capacity coefficient, and then calculate the pump station operation quality control index;

[0015] S9: Back up the pump station coordinated dispatch records and data processing records within the pump station operation data summary and analysis period to the cloud database according to date.

[0016] Preferably, the static information of the pump station in step S1 includes the pump station site name, number, affiliated drainage line, associated upstream and downstream pump station names, GIS location, and geographic information; the dynamic information of the pump station includes the number of water pumps contained in the pump station, including but not limited to the specific design parameters of the water pumps including the pump head and flow rate, the maximum number of operating units, the maximum volume of the pump pool, the maximum flow rate allowed by the pump station, and the minimum flow rate allowed by the pump station; the monitoring site information set in the drainage network at the front end of the pump station includes the monitoring site number and the spatial geographic location coordinates of the monitoring site.

[0017] Preferably, the specific process of constructing the pump pool dynamic liquid level prediction model in step S3 is as follows:

[0018] S31: Collect historical operation data of the pump station, historical monitoring data of the monitoring site, and historical environmental data as original data sets, wherein the historical operation data of the pump station includes the number of water pumps in operation, the head of the water pump used, the flow rate and motor speed, the number of starts and stops of the water pump used, and the output flow rate of the pump station; the historical monitoring data of the monitoring site is the liquid level and flow rate synchronized with the time point of the historical operation data of the pump station; and the historical environmental data is the rainfall synchronized with the time point of the historical operation data of the pump station;

[0019] S32, extracting the number of running pumps, running status of pumps, output flow of pump stations, liquid level of pump pools, liquid level and flow of monitoring points of drainage pipe network, and rainfall at the same time point in the original data set as characteristic parameters;

[0020] S33, selecting a neural network model as a prediction model, setting a network architecture and basic parameters of the neural network model based on actual needs, dividing the characteristic parameter historical data into a training data set and a test data set, and using the training data set to train the set prediction model;

[0021] S34, use the test data set to train the trained prediction model, and use the maximum error MAXE, average error ME, mean absolute error MAE, root mean square error RMSE and determination coefficient R 2 Evaluate the prediction accuracy of the pump pool dynamic level prediction model L z , the specific calculation formula is: , R 2 The value is between 0 and 1. The closer it is to 1, the better the effect. The closer the other four indicators are to 0, the better the model results. e is a natural constant;

[0022] S35. Compare the prediction accuracy obtained by the self-set algorithm with the expected value. If it is greater than or equal to the expected value, the pump pool dynamic liquid level prediction model is put into use. Otherwise, continue to update and optimize the pump pool dynamic liquid level prediction model until the prediction accuracy reaches the expected value.

[0023] Preferably, the pump pool liquid level predicted in step S4 satisfies the following formula: , where Q r,t , Q c,t and Z B,t are the input flow, output flow and liquid level of the pump pool at time t, respectively. B is the cross-sectional area of ​​the pump pool, and the maximum liquid level Z in the pump pool Bmax The calculation formula is: , where V Bmax is the maximum volume of the pump pool, and the theoretical required future time t lx The specific calculation formula is: .

[0024] Preferably, the construction of the pump station optimization scheduling model in step S5 includes the following steps:

[0025] S51. Determine the objective function of the model: Under the constraint of regulation and storage capacity, the pump station has the smallest flow rate fluctuation and the most stable operation, so the objective function 1 is , Q e is the average pumping flow rate of the pumping station, n tis the number of nodes at which the pump station operates throughout the day. When the flow rate is constant, the pump unit consumes the least energy and the pump station operates most reliably. Therefore, objective function 2 is , n B is the number of pumps in the pump station, i,t is the start and stop state of the i-th water pump at time t, λ i,t =1 means the water pump is on at time t, λ i,t =0 means the pump is stopped at time t, t e is the unit time, η bi is the motor efficiency of the i-th water pump, N i,t is the shaft power of the water pump at time t, and the specific calculation formula is: , ρ w is the sewage density, g is the gravitational acceleration, Q i,t is the flow rate of the i-th water pump at time t, H i,t is the head of the i-th pump at time t, η ai is the efficiency of the i-th water pump under the working condition at time t. The water pump with the least number of starts and stops corresponds to the longest life and the lowest maintenance cost. Therefore, objective function 3 is ,ξ i is the start-stop loss conversion coefficient of the i-th water pump;

[0026] S52. Determine the constraints of the model: the water flow rate of each water pump in the water pump unit should meet the minimum and maximum flow requirements, that is, , Q i,max is the maximum water flow rate of the i-th water pump in the high-efficiency range. When calculating the minimum flow rate variation, the sum of the flow rates at each moment of the day should be kept consistent with the total flow rate, that is, When calculating the optimal outflow of a pumping station, the maximum and minimum constraints of the storage capacity of the pumping station sump should be met, that is, , Δt is the time change, V B,t is the volume of the pump pool at time t, V B,t-1 is the volume of the pump pool at time t-1. Ensuring that the pump can operate within the high-efficiency range can reduce the energy consumption of the sewage pumping station. Then, , Q i,min is the minimum water flow rate in the high-efficiency range of the i-th water pump. The head of the water pump unit should be greater than or equal to the sum of the static head and the head loss of the pressure pipe, that is, , ε i is the loss coefficient of the i-th water pump, and the pump head and flow rate should be greater than or equal to 0;

[0027] S53, based on the hierarchical sequence method, the objective function is divided into two steps to solve the problem in order. The total flow rate of the pump station and the volume of the sump in objective function 1 are used to solve the optimal uniformity of the sewage pump station. Objective function 2 and objective function 3 can be combined into one objective function because the units of the pump start-stop loss conversion coefficient are consistent. According to the total outflow of the pump station and the flow rate, head and other constraints of each pump obtained by objective 1, the flow distribution and start-stop optimization between units are carried out, and finally the optimal operating conditions are obtained. Therefore, the sewage pump station optimization scheduling model is as follows: , , ;

[0028] S54. After the sewage pump station optimization scheduling model is established, a genetic algorithm is used to optimize the model and finally obtain an optimized scheduling plan.

[0029] Preferably, the specific data processing process in step S7 is as follows:

[0030] S71, retrieve the expected equipment operation energy consumption, the target particle impurity removal size in sewage, and the expected retention volume of the target particle impurities removed per cubic meter of sewage set by the pump station operation management center, and set the pump station operation data summary and analysis cycle;

[0031] S72. Expected equipment operation energy consumption E during the period of summary and analysis based on pump station operation data y and actual equipment operating energy consumption E x Calculate the energy consumption control capability coefficient X CE , the specific formula is: , based on the expected retention volume V of the remaining target particle impurities per cubic meter of sewage y and the volume V of the remaining target particle impurities to be removed per cubic meter of sewage after pretreatment x Calculate the particle impurity removal accuracy α c , the specific formula is: ;

[0032] S73, based on the number of pumps m required in the i-th scheduling configuration plan bxi and the number of pumps that can be used normally m bzi Calculate the average pump availability L Be , the specific formula is: , n fa is the number of dispatch configuration schemes generated during the pump station operation data summary and analysis period, and the response time t of the jth device in the pump station after the i-th dispatch configuration scheme is generated. xyij Calculate the average device response timeliness rate L sxe , the specific formula is: , n sbiis the number of equipment used after the i-th dispatch configuration plan is generated, and the number of starts and stops of the j-th water pump that can be normally activated in the pump station after the i-th dispatch configuration plan is generated is m qtij Calculate the average pump start-stop rate L qte , the specific formula is: , based on the pump station operation data summary analysis cycle t fx Internal sewage overflow times m wy Calculate the sewage overflow control capacity coefficient X CW , the specific formula is: ;

[0033] S74, based on energy consumption control capability coefficient X CE , particle impurity removal accuracy α c , average pump availability L Be , average equipment response timeliness rate L sxe , average pump start-stop rate L qte , Sewage overflow control capacity coefficient X CW Calculate the pump station operation quality control index P CY The specific formula is:

[0034] Technical effects and advantages of the present invention:

[0035] 1. The present invention provides a method for constructing a pump pool dynamic liquid level prediction model, which is beneficial to improving the accuracy of the dynamic liquid level prediction of the pump pool and providing data support for the subsequent scheduling and analysis process. The present invention provides a method for constructing a pump station optimization scheduling model, which takes the three objectives of the minimum fluctuation of the pump station flow rate under the constraint of storage capacity, the minimum energy consumption of the water pump unit when the flow rate is constant, and the minimum number of start-stop times of the water pump unit as the objective function, which is beneficial to improving the reliability and stability of the pump station operation, and is beneficial to maximizing the life of the water pump and reducing maintenance costs, meeting the requirements of economic benefits and operation quality. At the same time, constraint conditions and optimization algorithms are set, and the optimization scheduling scheme solved by the pump station optimization scheduling model is finally obtained, which is more conducive to the operation of the pump station.

[0036] 2. The present invention collects actual equipment operation energy consumption data during the coordinated scheduling of sewage pumping stations, the number of pumps that can be normally activated and the number of pumps that need to be activated in the pumping station, the response time of each device in the pumping station after the scheduling configuration plan is generated, the volume of target particulate impurities remaining in each cubic meter of sewage after pretreatment, the number of starts and stops of each pump used, and the number of sewage overflows, sets a pumping station operation data summary and analysis cycle, and processes the above data to calculate the energy consumption control capacity coefficient, particulate impurity removal accuracy, average water pump availability, average equipment response timeliness rate, average water pump start-stop rate, and sewage overflow control capacity coefficient, and then calculates the pumping station operation quality control index, providing a method for evaluating the operation quality of a pumping station, which can understand the operation of the pumping station through intuitive data, and is conducive to decision makers to adjust strategies directly based on the pumping station operation data. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a diagram of the method steps of the present invention.

[0038] Figure 2 It is a system structure block diagram of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] like Figure 1 The embodiment shown provides a sewage pump station coordinated scheduling method based on pipe network predicted liquid level, comprising the following steps:

[0041] S1: Input the static information, dynamic information of each pumping station and the monitoring site information set in the drainage network in front of the pumping station;

[0042] Furthermore, the static information of the pump station in step S1 includes the pump station site name, number, affiliated drainage line, associated upstream and downstream pump station names, GIS location, and geographic information; the dynamic information of the pump station includes the number of water pumps contained in the pump station, specific design parameters of the water pumps including but not limited to water pump head and flow, the maximum number of operating units, the maximum volume of the pump pool, the maximum flow allowed by the pump station, and the minimum flow allowed by the pump station; the monitoring site information set in the drainage network at the front end of the pump station includes the monitoring site number and the spatial geographic location coordinates of the monitoring site.

[0043] S2: A screen or mesh is installed at the pump station pump pool as a pre-treatment device to remove particulate impurities of a preset size in the sewage entering the pump;

[0044] S3: Obtain real-time liquid level and flow data at different monitoring locations through sensors, and simultaneously obtain real-time dynamic information of the pump station and input it into the pump pool dynamic liquid level prediction model to predict the pump pool liquid level of the pump station;

[0045] Further, the specific process of constructing the pump pool dynamic liquid level prediction model in step S3 is as follows:

[0046] S31: Collect historical operation data of the pump station, historical monitoring data of the monitoring site, and historical environmental data as original data sets, wherein the historical operation data of the pump station includes the number of water pumps in operation, the head of the water pump used, the flow rate and motor speed, the number of starts and stops of the water pump used, and the output flow rate of the pump station; the historical monitoring data of the monitoring site is the liquid level and flow rate synchronized with the time point of the historical operation data of the pump station; and the historical environmental data is the rainfall synchronized with the time point of the historical operation data of the pump station;

[0047] S32, extracting the number of running pumps, running status of pumps, output flow of pump stations, liquid level of pump pools, liquid level and flow of monitoring points of drainage pipe network, and rainfall at the same time point in the original data set as characteristic parameters;

[0048] S33, selecting a neural network model as a prediction model, setting a network architecture and basic parameters of the neural network model based on actual needs, dividing the characteristic parameter historical data into a training data set and a test data set, and using the training data set to train the set prediction model;

[0049] S34, use the test data set to train the trained prediction model, and use the maximum error MAXE, average error ME, mean absolute error MAE, root mean square error RMSE and determination coefficient R 2 Evaluate the prediction accuracy of the pump pool dynamic level prediction model L z , the specific calculation formula is: , R 2 The value is between 0 and 1. The closer it is to 1, the better the effect. The closer the other four indicators are to 0, the better the model results. e is a natural constant;

[0050] S35. Compare the prediction accuracy obtained by the self-set algorithm with the expected value. If it is greater than or equal to the expected value, the pump pool dynamic liquid level prediction model is put into use. Otherwise, continue to update and optimize the pump pool dynamic liquid level prediction model until the prediction accuracy reaches the expected value.

[0051] S4: inferring the predicted pump pool liquid level to calculate the future time required for the pump pool liquid level to reach the maximum pump pool liquid level in the current state;

[0052] Further, the pump pool liquid level predicted in step S4 satisfies the following formula: , where Q r,t, Q c,t and Z B,t are the input flow, output flow and liquid level of the pump pool at time t, respectively. B is the cross-sectional area of ​​the pump pool, and the maximum liquid level Z in the pump pool Bmax The calculation formula is: , where V Bmax is the maximum volume of the pump pool, and the theoretical required future time t lx The specific calculation formula is: .

[0053] S5: When the calculated theoretical required future duration is less than or equal to the preset scheduling analysis duration, the current monitoring data and the real-time dynamic information of the pumping station are input into the pumping station optimization scheduling model to generate a scheduling configuration plan;

[0054] Further, the construction of the pump station optimization scheduling model in step S5 includes the following steps:

[0055] S51. Determine the objective function of the model: Under the constraint of regulation and storage capacity, the pump station has the smallest flow rate fluctuation and the most stable operation, so the objective function 1 is , Q e is the average pumping flow rate of the pumping station, n t is the number of nodes at which the pump station operates throughout the day. When the flow rate is constant, the pump unit consumes the least energy and the pump station operates most reliably. Therefore, objective function 2 is , n B is the number of pumps in the pump station, i,t is the start and stop state of the i-th water pump at time t, λ i,t =1 means the water pump is on at time t, λ i,t =0 means the pump is stopped at time t, t e is the unit time, η bi is the motor efficiency of the i-th water pump, N i,t is the shaft power of the water pump at time t, and the specific calculation formula is: , ρ w is the sewage density, g is the gravitational acceleration, Q i,t is the flow rate of the i-th water pump at time t, H i,t is the head of the i-th pump at time t, η ai is the efficiency of the i-th water pump under the working condition at time t. The water pump with the least number of starts and stops corresponds to the longest life and the lowest maintenance cost. Therefore, objective function 3 is ,ξ i is the start-stop loss conversion coefficient of the i-th water pump;

[0056] What needs to be specifically explained in this embodiment is that the water pump start-stop loss conversion coefficient is related to the performance of the water pump and should be determined according to the water pump's own conditions. The water pump's loss coefficient is related to the water pump's physical properties.

[0057] S52. Determine the constraints of the model: the water flow rate of each water pump in the water pump unit should meet the minimum and maximum flow requirements, that is, , Q i,max is the maximum water flow rate of the i-th water pump in the high-efficiency range. When calculating the minimum flow rate variation, the sum of the flow rates at each moment of the day should be kept consistent with the total flow rate, that is, When calculating the optimal outflow of a pumping station, the maximum and minimum constraints of the storage capacity of the pumping station sump should be met, that is, , Δt is the time change, V B,t is the volume of the pump pool at time t, V B,t-1 is the volume of the pump pool at time t-1. Ensuring that the pump can operate within the high-efficiency range can reduce the energy consumption of the sewage pumping station. Then, , Q i,min is the minimum water flow rate in the high-efficiency range of the i-th water pump. The head of the water pump unit should be greater than or equal to the sum of the static head and the head loss of the pressure pipe, that is, , ε i is the loss coefficient of the i-th water pump, and the pump head and flow rate should be greater than or equal to 0;

[0058] S53, based on the hierarchical sequence method, the objective function is divided into two steps to solve the problem in order. The total flow rate of the pump station and the volume of the sump in objective function 1 are used to solve the optimal uniformity of the sewage pump station. Objective function 2 and objective function 3 can be combined into one objective function because the units of the pump start-stop loss conversion coefficient are consistent. According to the total outflow of the pump station and the flow rate, head and other constraints of each pump obtained by objective 1, the flow distribution and start-stop optimization between units are carried out, and finally the optimal operating conditions are obtained. Therefore, the sewage pump station optimization scheduling model is as follows: , , ;

[0059] S54. After the sewage pump station optimization scheduling model is established, a genetic algorithm is used to optimize the model and finally obtain an optimized scheduling plan.

[0060] S6: Perform scheduling based on the generated scheduling configuration plan, and update the scheduling strategy based on real-time data;

[0061] S7: Collect the actual equipment operation energy consumption data during the coordinated dispatching process of the sewage pumping station, the number of pumps that can be normally activated and the number of pumps that need to be activated in the pumping station, the response time of each device in the pumping station after the dispatching configuration plan is generated, the volume of target particulate impurities remaining in each cubic meter of sewage after pretreatment, the number of starts and stops of each pump used, and the number of sewage overflows;

[0062] Further, the specific data processing process in step S7 is as follows:

[0063] S71, retrieve the expected equipment operation energy consumption, the target particle impurity removal size in sewage, and the expected retention volume of the target particle impurities removed per cubic meter of sewage set by the pump station operation management center, and set the pump station operation data summary and analysis cycle;

[0064] S72. Expected equipment operation energy consumption E during the period of summary and analysis based on pump station operation data y and actual equipment operating energy consumption E x Calculate the energy consumption control capability coefficient X CE , the specific formula is: , based on the expected retention volume V of the remaining target particle impurities per cubic meter of sewage y and the volume V of the remaining target particle impurities to be removed per cubic meter of sewage after pretreatment x Calculate the particle impurity removal accuracy α c , the specific formula is: ;

[0065] S73, based on the number of pumps m required in the i-th scheduling configuration plan bxi and the number of pumps that can be used normally m bzi Calculate the average pump availability L Be , the specific formula is: , n fa is the number of dispatch configuration schemes generated during the pump station operation data summary and analysis period, and the response time t of the jth device in the pump station after the i-th dispatch configuration scheme is generated. xyij Calculate the average device response timeliness rate L sxe , the specific formula is: , n sbi is the number of equipment used after the i-th dispatch configuration plan is generated, and the number of starts and stops of the j-th water pump that can be normally activated in the pump station after the i-th dispatch configuration plan is generated is m qtij Calculate the average pump start-stop rate L qte , the specific formula is: , based on the pump station operation data summary analysis cycle t fx Internal sewage overflow times m wy Calculate the sewage overflow control capacity coefficient X CW , the specific formula is: ;

[0066] S74, based on energy consumption control capability coefficient X CE , particle impurity removal accuracy α c , average pump availability L Be , average equipment response timeliness rate L sxe , average pump start-stop rate L qte , Sewage overflow control capacity coefficient X CW Calculate the pump station operation quality control index P CY The specific formula is: .

[0067] S8: Set the pump station operation data summary analysis cycle, and process the above data to calculate the energy consumption control capacity coefficient, particle impurity removal accuracy, average water pump availability, average equipment response timeliness rate, average water pump start-stop rate, and sewage overflow control capacity coefficient, and then calculate the pump station operation quality control index;

[0068] S9: Back up the pump station coordinated dispatch records and data processing records within the pump station operation data summary and analysis period to the cloud database according to date.

[0069] It should be specifically noted in this embodiment that the preset values, expected values, and coefficients that are not specifically calculated are selected based on actual needs and are not limited to specific values.

[0070] like Figure 2 The present embodiment provides a sewage pump station collaborative scheduling system based on pipeline network predicted liquid level, including a pump station information entry module, a sewage particle impurity pretreatment module, an online monitoring module, a pump pool dynamic liquid level prediction module, a scheduling analysis and determination module, a scheduling configuration scheme generation module, a scheduling configuration scheme execution module, a pump station collaborative scheduling data collection module, a pump station collaborative scheduling data processing module, a pump station operation quality control index calculation module, and a cloud database. The online monitoring module, the pump pool dynamic liquid level prediction module, the scheduling analysis and determination module, the scheduling configuration scheme generation module, the scheduling configuration scheme execution module, the pump station collaborative scheduling data collection module, the pump station collaborative scheduling data processing module, and the pump station operation quality control index calculation module are connected in sequence, the pump station information entry module is connected to the pump pool dynamic liquid level prediction module and the scheduling configuration scheme generation module, the sewage particle impurity pretreatment module is connected to the pump station collaborative scheduling data collection module, and all modules in the system are connected to the cloud database.

[0071] The pump station information input module is used to input the static information, dynamic information of each pump station and the monitoring site information set in the drainage network at the front end of the pump station;

[0072] The sewage particle impurity pretreatment module is used to set a grille or screen at the pump station pump pool as a pretreatment device to remove particle impurities of a preset size in the sewage entering the pump;

[0073] The online monitoring module obtains the real-time liquid level and flow data of different monitoring locations through sensors, and simultaneously obtains the real-time dynamic information of the pump station and sends it to the dynamic liquid level prediction module of the pump pool;

[0074] The pump pool dynamic liquid level prediction module inputs the real-time liquid level and flow data of different monitoring sites and the real-time dynamic information of the pump station into the pump pool dynamic liquid level prediction model to predict the pump pool liquid level of the pump station;

[0075] The scheduling analysis determination module compares the calculated theoretical required future duration with the preset scheduling analysis duration, and determines that scheduling analysis is required when the calculated theoretical required future duration is less than or equal to the preset scheduling analysis duration, otherwise it determines that scheduling analysis is not required;

[0076] When the calculated theoretical required future duration is less than or equal to the preset scheduling analysis duration, the scheduling configuration scheme generation module inputs the current monitoring data and the real-time dynamic information of the pumping station into the pumping station optimization scheduling model to generate a scheduling configuration scheme;

[0077] The scheduling configuration scheme execution module performs scheduling based on the generated scheduling configuration scheme and updates the scheduling strategy based on real-time data;

[0078] The pump station coordinated scheduling data collection module is used to collect the actual equipment operation energy consumption data during the coordinated scheduling of the sewage pump station, the number of pumps that can be normally activated and the number of pumps that need to be activated in the pump station, the response time of each device in the pump station after the scheduling configuration plan is generated, the volume of target particulate impurities remaining in each cubic meter of sewage after pretreatment, the number of starts and stops of each pump used, and the number of sewage overflows;

[0079] The pump station collaborative dispatching data processing module sets the pump station operation data summary and analysis cycle, and calculates the energy consumption control capacity coefficient, particulate impurity removal accuracy, average water pump availability, average equipment response timeliness rate, average water pump start-stop rate, and sewage overflow control capacity coefficient based on the collected pump station collaborative dispatching data;

[0080] The pump station operation quality control index calculation module calculates the pump station operation quality control index based on the data indicators calculated in the pump station collaborative scheduling data processing module;

[0081] The cloud database summarizes and analyzes the pump station operation data and backs up the data of all modules in the system according to date.

[0082] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A sewage pump station coordinated dispatching method based on pipe network predicted liquid level, characterized by: The following steps are involved: S1: Input the static information, dynamic information of each pumping station and the monitoring site information set in the drainage network in front of the pumping station; S2: A screen or mesh is installed at the pump station pump pool as a pre-treatment device to remove particle impurities of a preset size in the sewage entering the pump; S3: Obtain real-time liquid level and flow data at different monitoring locations through sensors, and simultaneously obtain real-time dynamic information of the pump station and input it into the pump pool dynamic liquid level prediction model to predict the pump pool liquid level of the pump station; The specific process of constructing the pump pool dynamic liquid level prediction model in step S3 is as follows: S31: Collect historical operation data of the pump station, historical monitoring data of the monitoring site, and historical environmental data as original data sets, wherein the historical operation data of the pump station includes the number of water pumps in operation, the head of the water pump used, the flow rate and motor speed, the number of starts and stops of the water pump used, and the output flow rate of the pump station; the historical monitoring data of the monitoring site is the liquid level and flow rate synchronized with the time point of the historical operation data of the pump station; and the historical environmental data is the rainfall synchronized with the time point of the historical operation data of the pump station; S32, extracting the number of running pumps, running status of pumps, output flow of pump stations, liquid level of pump pools, liquid level and flow of monitoring points of drainage pipe network, and rainfall at the same time point in the original data set as characteristic parameters; S33, selecting a neural network model as a prediction model, setting a network architecture and basic parameters of the neural network model based on actual needs, dividing the characteristic parameter historical data into a training data set and a test data set, and using the training data set to train the set prediction model; S34, use the test data set to train the trained prediction model, and use the maximum error MAXE, average error ME, mean absolute error MAE, root mean square error RMSE and determination coefficient R 2 Evaluate the prediction accuracy of the pump pool dynamic level prediction model L z , the specific calculation formula is: R 2 The value is between 0 and 1. The closer it is to 1, the better the effect. The closer the other four indicators are to 0, the better the model results. e is a natural constant; S35, comparing the prediction accuracy obtained by the self-set algorithm with the expected value, if it is greater than or equal to the expected value, the pump pool dynamic liquid level prediction model is put into use, otherwise, the pump pool dynamic liquid level prediction model is continuously updated and optimized until the prediction accuracy reaches the expected value; S4: inferring the predicted pump pool liquid level to calculate the future time required for the pump pool liquid level to reach the maximum pump pool liquid level in the current state; The pump pool liquid level predicted in step S4 satisfies the following formula: Q r,t -Q c,t =Z B,t *A B , where Q r,t , Q c,t and Z B,t are the input flow, output flow and liquid level of the pump pool at time t, respectively. B is the cross-sectional area of ​​the pump pool, and the maximum liquid level Z in the pump pool Bmax The calculation formula is: Where V Bmax is the maximum volume of the pump pool, and the theoretical required future time t lx The specific calculation formula is: S5: When the calculated theoretical required future duration is less than or equal to the preset scheduling analysis duration, the current monitoring data and the real-time dynamic information of the pumping station are input into the pumping station optimization scheduling model to generate a scheduling configuration plan; The construction of the pump station optimization scheduling model in step S5 includes the following steps: S51. Determine the objective function of the model: Under the constraint of storage capacity, the pumping station has the smallest flow rate fluctuation and the most stable operation, so the objective function 1 is Q e is the average pumping flow rate of the pumping station, n t is the number of nodes at which the pump station operates throughout the day. When the flow rate is constant, the pump unit consumes the least energy and the pump station operates most reliably. Therefore, objective function 2 is n B is the number of pumps in the pump station, i,t is the start / stop state of the i-th water pump at time t, λ i,t =1 means the water pump is in the on state at time t, λ i,t =0 means the water pump is in shutdown state at time t, t e is the unit time, η bi is the motor efficiency of the i-th water pump, N i,t is the shaft power of the water pump at time t, and the specific calculation formula is: ρ w is the sewage density, g is the gravitational acceleration, Q i,t is the flow rate of the i-th water pump at time t, H i,t is the head of the i-th pump at time t, η ai is the efficiency of the i-th water pump under the working condition at time t. The water pump with the least number of starts and stops corresponds to the longest life and the lowest maintenance cost. Therefore, objective function 3 is ξ i is the start-stop loss conversion coefficient of the i-th water pump; S52. Determine the constraints of the model: the water flow rate of each pump in the pump unit should meet the minimum and maximum flow requirements, that is, 0≤Q i,t ≤Q i,max , Q i,max is the maximum water flow rate of the i-th water pump in the high-efficiency range. When calculating the minimum flow rate variation, the sum of the flow rates at each moment of the day should be kept consistent with the total flow rate, that is, When calculating the optimal outflow of a pump station, the maximum and minimum constraints of the storage capacity of the pump station sump should be met, that is, 0≤V B,t =V B,t-1 +Q r,t *Δt-Q c,t *Δt≤V Bmax , Δt is the time change, V B,t is the volume of the pump pool at time t, V B,t-1 is the volume of the pump pool at time t-1. Ensuring that the pump can operate within the high-efficiency range can reduce the energy consumption of the sewage pumping station. Then, Q i,min ≤Q i,t ≤Q i,max , Q i,min is the minimum water flow rate in the high-efficiency range of the i-th water pump. The head of the water pump unit should be greater than or equal to the sum of the static head and the head loss of the pressure pipe, that is, H i,t ≥Z Bmax -Z B,t +ε i *Q i,t 2 , ε i is the loss coefficient of the i-th water pump, and the pump head and flow rate should be greater than or equal to 0; S53, based on the hierarchical sequence method, the solution is divided into two steps according to the sequence of the objective functions. The total flow rate of the pump station and the volume of the sump in objective function 1 are used to solve the optimal uniformity of the sewage pump station. Objective function 2 and objective function 3 can be combined into one objective function because the units of the pump start-stop loss conversion coefficient are consistent. According to the total outflow of the pump station and the flow and head constraints of each pump obtained by objective function 1, the flow distribution and start-stop optimization between units are carried out, and finally the optimal operating conditions are obtained. Therefore, the sewage pump station optimization scheduling model is as follows: minF=F2+F3, S54. After the sewage pump station optimization dispatching model is established, a genetic algorithm is used to optimize the model and finally obtain an optimized dispatching plan; S6: Perform scheduling based on the generated scheduling configuration plan, and update the scheduling strategy based on real-time data; S7: Collect the actual equipment operation energy consumption data during the coordinated dispatching process of the sewage pumping station, the number of pumps that can be normally activated and the number of pumps that need to be activated in the pumping station, the response time of each device in the pumping station after the dispatching configuration plan is generated, the volume of target particulate impurities remaining in each cubic meter of sewage after pretreatment, the number of starts and stops of each pump used, and the number of sewage overflows; S8: Set the pump station operation data summary analysis cycle, and process the above data to calculate the energy consumption control capacity coefficient, particle impurity removal accuracy, average water pump availability, average equipment response timeliness rate, average water pump start-stop rate, and sewage overflow control capacity coefficient, and then calculate the pump station operation quality control index; The specific data processing process in step S8 is as follows: S81. Retrieve the expected equipment operation energy consumption, target particle impurity removal size in sewage, and expected retention volume of target particle impurities removed per cubic meter of sewage set by the pump station operation management center, and set a pump station operation data summary and analysis cycle; S82. Expected equipment operation energy consumption E during the period of summary and analysis based on pump station operation data y and actual equipment operating energy consumption E x Calculate the energy consumption control capability coefficient X CE , the specific formula is: Expected retention volume V based on the remaining target removal of particulate impurities per cubic meter of wastewater y and the volume V of the remaining target particle impurities to be removed per cubic meter of sewage after pretreatment x Calculate the particle impurity removal accuracy α c , the specific formula is: S83, based on the number of pumps m required in the i-th scheduling configuration plan bxi and the number of pumps that can be used normally m bzi Calculate the average pump availability L Be , the specific formula is: n fa is the number of dispatch configuration schemes generated during the pump station operation data summary and analysis period, and the response time t of the jth device in the pump station after the i-th dispatch configuration scheme is generated. xyij Calculate the average device response timeliness rate L sxe , the specific formula is: n sbi is the number of equipment used after the i-th dispatch configuration plan is generated, and the number of starts and stops of the j-th water pump that can be normally activated in the pump station after the i-th dispatch configuration plan is generated is m qtij Calculate the average pump start-stop rate L qte , the specific formula is: Based on the summary and analysis cycle of pump station operation data t fx Internal sewage overflow times m wy Calculate the sewage overflow control capacity coefficient X CW , the specific formula is: S84, based on energy consumption control capability coefficient X CE , particle impurity removal accuracy α c , average pump availability L Be , average equipment response timeliness rate L sxe , average pump start-stop rate L qte , Sewage overflow control capacity coefficient X CW Calculate the pump station operation quality control index P CY The specific formula is: S9: Back up the pump station coordinated dispatch records and data processing records within the pump station operation data summary and analysis period to the cloud database according to date.

2. A sewage pump station coordinated dispatching method based on pipe network predicted liquid level according to claim 1, characterized in that: The static information of the pump station in step S1 includes the pump station site name, number, affiliated drainage line, associated upstream and downstream pump station names, GIS location, and geographic information; the dynamic information of the pump station includes the number of water pumps contained in the pump station, specific design parameters of the water pumps including water pump head and flow, the maximum number of operating units, the maximum volume of the pump pool, the maximum allowable flow of the pump station, and the minimum allowable flow of the pump station; the monitoring site information set in the drainage network at the front end of the pump station includes the monitoring site number and the spatial geographic location coordinates of the monitoring site.

Images