Global water supply scheduling method and system based on two-stage mixed integer programming algorithm

Abstract

The application belongs to the technical field of city water supply, and provides a global water supply scheduling method based on a two-stage mixed integer programming algorithm, which comprises the following steps: according to historical water consumption data of each water tank in the city, the predicted water consumption of each water tank on the day is calculated; based on the predicted water consumption data, a mathematical model for intelligently determining the peak of a single water tank is established, and the in-out time of the water consumption peak of the single water tank is calculated; based on the in-out time of the water consumption peak of each water tank, a mathematical optimization model for peak-shifting water replenishment of the entire city is established, taking the peak-shifting water replenishment and the minimum valve switching frequency as the target, and the water replenishment parameters of each water tank are determined. According to the pipe connection mode between municipal water supply and the community, the intelligent control means is adopted, the water storage capacity of the water tank is fully utilized, the flow rate of the pipe network in different periods is adjusted, the water supply pressure fluctuation of the municipal pipe network is finally balanced, the balance performance of the entire municipal water supply system is optimized, and the energy-saving purpose is achieved.

Description

Technical Field

[0001] This invention relates to the field of urban water supply scheduling technology, specifically to an intelligent optimization scheduling system and method for balancing water supply and demand across the entire urban area. Background Technology

[0002] The water supply for urban residential communities is provided by the municipal water supply system. The municipal water supply system draws water from the source, which is then treated at the water plant. The clean water is then transported through water pipelines and further pressurized at the community pumping station before being delivered to users.

[0003] Urban residential water supply typically consists of two parts. The first part is direct supply from the municipal pipeline to the residential area. The pressure of the direct supply is relatively low, and the direct supply mainly serves residents with 4 floors or less. The other part is secondary water tank storage. Water is delivered to each water tank through water pipes, and then the water tanks supply the water.

[0004] Because urban residents' water usage patterns exhibit distinct peak and off-peak periods, the scheduling of secondary water tanks requires replenishing water during off-peak times and utilizing the tank's stored water to meet user demand during peak periods. This necessitates predicting peak water usage times and automatically replenishing the tank after the set peak time has elapsed until the tank is full.

[0005] In addition, the water supply modes vary from community to community. Some communities use a pure water tank mode, meaning that the entire water supply for the community is provided by a water tank. In this mode, to meet residents' water needs, the valve is kept open during off-peak hours; that is, the valve is opened immediately after peak hours to replenish water until the next peak period begins and then closed. Other communities use a combined or jet-type mode. In these two modes, two water pumps are installed at the water tank inlet and the municipal water network, allowing residents to draw water from both the tank and the municipal network simultaneously. This requires replenishment after peak hours, and replenishment stops once a certain amount of water has been replenished.

[0006] However, the current peak hours for water inflow and outflow are set by technicians using their experience and knowledge. But because there are a large number of water tanks in the city (more than 100), and the peak hours for each community are very close, the manual experience setting leads to a large number of water tanks being replenished at the same time after the peak hours, which causes the pressure fluctuation of the municipal water supply network to increase. Pressure fluctuation will reduce the service life of the water supply network and increase energy consumption. Summary of the Invention

[0007] The present invention aims to at least partially solve one of the technical problems existing in the related art.

[0008] Therefore, one objective of this invention is to provide a global water supply scheduling method based on a two-stage mixed integer programming algorithm, which can reduce pressure fluctuations in the municipal water supply network while meeting the water supply needs of water storage tanks, thereby achieving the rational allocation and utilization of urban water resources.

[0009] Another objective of this invention is to provide a global water supply scheduling system based on a two-stage hybrid integer programming algorithm.

[0010] To achieve the above objectives, the present invention provides a method for global water supply scheduling based on a two-stage mixed integer programming algorithm, comprising the following steps:

[0011] S100. Based on the historical water consumption data of each water tank in the city, calculate the predicted water consumption of each water tank for the day.

[0012] S200. Based on predicted water consumption data, establish a mathematical model for intelligently determining peak water usage for a single water tank, and calculate and obtain the entry and exit times of peak water usage for a single water tank.

[0013] S300. Based on the peak water usage times of each water tank, and with the goal of staggered water replenishment and minimizing the number of valve switching times, a mathematical optimization model for staggered water replenishment of the entire city is established to determine the water replenishment parameters of each water tank.

[0014] Preferably, step S100, which involves calculating the predicted water consumption of each water tank for the day based on historical water consumption data of each water tank in the city, specifically includes:

[0015] S110. First, collect historical water consumption data for a single water tank;

[0016] S120. Discretize the collected historical water consumption data and divide the daily water consumption data into several time intervals.

[0017] S130. Classify historical water consumption data according to whether it is a working day or not, and take the average of recent data of the same type as the predicted day as the predicted water consumption for that day. The predicted water consumption for that day is displayed in the corresponding time interval.

[0018] Preferably, the historical water consumption data is obtained based on the level difference of the water tank level gauge and the bottom area of ​​the water tank.

[0019] Preferably, step S200, which involves calculating the peak water usage times for a single water tank based on predicted water consumption data, specifically includes:

[0020] S210. According to the difference in water supply mode, the water tanks are classified into pure water tank mode, combined mode and jet mode.

[0021] S220. Determine the peak water usage time for each water tank based on its water supply mode:

[0022] A. Peak water usage times in pure water tank mode:

[0023] In pure water tank mode, the valve is always open during off-peak hours, and all residential water use is provided by the water tank.

[0024] The peak water usage time in pure water tank mode is then determined using a peak detection algorithm.

[0025] B. Determine the peak water usage times for the combined or jet mode:

[0026] The combined mode and jet mode replenish water to the water tank during off-peak hours, and the water supply for residents is provided by the water tank and the municipal water network at the same time.

[0027] Based on the predicted water consumption data, a peak detection algorithm is used to determine the upper and lower bounds of the morning and evening peak times. Then, the predicted water consumption data and the upper and lower bounds of the morning and evening peak times are used as inputs. With the goal of maximizing peak time, minimizing valve switching, and pumping municipal water from the municipal pipeline to replenish the water tank during off-peak hours, a mathematical model for intelligently determining peak times for a single water tank is established, and the entry and exit times of peak water consumption for each water tank are solved.

[0028] Preferably, the step of determining the peak water usage time in pure water tank mode using the peak detection algorithm is as follows:

[0029] First, based on the predicted water consumption data, 5:00-14:00 and 17:00-24:00 were selected as the morning and evening peak times, respectively.

[0030] Then, using a peak detection algorithm, the peak water consumption and the time of entering the peak water consumption are calculated for the two alternative time periods: 5:00-14:00 and 17:00-24:00.

[0031] Finally, the peak water usage time is searched from the peak water usage time to both ends, and the effective storage capacity of the water tank is used as the search termination criterion to determine the peak water usage time.

[0032] Preferably, in step S300:

[0033] Based on the peak water usage inflow and outflow times calculated from the water tanks using the combined mode and jet mode, a mathematical optimization model for staggered water replenishment for the entire city is established to determine the water replenishment parameters for each water tank.

[0034] The water replenishment volume at each time point is calculated based on the peak water usage time of the pure water tank mode, and this is used as the benchmark for the municipal pipeline flow rate.

[0035] On the other hand, the present invention provides a global water supply scheduling system based on a two-stage hybrid integer programming algorithm, comprising:

[0036] The water plant, several water tanks supplying water to residents, and an intelligent control center;

[0037] The water plant is connected to the water tanks through the municipal pipeline network, and several sensors are installed in the municipal pipeline network and the water tanks; the inlet and outlet of each water tank are controlled by control valves.

[0038] The control valves and sensors are connected to the intelligent control center via a communication network.

[0039] The intelligent control center includes:

[0040] The water tank water consumption acquisition module collects historical water consumption data of each water tank in the city through the sensor and calculates the predicted water consumption of each water tank for the day.

[0041] The module for obtaining peak water usage time of water tanks is used to calculate the peak water usage time of a single water tank based on the obtained predicted water usage data.

[0042] The peak-shaving water replenishment intelligent decision-making module is used to establish a mathematical optimization model for peak-shaving water replenishment of the entire city based on the peak water usage times of each water tank, with the goal of peak-shaving water replenishment and minimizing the number of valve switching times. It determines the water replenishment parameters for each water tank and sends opening and closing control signals to the control valves through the communication network.

[0043] The present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the above-described global water supply scheduling method based on a two-stage hybrid integer programming algorithm.

[0044] The present invention also provides a non-transitory computer-readable storage medium storing computer instructions thereon, which cause the computer to execute the above-described global water supply scheduling method based on a two-stage mixed integer programming algorithm.

[0045] In another aspect, the present invention provides an electronic device, comprising: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus, and the processor calls logical instructions in the memory to execute the above-mentioned global water supply scheduling method based on a two-stage mixed integer programming algorithm.

[0046] This invention solves a mathematical optimization problem to determine the replenishment time and duration of urban water tanks, thereby maximizing the tank's regulation and storage capacity. This ensures water supply during peak usage periods and replenishment during off-peak periods, reducing peak replenishment rates and achieving balanced water supply pressure. Furthermore, based on the pipeline connection between municipal water supply and residential areas, this invention employs intelligent control methods to fully utilize the water tank's storage capacity, regulate the flow rate of the pipe network at different times, and ultimately balance the fluctuations in municipal water supply pressure, optimizing the overall balance performance of the municipal water supply system and achieving energy conservation. Attached Figure Description

[0047] Figure 1 This is a flowchart of the global water supply scheduling method based on a two-stage hybrid integer programming algorithm of the present invention;

[0048] Figure 2 This is a predicted water consumption chart for the current day based on historical water consumption data in an embodiment of the present invention;

[0049] Figure 3 This is a diagram illustrating the effect of staggered water replenishment according to the present invention;

[0050] Figure 4 This invention adds a water replenishment effect diagram showing the water replenishment rate at times 0-5;

[0051] Figure 5 This is a comparison chart of the number of residential areas experiencing morning rush hour at different times according to the present invention. Detailed Implementation

[0052] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the embodiments described.

[0053] Example 1: This example provides a global water supply scheduling method based on a two-stage mixed integer programming algorithm, including the following steps:

[0054] S100. Based on the historical water consumption data of each water tank in the city, calculate the predicted water consumption of each water tank for the day.

[0055] Specifically, it includes:

[0056] S110. First, collect historical water consumption data for a single water tank. The water consumption at different times of the day can be obtained based on the liquid level difference of the level gauge and the bottom area of ​​the water tank.

[0057] S120. The collected historical water consumption data is discretized. In this embodiment, time discretization is performed at 5-minute intervals, dividing the day into 288 time intervals such as 0:00-0:05, 0:05-0:10, ..., 23:45-0:00. Since the water tank is only equipped with a level gauge, if the level in a certain interval does not change, the data is filled with zeros.

[0058] S130. Historical water consumption data is categorized according to whether it is a working day. In this embodiment, dates are divided into four types: working days, Saturdays, Sundays, and public holidays. During prediction, data from the most recent 20 days of the same type as the current day is extracted from the historical water consumption data, and the average flow rate of these 20 days is used as the predicted water consumption for that day. The predicted water consumption for the day is displayed in the corresponding time interval.

[0059] Figure 2 This is a comparison chart of the peak flow prediction based on liquid level output from the prediction model and the actual peak water consumption in the example. Figure 2 It can be seen that peak water consumption exhibits a regularity historically, meaning that peak water consumption at the same time on different dates is relatively similar. The prediction results show that the output flow curve matches the actual flow curve, validating the accuracy of the prediction model.

[0060] S200. Based on the predicted water consumption data, calculate and obtain the peak water consumption entry and exit times for a single water tank;

[0061] Specifically, it includes:

[0062] S210. According to the difference in water supply mode, the water tanks are classified into pure water tank mode, combined mode and jet mode.

[0063] S220. Determine the peak water usage time for each water tank based on its water supply mode:

[0064] A. Peak water usage times in pure water tank mode:

[0065] For the pure water tank mode, to meet residents' water demand, the valve is kept open during off-peak hours. This means that the valve opens immediately upon the start of peak hours to replenish water, and remains open until the next peak period begins. Due to this unique characteristic of the pure water tank, replenishment cannot be delayed after peak hours to stagger peak demand, resulting in very limited peak-shaving capacity. Therefore, a peak time is determined directly using a peak detection algorithm. The goal is to maximize peak time within the tank's storage capacity, near peak water usage periods. The specific process is as follows:

[0066] (1) Based on the predicted water consumption data, 5:00-14:00 and 17:00-24:00 are selected as the morning and evening peak times, respectively.

[0067] (2) Using the peak detection algorithm, calculate the peak water consumption and the time of entering the peak water consumption for the two alternative time periods of 5:00-14:00 and 17:00-24:00.

[0068] (3) Search for peak time from peak water usage time to both ends, and use the effective storage capacity of the water tank as the search termination criterion to determine the peak water usage time.

[0069] This algorithm can determine the peak inflow and outflow times of the water tank in pure water tank mode, with the peak outflow time being the starting point for determining water replenishment. Based on the tank capacity and replenishment rate, the amount of water to be replenished at various times of the day can be determined. If no water is replenished at a certain time, the amount of water replenished at that time is 0.

[0070] B. Determine the peak water usage times for the combined or jet mode:

[0071] For the combined mode and the jet mode, two water pumps were installed at the water tank inlet and the municipal pipe network, so that residents can use water from both the water tank and the municipal pipe network at the same time.

[0072] Based on the predicted water consumption data, a peak detection algorithm is used to determine the upper and lower bounds of the morning and evening peak times. Then, the predicted water consumption data and the upper and lower bounds of the morning and evening peak times are used as inputs. With the goal of maximizing peak time, minimizing valve switching, and pumping municipal water from the municipal pipeline to replenish the water tank during off-peak hours, a mathematical model for intelligently determining peak times for a single water tank is established, and the entry and exit times of peak water consumption for each water tank are solved.

[0073] The mathematical model for intelligent peak determination in a single water tank is shown below:

[0074]

[0075] V t-1 +y t +M t =D t +V t ,t=1,…,T;

[0076] M t =D t m t ,t=1,…,T;

[0077] V t ≤V max ,t=1,…,T;

[0078]

[0079] p1≥x1-Δ;

[0080] p t ≥x t ,-x t-1 t = 2, ..., T;

[0081]

[0082]

[0083] x t ∈{0,1},t=1,…,T;

[0084] m t ∈{0,1},t=1,…,T;

[0085] z j,t ∈{0,1},t=1,…,T;

[0086] In the above formula, β: model normalization parameter; ζ: model normalization parameter; D t : Predicted water consumption of the community at time t; V max : Maximum regulating capacity of the water tank; q j The water tank replenishment rate when the valve is in position J; The lower bound of the start time of the morning rush hour; The upper limit of the start time of the morning rush hour; The lower bound of the morning rush hour end time; The upper limit of the morning rush hour end time; The lower bound of the start time of the evening rush hour; The upper limit of the start time of the evening rush hour; The lower bound of the evening rush hour end time; Upper bound of the evening rush hour end time; Δ: Valve status at dawn; T: Number of moments in a day after discretization;

[0087] The relevant variables for the problem are:

[0088] Morning rush hour start time; Morning rush hour end time; Evening rush hour start time; Evening rush hour end time; y t : Water supply over t time periods; p t : Whether the valve is switched at time t; m_t: Whether the municipal direct supply valve is opened during time period t; M t : Virtual traffic flowing into the community during time period t; z jt : Whether the valve opening at time t is at position j; x t : Whether the valve is opened at time t; V t : The water volume in the water tank of the i-th cell at time t.

[0089] S300. By solving the above mathematical optimization model, the peak time for each community can be obtained. In the planning of all water tanks in the entire city, it is only necessary to consider the water replenishment time of the combined mode and jet mode water tanks. Then, the water replenishment volume calculated by the pure water tank at each time is used as the benchmark for the pipeline flow. This allows the water replenishment time of the combined mode and jet mode water tanks to be staggered with that of the pure water tank mode water tanks, reducing the fluctuation range of the pipeline flow and achieving the purpose of energy saving.

[0090] Based on the peak water usage times of water tanks in the combined mode and jet mode, and with the goal of staggered water replenishment and minimizing the number of valve switching times, a mathematical optimization model for staggered water replenishment of the entire city is established to determine the water replenishment parameters of each water tank.

[0091] The established mathematical optimization model for staggered water replenishment is as follows:

[0092]

[0093] V i,t-1 +y it =D i,t +V i,t ,i∈1,…,N,t=1,…,T;

[0094]

[0095] p i1 ≥x 1,t -Δ i i = 1, ..., N;

[0096] p it ≥x i,t -x i,t-1 ,t=2,…,T,i=1,…,N;

[0097]

[0098] x i,t ∈{0,1},i=1…,N,t=1,…,T;

[0099] z i,j,t ∈{0,1},i=1…,N,j=1,…J i ,t=1,…,T;

[0100] p it ≥0, t=1,…,T,i=1,…,N.

[0101] in,

[0102] β: Model normalization parameter; r: The r-th region; The set of all regions; N: the number of cells in the joint mode and jet mode; T: the number of times after discretization of a day; i: cell i; D i,t : Predicted water consumption of cell i at time t; The maximum storage capacity of the water tank in the i-th community; q ij : The water tank replenishment rate when the valve of the i-th community is opened to position j; The start time of the morning rush hour in the i-th residential area; The end time of the morning rush hour for the i-th residential area; The start time of the morning rush hour in the i-th residential area; The end time of the morning rush hour for the i-th residential area; Δ i : Valve status of the i-th cell at midnight; Water replenishment volume of the pure water tank in the r-th area during the t-th time period;

[0103] The relevant variables for the problem are:

[0104] η r : Peak water replenishment rate for the r-th area; η: Peak water replenishment rate for all water tanks in the entire city's pipe network; y it p: Water supply for the i-th cell during the t-th time period; i,t : Whether the valve is switched at time t in the i-th cell; z ijt : Whether the valve opening at time t in the i-th cell is at position j; x it : Whether the valve is opened at time t in the i-th cell; V it : The water volume in the water tank of the i-th cell at time t.

[0105] Example 2: A global water supply scheduling system based on a two-stage mixed integer programming algorithm, comprising:

[0106] The water plant, several water tanks supplying water to residents, and an intelligent control center;

[0107] The water plant is connected to the water tanks through the municipal pipeline network, and several sensors are installed in the municipal pipeline network and the water tanks; the inlet and outlet of each water tank are controlled by control valves.

[0108] The control valves and sensors are connected to the intelligent control center via a communication network.

[0109] The intelligent control center includes:

[0110] The water tank water consumption acquisition module collects historical water consumption data of each water tank in the city through the sensor and calculates the predicted water consumption of each water tank for the day.

[0111] The module for obtaining peak water usage time of water tanks is used to calculate the peak water usage time of a single water tank based on the obtained predicted water usage data.

[0112] The peak-shaving water replenishment intelligent decision-making module is used to establish a mathematical optimization model for peak-shaving water replenishment of the entire city based on the peak water usage times of each water tank, with the goal of peak-shaving water replenishment and minimizing the number of valve switching times. It determines the water replenishment parameters for each water tank and sends opening and closing control signals to the control valves through the communication network.

[0113] Figure 3 This is a diagram illustrating the effect of the staggered water replenishment method of this invention. The experiment used 98 water tanks, and the model involved tens of thousands of decision variables. Therefore, the total solution time for the mathematical model was set to 5 minutes, and the relative gap in the second stage was set to 15%. Within this relatively limited time, the model of this invention can run stably and output the specified valve opening and closing. Figure 3 The blue line in the image represents the water supply rate when all valves are opened simultaneously, which is close to 500m³ / h. 2 The peak water replenishment rate is very high, at 5 minutes per minute. The current system uses manually set peak entry and exit times, indicated by the yellow line in the diagram. Due to manual peak-shaving measures implemented in some residential areas, the peak water replenishment rate has decreased to approximately 300 minutes per minute. 2 / 5min. The output of the two-stage mathematical model of this invention is shown as a red line, which achieves peak shaving and water replenishment, reducing the peak water replenishment rate to slightly over 100m. 2 / 5min. The reduction in peak water replenishment also significantly reduces energy consumption. In particular, the reduction in peak water replenishment allows the pipeline network to maintain a relatively stable pressure value. Based on the relationship between energy consumption, peak water replenishment, and pipeline pressure, it can be calculated that the theoretical energy consumption value of this off-peak water replenishment can be reduced by 10.4%, achieving a significant energy-saving and consumption-reducing effect.

[0114] In actual pipe network systems, the pressure on the network varies due to different water consumption patterns at different times. Ideally, the chosen water replenishment time should not only smooth out peak flows in the considered pump stations but also ensure that these communities receive water during the overall low water consumption period of the pipe network. The advantage of replenishing water during the overall low water consumption period is that the pipe network pressure is sufficient at this time, thus minimizing the pressure impact on the pipe network caused by replenishing water to the pump station tanks. Currently, due to inaccurate instruments and significant noise from pressure fluctuations, only the periods of lower pipe network pressure can be roughly identified. Empirical data shows that water consumption is low and pressure is relatively sufficient between 0:00 and 5:00 AM, making it suitable for replenishing water tanks. For this period of lower overall pipe network water consumption, the peak water replenishment rate can be appropriately increased to fully utilize the water tank's storage capacity and reduce the replenishment pressure at the morning peak. Considering this practical situation, a mathematical model is used to recalculate the water replenishment rate, such as... Figure 4 As shown. By Figure 4 It can be seen that the water replenishment rate from 0:00 to 5:00 is relatively... Figure 3 It increased by approximately 20%. However, the peak rate of water replenishment during the period from the end of the morning rush hour to the beginning of the evening rush hour was compared to... Figure 3 It decreased by 20%, reducing the pressure on the pipeline network before the evening peak.

[0115] Figure 5 The number of communities experiencing peak hours at different times was described. It is not difficult to find that, overall, the morning peak time given by the intelligent decision-making program is delayed compared to the historical strategy. However, the intelligent decision-making program can uniformly determine the peak time at 13:30 in the afternoon and carry out water replenishment in an orderly manner to prepare for the evening peak water replenishment.

[0116] The system of this invention relies on a central control server. By adjusting the electric valves of the secondary water supply tanks in each community within the regional system, the system achieves decentralized water replenishment time, effectively reducing peak flow during the replenishment process. During peak water usage periods, the system automatically closes the electric valves and utilizes the water stored in the tanks for supply, thereby alleviating water pressure during peak hours and achieving energy conservation.

[0117] This invention relates to the installation of level gauges, flow meters, and electric valves in each residential community. These devices can communicate with a control server center in real time. The community's level data is sent to a regional water supply balance intelligent optimization scheduling system, which converts the level and flow data into the actual water consumption of the community's residents. Based on historical water consumption data, the system predicts the water consumption data for the following day. Based on the water consumption data, the system solves a two-stage mathematical optimization model. In the first stage, based on the predicted water consumption data and considering the independence of each community, the system determines the peak time for each community in parallel. After solving the model, the peak entry and exit times for each community are distributed to each pumping station, restricting water supply to use only water stored in water tanks during peak hours. In the second stage, based on the peak times from the first stage, the system solves a second-stage mathematical optimization model to obtain the valve opening and closing times for each community, and distributes this information to each pumping station to control water consumption scheduling.

[0118] This invention solves a mathematical optimization problem to determine the opening and closing times and duration of the electric valve, thereby maximizing the water tank's regulation and storage capacity. This ensures water supply from the tank during peak usage periods and replenishment during off-peak periods, reducing peak replenishment rates and achieving balanced water supply pressure. Based on the pipeline connection between municipal water supply and residential areas, this invention employs intelligent control methods to fully utilize the water tank's storage capacity, regulate the flow rate in the pipe network at different times, and ultimately balance the fluctuations in municipal water supply pressure, optimizing the overall balance performance of the municipal water supply system and achieving energy conservation.

[0119] In some embodiments of the present invention, a computer program product is provided, the computer program product including a computer program that can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute the above-described global water supply scheduling method based on a two-stage hybrid integer programming algorithm.

[0120] In some embodiments of the present invention, a non-transitory computer-readable storage medium is provided, on which computer instructions are stored, which cause the computer to execute the above-described global water supply scheduling method based on a two-stage hybrid integer programming algorithm.

[0121] In some embodiments of the present invention, an electronic device is provided, which may include a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The processor can call logical instructions in the memory to execute the above-described global water supply scheduling method based on a two-stage mixed integer programming algorithm.

[0122] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0123] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0124] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0125] As described above, although the invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the invention itself. Various changes in form and detail may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A global water supply scheduling method based on a two-stage mixed integer programming algorithm, characterized in that, Includes the following steps: S100. Based on the historical water consumption data of each water tank in the city, calculate the predicted water consumption of each water tank for the day. S200. According to the difference in water supply mode, the water tanks are classified into pure water tank mode, combined mode and jet mode; and based on the predicted water consumption data, a mathematical model for intelligently determining the peak water consumption of a single water tank is established to calculate the inflow and outflow time of the peak water consumption of a single water tank. S300: Based on the peak water usage times of water tanks in the combined mode and jet mode, a mathematical optimization model for peak water usage is established for the entire city with the goal of staggered water replenishment and minimizing the number of valve switching. The water replenishment parameters of each water tank are determined. The water replenishment volume at each time is calculated based on the peak water usage times of the pure water tank mode, and this is used as the benchmark for the municipal pipeline flow. The established mathematical optimization model for staggered water replenishment is as follows: in, : Model normalization parameters; : No. Each area; A collection of all regions; The number of cells in the combined mode and the jet mode; The number of moments after a day is discretized; : residential area ; Predicted Water consumption in the community at any given time; : The maximum storage capacity of the water tank in the i-th community; The water tank replenishment rate when the valve of the i-th community is opened to position j; : The start time of the morning rush hour for the i-th community, where 1 represents the morning rush hour; : The end time of the morning rush hour for the i-th residential area; : The start time of the evening rush hour for the i-th residential area, where 2 represents the evening rush hour; : The end time of the evening rush hour for the i-th residential area; : Valve status of the i-th cell at midnight; : The amount of pure water replenished to the water tank in the r-th area during the t-th time period; The relevant variables for the problem are: : The peak rate of water replenishment in the r-th region; Peak water replenishment rate for all water tanks in the entire city's pipe network; : The amount of water supplied to the i-th cell during the t-th time period; : Whether the valve is switched at time t in the i-th cell; : Whether the valve opening is at position j at time t in the i-th cell; : Whether the valve is opened in the i-th cell at time t; : The water volume in the water tank of the i-th cell at time t.

2. The global water supply scheduling method based on a two-stage mixed integer programming algorithm according to claim 1, characterized in that, Step S100, which involves calculating the predicted water consumption for each water tank on a given day based on historical water consumption data, specifically includes: S110. First, collect historical water consumption data for a single water tank; S120. Discretize the collected historical water consumption data and divide the daily water consumption data into several time intervals. S130. Classify historical water consumption data according to whether it is a working day or not, and take the average of recent data of the same type as the predicted day as the predicted water consumption for that day. The predicted water consumption for that day is displayed in the corresponding time interval.

3. The global water supply scheduling method based on a two-stage mixed integer programming algorithm according to claim 2, characterized in that, The historical water consumption data is obtained based on the level difference of the water tank level gauge and the bottom area of ​​the water tank.

4. The global water supply scheduling method based on a two-stage mixed integer programming algorithm according to claim 1, characterized in that, Step S200, which involves calculating the peak water usage times for a single water tank based on predicted water consumption data, specifically includes: S210. According to the difference in water supply mode, the water tanks are classified into pure water tank mode, combined mode and jet mode. S220. Determine the peak water usage time for each water tank based on its water supply mode: A. Peak water usage times in pure water tank mode: In pure water tank mode, the valve is always open during off-peak hours, and all residential water use is provided by the water tank. The peak water usage time in pure water tank mode is then determined using a peak detection algorithm. B. Determine the peak water usage times for the combined or jet mode: The combined mode and jet mode replenish water to the water tank during off-peak hours, and the water supply for residents is provided by the water tank and the municipal water network at the same time. Based on the predicted water consumption data, a peak detection algorithm is used to determine the upper and lower bounds of the morning and evening peak times. Then, the predicted water consumption data and the upper and lower bounds of the morning and evening peak times are used as inputs. With the goal of maximizing peak time, minimizing valve switching, and pumping municipal water from the municipal pipeline to replenish the water tank during off-peak hours, a mathematical model for intelligently determining peak times for a single water tank is established, and the entry and exit times of peak water consumption for each water tank are solved.

5. The global water supply scheduling method based on a two-stage mixed integer programming algorithm according to claim 4, characterized in that, The steps for determining peak water usage times in pure water tank mode using the peak detection algorithm are as follows: First, based on the predicted water consumption data, 5:00-14:00 and 17:00-24:00 were selected as alternative times for the morning and evening peak hours, respectively. Then, using a peak detection algorithm, the peak water consumption and the time of entering the peak water consumption are calculated for the two alternative time periods: 5:00-14:00 and 17:00-24:

00. Finally, the peak water usage time is searched from the peak water usage time to both ends, and the effective storage capacity of the water tank is used as the search termination criterion to determine the peak water usage time.

6. A global water supply scheduling system based on a two-stage mixed integer programming algorithm, characterized in that, include: The water plant, several water tanks supplying water to residents, and an intelligent control center; Each water tank is classified according to its water supply mode, including pure water tank mode, combined mode and jet mode; The water plant is connected to the water tanks through the municipal pipeline network, and several sensors are installed in the municipal pipeline network and the water tanks; the inlet and outlet of each water tank are controlled by control valves. The control valves and sensors are connected to the intelligent control center via a communication network. The intelligent control center includes: The water tank water consumption acquisition module collects historical water consumption data of each water tank in the city through the sensor and calculates the predicted water consumption of each water tank for the day. The module for obtaining peak water usage time of water tanks is used to calculate the peak water usage time of a single water tank based on the obtained predicted water usage data. The peak-shaving water replenishment intelligent decision-making module is used to determine the inflow and outflow times of water tanks in both combined and jet modes based on peak water usage. With the goal of peak-shaving water replenishment and minimizing valve switching times, it establishes a mathematical optimization model for peak-shaving water replenishment for the entire city, determines the water replenishment parameters for each water tank, and sends opening and closing control signals to the control valves via a communication network. Based on the peak water usage inflow and outflow times in pure water tank mode, it calculates the water replenishment volume for each time period and uses this as a benchmark for the municipal pipeline flow rate. The established mathematical optimization model for staggered water replenishment is as follows: in, : Model normalization parameters; : No. Each area; A collection of all regions; The number of cells in the combined mode and the jet mode; The number of moments after a day is discretized; : residential area ; Predicted Water consumption in the community at any given time; : The maximum storage capacity of the water tank in the i-th community; The water tank replenishment rate when the valve of the i-th community is opened to position j; : The start time of the morning rush hour for the i-th community, where 1 represents the morning rush hour; : The end time of the morning rush hour for the i-th residential area; : The start time of the evening rush hour for the i-th residential area, where 2 represents the evening rush hour; : The end time of the evening rush hour for the i-th residential area; : Valve status of the i-th cell at midnight; : The amount of pure water replenished to the water tank in the r-th area during the t-th time period; The relevant variables for the problem are: : The peak rate of water replenishment in the r-th region; Peak water replenishment rate for all water tanks in the entire city's pipe network; : The amount of water supplied to the i-th cell during the t-th time period; : Whether the valve is switched at time t in the i-th cell; : Whether the valve opening is at position j at time t in the i-th cell; : Whether the valve is opened in the i-th cell at time t; : The water volume in the water tank of the i-th cell at time t.

7. A computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is capable of executing the global water supply scheduling method based on a two-stage hybrid integer programming algorithm as described in any one of claims 1-5.

8. A non-transitory computer-readable storage medium having stored thereon computer instructions that cause a computer to execute the global water supply scheduling method based on a two-stage hybrid integer programming algorithm as described in any one of claims 1-5.

9. An electronic device, comprising: The system includes a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other through the communication bus. The processor calls logical instructions in the memory to execute the global water supply scheduling method based on a two-stage hybrid integer programming algorithm as described in any one of claims 1-5.

Images