An open water system leakage monitoring method and device based on statistical characteristics

Abstract

The application discloses an open water system leakage monitoring method and device based on statistical characteristics, and relates to the technical field of water system leakage monitoring. The method comprises the following steps: obtaining historical data and instant data of an open water system to be detected; constructing an open water system model according to the historical data; and obtaining a leakage condition and a leakage position of the open water system according to the historical data, the instant data, the open water system model and an online leakage monitoring algorithm. According to common data in the water system, the application establishes an open water system model; through analysis of statistical data of the water system under the model, in combination with instant data, an online leakage monitoring algorithm is designed, which is used for judging the leakage condition and determining the leakage position.

Description

Technical Field

[0001] This invention relates to the field of water system leakage monitoring technology, and in particular to an open water system leakage monitoring method and apparatus based on statistical characteristics. Background Technology

[0002] Open water systems are a type of water system distinct from closed water systems, characterized by the absence of a closed water path. They are widely used in various cooling water systems, such as cooling tower water systems in industries like food, pharmaceuticals, metallurgy, plastics, textiles, and chemicals. Open water systems are also commonly used in air conditioning and refrigeration equipment, as well as generator cooling systems, where precision requirements are less stringent. Furthermore, common water usage scenarios such as fire fighting, drainage systems, and agricultural irrigation can all be classified as open water systems; even water conservancy projects fall under the broader category of open water systems.

[0003] Compared to closed-loop water systems, which offer stable and rapid pressure response and a globally integrated water system structure, open-loop water systems exhibit poor pressure stability, slow response to pressure changes, and fluid exchange with the external environment. This makes it difficult to apply common methods such as Kirchhoff's laws to analyze water system characteristics. Taking the cooling water system of a continuous casting machine as an example, it uses nozzles supplied by an open water system to spray a mixture of water vapor onto the surface of the cast billet for cooling. The cooled water then flows into an open water tank for treatment and reuse. Furthermore, the upstream cooling water is supplied by a variable frequency pump and an elevated water tank, and the pressure cannot be guaranteed to remain constant throughout the entire operation from supply to drainage. Therefore, controlling and diagnosing faults in open-loop water systems presents significant challenges.

[0004] Existing literature has studied the fault diagnosis of specific components (such as valves, nozzles, water-consuming equipment, etc. in the water circuit where the valve is located) in open water systems. Pu Dongqing. Research on the characteristics and identification methods of blockage and leakage faults in cooling water systems [D]. Huazhong University of Science and Technology, 2017. This paper analyzes the main causes of nozzle blockage and leakage from the aspects of water quality control, process equipment configuration and maintenance, production water operation and maintenance management, and proposes relevant control measures. Duan Lanlan. Fault diagnosis of heating pipe networks based on genetic optimization BP neural network [D]. Taiyuan University of Technology, 2014. From the perspective of valve selection, this paper introduces the structural composition of the sliding plate valve, its advantages, and its specific application in the secondary cooling water regulation system of the billet continuous casting machine; Wei Xiangxiang. Research on pipeline leakage diagnosis method based on fuzzy model [D]. Northeastern University, 2011. Based on the HART7.0 communication protocol, this paper designs and develops a fault diagnosis tool for valves using Windows Forms in the Visual Studio 2019 environment, which can realize normal communication with valves and obtain and set valve parameters.

[0005] Furthermore, some research results have been achieved regarding the failure principles, solutions, and compensation measures for open water systems. For example, the flow rate variation of the generator stator cooling water system of a 660MW supercritical unit was analyzed, verifying its correlation with the copper corrosion rate; an energy efficiency control method for open-loop industrial cooling water systems was studied; a control method for open-loop circulating water cooling systems that compensates for evaporation was provided; research and modification of open cooling water systems in thermal power plants were discussed; a method for estimating leakage in high-pressure fire-fighting water supply systems was proposed to ensure that indoor temporary high-pressure fire-fighting water supply systems equipped with pressure stabilization facilities meet the requirements for automatic start-up of fire pumps by pressure switches; and a large-scale data acquisition solution was proposed for pipe networks using sensor testing technology, along with an equivalent pipe network model for leakage conditions, providing a theoretical explanation for the analysis of leakage patterns.

[0006] Unlike open water systems, closed water systems have pressure stabilizing equipment such as pressure tanks, resulting in stable water pressure and fast pressure signal transmission speed, making them suitable for a wide range of leak detection methods. For example, Cao Huizhe, He Zhihong, and He Zhongyi proposed a second-order BP neural network-based model for leak diagnosis of heating pipe networks based on genetic optimization in their work "Calculation Research on Slow Flow of Ring Pipe Network Based on Graph Theory [J]. Journal of Harbin Institute of Technology, 2007(10):1559-1563". Wu Yuebin and Liu Tianshun conducted an in-depth study on leak detection of oil pipelines based on the fuzzy TS model and the generalized fuzzy hyperbolic tangent model, taking the fault diagnosis problem in pipeline transportation as the background. Wu Shaoke. Research on Leakage Location of Ship Water Pipeline System Based on Pressure Gradient Method [D]. Dalian Maritime University, 2017. The hydraulic process of pipeline load variation can be abstracted into a slow variable flow in unsteady flow for research, and a calculation method for slow variable flow in ring pipeline based on graph theory is proposed; Li Wuyang, Ge Wei. Computer Simulation of Mathematical Model of Low Pressure Micro Leakage [J]. Hydraulics, Pneumatics and Sealing, 1992(02):2-5. The theoretical framework of leakage numerical simulation based on hydraulic transient analysis is proposed, pointing out that the essence of this problem is to solve the inverse problem of system identification and perform parameter identification; a ship pipeline transmission model is established, a ship ballast water system is built, and the equivalent length method is used to study the leakage location of the system; a mathematical model of low pressure micro leakage working condition is established, providing a basis for the design of low pressure micro leakage detection device.

[0007] In summary, existing literature on water system leak detection methods mainly focuses on closed water systems; while research on open water systems either emphasizes the fault diagnosis of a specific component or the mechanistic study of the fault formation principle, solutions, and compensation measures. Leak detection methods and algorithms for open water systems are still relatively rare. Summary of the Invention

[0008] This invention addresses the problem that existing methods and algorithms for leak detection in open water systems are still relatively rare.

[0009] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0010] On one hand, the present invention provides a method for monitoring leaks in open water systems based on statistical characteristics. This method is implemented by electronic devices and includes:

[0011] S1. Obtain historical and real-time data of the open water system to be tested.

[0012] S2. Construct an open water system model based on historical data.

[0013] S3. Based on historical data, real-time data, open water system models, and online leak monitoring algorithms, obtain the leakage situation and location of the open water system.

[0014] Optionally, the open water system model constructed in S2 based on historical data includes:

[0015] S21. Obtain historical data for open water systems.

[0016] The historical data includes the outlet pressure of the main water supply pipe of the open water system, the topology of the open water system, and the valve opening and flow rate of each outlet of the open water system.

[0017] S22. Based on historical data, construct a normal operation model and a leakage model for the open water system.

[0018] Optionally, S3 obtains the leakage status and location of the open water system based on real-time data, an open water system model, and an online leak monitoring algorithm, including:

[0019] S31. Based on historical data, calculate the average opening degree of each valve in the open water system, and then calculate the ratio of each valve opening degree to the average opening degree of each valve, and compile ratio data based on the ratio.

[0020] S32. Take the average value of the comparison data to obtain the average value and standard deviation.

[0021] S33. Based on real-time data, calculate the real-time average opening of each valve in the open water system, and then calculate the real-time ratio of each valve opening to the real-time average opening of each valve, and compile real-time ratio data based on the real-time ratios.

[0022] S34. Compare the average data with the real-time ratio data, and determine whether the real-time data is normal or whether there is a leak in the open water system based on the comparison results.

[0023] S35. If the real-time data is determined to be normal data, then the real-time data is added to the historical data, and the average value and standard deviation of the historical data after the real-time data is added are updated.

[0024] If a leak is determined in an open water system, the leak details and location can be determined based on the comparison results.

[0025] Optionally, the ratio data X′ in S31 i As shown in equation (1):

[0026]

[0027] Where m is the number of valves, L′ i,1 ,L′ i,2 ,...,L′ i,m This refers to the opening degree of the branch valve.

[0028] Optionally, the average data in S32 As shown in equation (2):

[0029]

[0030] Where m is the number of valves, L′ i,1 ,L′ i,2 ,...,L′ i,m For the opening degree of the branch valve, N represents the average opening degree of the branch valves, and N is the number of historical data points.

[0031] Alternatively, the standard deviation σ in S32 is shown in equations (3) and (4) below:

[0032] σ=(σ1,σ2,...,σ m (3)

[0033]

[0034] Where m is the number of valves, N is the number of historical data points, and L′ i,j For the opening degree of the branch valve, This represents the average opening degree of the branch valves.

[0035] Optionally, in S34, the average data is compared with the instantaneous ratio data, as shown in equation (5) below:

[0036]

[0037] Among them, L′ t,j To determine the opening degree of the branch valve in real time, ε represents the average opening degree of the branch valves. j This is an adjustable parameter.

[0038] Optionally, the average data in S35 is updated as shown in equation (6):

[0039]

[0040] in, L′ represents the average opening degree of the branch valves, N is the number of historical data points, and L′ is the average opening degree of the branch valves. t,1 This refers to the immediate opening degree of the branch valve.

[0041] Optionally, the standard deviation in S35 is updated as shown in equation (7):

[0042]

[0043] Where N is the number of historical data points, σ j L′ is the standard deviation. t,j To determine the opening degree of the branch valve in real time, This represents the average opening degree of the branch valves.

[0044] On the other hand, the present invention provides a statistically characteristic-based open water system leakage monitoring device, which is used to implement a statistically characteristic-based open water system leakage monitoring method. The device includes:

[0045] The acquisition module is used to acquire historical and real-time data of the open water system to be tested.

[0046] Build modules are used to construct open water system models based on historical data.

[0047] The output module is used to obtain the leakage status and location of the open water system based on historical data, real-time data, open water system model, and online leakage monitoring algorithm.

[0048] Optionally, building modules are further used for:

[0049] S21. Obtain historical data for open water systems.

[0050] The historical data includes the outlet pressure of the main water supply pipe of the open water system, the topology of the open water system, and the valve opening and flow rate of each outlet of the open water system.

[0051] S22. Based on historical data, construct a normal operation model and a leakage model for the open water system.

[0052] Optionally, the output module is further used for:

[0053] S31. Based on historical data, calculate the average opening degree of each valve in the open water system, and then calculate the ratio of each valve opening degree to the average opening degree of each valve, and compile ratio data based on the ratio.

[0054] S32. Take the average value of the comparison data to obtain the average value and standard deviation.

[0055] S33. Based on real-time data, calculate the real-time average opening of each valve in the open water system, and then calculate the real-time ratio of each valve opening to the real-time average opening of each valve, and compile real-time ratio data based on the real-time ratios.

[0056] S34. Compare the average data with the real-time ratio data, and determine whether the real-time data is normal or whether there is a leak in the open water system based on the comparison results.

[0057] S35. If the real-time data is determined to be normal data, then the real-time data is added to the historical data, and the average value and standard deviation of the historical data after the real-time data is added are updated.

[0058] If a leak is determined in an open water system, the leak details and location can be determined based on the comparison results.

[0059] Optionally, the ratio data X′ i As shown in equation (1):

[0060]

[0061] Where m is the number of valves, L′ i,1 ,L′ i,2 ,...,L′ m,m This refers to the opening degree of the branch valve.

[0062] Optionally, average data As shown in equation (2):

[0063]

[0064] Where m is the number of valves, L′ i,1 ,L′ i,2 ,...,L′ i,m For the opening degree of the branch valve, N represents the average opening degree of the branch valves, and N is the number of historical data points.

[0065] Alternatively, the standard deviation σ is shown in equations (3) and (4) below:

[0066] σ=(σ1,σ2,...,σ m (3)

[0067]

[0068] Where m is the number of valves, N is the number of historical data points, and L′ i,j For the opening degree of the branch valve, This represents the average opening degree of the branch valves.

[0069] Optionally, the average data can be compared with the instantaneous ratio data, as shown in equation (5) below:

[0070]

[0071] Among them, L′ t,j To determine the opening degree of the branch valve in real time, ε represents the average opening degree of the branch valves. j This is an adjustable parameter.

[0072] Optionally, the average data is updated as shown in equation (6):

[0073]

[0074] in, L′ represents the average opening degree of the branch valves, N is the number of historical data points, and L′ is the average opening degree of the branch valves. t,1 This refers to the immediate opening degree of the branch valve.

[0075] Optionally, the standard deviation is updated as shown in equation (7):

[0076]

[0077] Where N is the number of historical data points, σ j L′ is the standard deviation. t,j To determine the opening degree of the branch valve in real time, This represents the average opening degree of the branch valves.

[0078] On the one hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, which is loaded and executed by the processor to implement the above-described open water system leakage monitoring method based on statistical characteristics.

[0079] On the one hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, the at least one instruction being loaded and executed by a processor to implement the above-described open water system leakage monitoring method based on statistical characteristics.

[0080] The above technical solution has at least the following advantages compared with the existing technology:

[0081] The above-mentioned scheme proposes a leakage monitoring algorithm for open water systems. This algorithm targets open water systems where flow is controlled by valves, where the flow rate at the end of the water path is controlled by adjusting the valve opening. Based on commonly used data in such water systems, this invention establishes an open water system model. Under this model, through analysis of water system statistical data and combined with real-time data, an online leakage monitoring algorithm is designed for judging leakage conditions and determining leakage locations, achieving effective data utilization. Simulations verify the effectiveness of the method. Attached Figure Description

[0082] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0083] Figure 1 This is a schematic diagram of the open water system leakage monitoring method based on statistical characteristics provided in an embodiment of the present invention;

[0084] Figure 2 This is a normal operating model diagram of an open water system provided in an embodiment of the present invention;

[0085] Figure 3 This is a leakage model diagram of an open water system provided in an embodiment of the present invention;

[0086] Figure 4 This is a graph showing the relationship between accuracy and experimental sample size provided in an embodiment of the present invention;

[0087] Figure 5 This is a block diagram of an open water system leakage monitoring device based on statistical characteristics provided in an embodiment of the present invention;

[0088] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0089] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0090] like Figure 1 As shown, this embodiment of the invention provides a method for monitoring leaks in open water systems based on statistical characteristics, which can be implemented by electronic equipment. Figure 1 The flowchart shown is for a statistically based leak monitoring method for open water systems. The processing flow of this method may include the following steps:

[0091] S1. Obtain historical and real-time data of the open water system to be tested.

[0092] The known information for the open water system includes the following: the outlet pressure of the main water supply pipe, the topology of the water system, the real-time valve opening and flow rate of each outlet; the system adjusts the valve opening online in real time to maintain the set flow rate during operation.

[0093] S2. Construct an open water system model based on historical data.

[0094] Optionally, step S2 above may include the following steps S21-S22:

[0095] S21. Obtain historical data for open water systems.

[0096] The historical data includes the outlet pressure of the main water supply pipe of the open water system, the topology of the open water system, and the valve opening and flow rate of each outlet of the open water system.

[0097] S22. Based on historical data, construct a normal operation model and a leakage model for the open water system.

[0098] In one feasible implementation, based on existing information, an open water system model can be established, such as... Figure 2 As shown in the figure. P and Q are the pressure and flow measurement points, respectively, and the ends of each load (spray device) can be considered to be directly connected to the open water pool.

[0099] Furthermore, when a water spray device leaks, the model changes as follows: Figure 3 As shown.

[0100] If a leak occurs at load 1 on the right, it can be seen in the system structure diagram as a line directly connected to the open water pool on the load. Therefore, the flow from this branch has a new consumption path in addition to the consumption by the water spray device and the discharge into the water pool.

[0101] In this model, since the flow rate Q of each water spray device remains basically constant, and the flow rate Q is only adjusted by its valve opening L, the relationship between the main pipe pressure P and the water spray flow rate Q of each branch pipe can be analyzed by analyzing its operating data over a period of time, as well as the variation law of the valve opening L of each branch under similar operating conditions, thereby realizing the monitoring of leakage conditions.

[0102] S3. Based on historical data, real-time data, open water system models, and online leak monitoring algorithms, obtain the leakage situation and location of the open water system.

[0103] Optionally, step S3 above may include the following steps S31-S35:

[0104] S31. Based on historical data, calculate the average opening degree of each valve in the open water system, and then calculate the ratio of each valve opening degree to the average opening degree of each valve, and compile ratio data based on the ratio.

[0105] In one feasible implementation, it is assumed that there exists an open water system conforming to the aforementioned model, where the pressure P of the main pipe at each moment is known; the main pipe has a distributor with m branch pipes, and the flow rate of water ejected from the valve of each branch pipe at each moment is known and assumed to be basically constant, and the opening degree of the branch pipe valves is L1, L2, ..., L m It is known that the valve flow rate setpoint is constant, and leakage will not affect the flow characteristic curves of the leakage branch and other branches.

[0106] Take N normal data points after maintenance as the normal dataset, and each data point X i (P i ,L i,1 ,L i,2 ,L i,3 ,...,L i,m The pressure P in the main pipe i and the opening degree L of m branch valves i,1 ,L i,2 ,L i,3 ,...,L i,m Composition. The actual data during device operation is denoted as X. t .

[0107] Assuming that under constant pressure, the flow rate Q of a certain valve is directly proportional to the valve opening L, i.e., L = kQ, where k is a coefficient negatively correlated with pressure. Based on this assumption, while keeping the set flow rate Q constant, the flow rates Q1:Q2:...:Q of each valve are... m It remains unchanged, so the ratio of the opening degrees of each valve is as shown in equation (1):

[0108] L1:L2:...:L m =kQ1:kQ2:...:kQ m =Q1:Q2:...:Q m (1)

[0109] The ratio remains constant; consequently, the ratio of a certain valve opening at a given moment to the average valve opening at that moment also remains constant, i.e. The value of is a constant, which is independent of the real-time pressure P.

[0110] Furthermore, regarding data X i Calculate the ratio of each valve opening degree to the average valve opening degree. And generate new data consisting of this ratio:

[0111]

[0112] Where m is the number of valves, L′ i,1 ,L′ i,2 ,...,L′ i,m This refers to the opening degree of the branch valve.

[0113] S32. Take the average value of the comparison data to obtain the average value and standard deviation.

[0114] Optionally, the average data in S32 As shown in equation (3):

[0115]

[0116] Where m is the number of valves, L′ i,1 ,L′ i,2 ,...,L′ i,m For the opening degree of the branch valve, N represents the average opening degree of the branch valves, and N is the number of historical data points.

[0117] Alternatively, the standard deviation σ in S32 is shown in equations (4) and (5) below:

[0118] σ=(σ1,σ2,...,σ m (4)

[0119]

[0120] Where m is the number of valves, N is the number of historical data points, and L′ i,j For the opening degree of the branch valve, This represents the average opening degree of the branch valves.

[0121] S33. Based on real-time data, calculate the real-time average opening of each valve in the open water system, and then calculate the real-time ratio of each valve opening to the real-time average opening of each valve, and compile real-time ratio data based on the real-time ratios.

[0122] S34. Compare the average data with the real-time ratio data, and determine whether the real-time data is normal or whether there is a leak in the open water system based on the comparison results.

[0123] In one feasible implementation, X is used. t Represents real-time data, ε j This is an adjustable parameter, taken as ε here. j =3σ j j = 1, 2, ..., m. Compared with the real-time data X processed by method (1) t Compare them.

[0124]

[0125] Among them, L′ t,j To determine the opening degree of the branch valve in real time, ε represents the average opening degree of the branch valves. j This is an adjustable parameter.

[0126] S35. If the real-time data is determined to be normal data, then the real-time data is added to the historical data, and the average value and standard deviation of the historical data after the real-time data is added are updated.

[0127] In one feasible implementation, for each real-time data X t If, after the aforementioned comparison, it is concluded that the data is normal, then this data X... t Adding a normal dataset, at this point the normal dataset... The standard deviation σ is updated as follows:

[0128] Optionally, the average data in S35 is updated as shown in equation (7):

[0129]

[0130] in, L′ represents the average opening degree of the branch valves, N is the number of historical data points, and L′ is the average opening degree of the branch valves. t,1 This refers to the immediate opening degree of the branch valve.

[0131] Optionally, the standard deviation in S35 is updated as shown in equation (8):

[0132]

[0133] Where N is the number of historical data points, σ j L′ is the standard deviation. t,j To determine the opening degree of the branch valve in real time, This represents the average opening degree of the branch valves.

[0134] If a leak is determined in an open water system, the leak details and location can be determined based on the comparison results.

[0135] In one feasible implementation, real-time data X t The j-th term and The absolute value of the difference is greater than ε j If so, then it is considered that there is a leak in branch j.

[0136] Considering that the occurrence of leakage does not affect the flow characteristic curves of the leaking branch and other branches, the above algorithm can give the location of each leakage point when multiple leakage points occur, so the algorithm has universality.

[0137] Algorithm 1: Leakage monitoring algorithm for open water systems.

[0138] Input: N sets of normal data X i (P i ,L i,1 ,L i,2 ,L i,3 ,...,L i,m Real-time data X t (P t ,L t,1 ,L t,2 ,L t,3 ,...,L t,m );

[0139] Step 1:

[0140]

[0141]

[0142]

[0143]

[0144] Step 2:

[0145] For each input X t :

[0146]

[0147] If it exists

[0148] Output t,j;

[0149] If it does not exist

[0150]

[0151] Add this input to the normal data.

[0152] Furthermore, simulation verification:

[0153] Based on the aforementioned model, construct an open water system with potential leakage, setting N = 5000 and m = 9; generate normal datasets and experimental sets according to the following rules.

[0154] The pressure P in the main pipe at each moment is known, and P takes a random value within [80, 120].

[0155] The main pipe has a distributor with m=9 branch pipes. The flow rate of water sprayed from the valve of each branch pipe at any given time is known. While the sum of the flow rates of each branch pipe remains basically constant, there are certain fluctuations in the flow rate of each branch pipe, and the fluctuation range does not exceed 1% of the flow rate of the main pipe.

[0156] The probability of leakage at any given time is p = 0.1, and the leakage amount shall not exceed 8% of the total pipe flow rate and shall not be less than 3% of the total pipe flow rate.

[0157] It is believed that the occurrence of leakage will not affect the flow characteristic curves of the leaking branch and other branches.

[0158] A dataset of 10,000 data points was generated as the experimental set, and the proposed algorithm was used for the experiment. Simulation results show that the final accuracy of the algorithm is 92.3%. Figure 4 The trend of the accuracy of leakage detection as the number of samples in the experimental set increases.

[0159] This invention proposes a leakage monitoring algorithm for open water systems. The algorithm targets open water systems where flow is controlled by valves, where the flow rate at the end of the water path is controlled by adjusting the valve opening. Based on commonly used data in such water systems, this invention establishes an open water system model. Under this model, through analysis of water system statistical data and combined with real-time data, an online leakage monitoring algorithm is designed for judging leakage conditions and determining leakage locations, achieving effective data utilization. Simulations verify the effectiveness of the method.

[0160] like Figure 5As shown, this embodiment of the invention provides a statistical feature-based open water system leakage monitoring device 500. This device 500 is used to implement a statistical feature-based open water system leakage monitoring method. The device 500 includes:

[0161] The acquisition module 510 is used to acquire historical and real-time data of the open water system to be tested.

[0162] Module 520 is used to build open water system models based on historical data.

[0163] The output module 530 is used to obtain the leakage status and location of the open water system based on historical data, real-time data, open water system model and online leakage monitoring algorithm.

[0164] Optionally, module 520 is further used for:

[0165] S21. Obtain historical data for open water systems.

[0166] The historical data includes the outlet pressure of the main water supply pipe of the open water system, the topology of the open water system, and the valve opening and flow rate of each outlet of the open water system.

[0167] S22. Based on historical data, construct a normal operation model and a leakage model for the open water system.

[0168] Optionally, the output module 530 is further used for:

[0169] S31. Based on historical data, calculate the average opening degree of each valve in the open water system, and then calculate the ratio of each valve opening degree to the average opening degree of each valve, and compile ratio data based on the ratio.

[0170] S32. Take the average value of the comparison data to obtain the average value and standard deviation.

[0171] S33. Based on real-time data, calculate the real-time average opening of each valve in the open water system, and then calculate the real-time ratio of each valve opening to the real-time average opening of each valve, and compile real-time ratio data based on the real-time ratios.

[0172] S34. Compare the average data with the real-time ratio data, and determine whether the real-time data is normal or whether there is a leak in the open water system based on the comparison results.

[0173] S35. If the real-time data is determined to be normal data, then the real-time data is added to the historical data, and the average value and standard deviation of the historical data after the real-time data is added are updated.

[0174] If a leak is determined in an open water system, the leak details and location can be determined based on the comparison results.

[0175] Optionally, the ratio data X′ i As shown in equation (1):

[0176]

[0177] Where m is the number of valves, L′ i,1 ,L′ i,2 ,...,L′ i,m This refers to the opening degree of the branch valve.

[0178] Optionally, average data As shown in equation (2):

[0179]

[0180] Where m is the number of valves, L′ i,1 ,L′ i,2 ,...,L′ i,m For the opening degree of the branch valve, N represents the average opening degree of the branch valves, and N is the number of historical data points.

[0181] Alternatively, the standard deviation σ is shown in equations (3) and (4) below:

[0182] σ=(σ1,σ2,...,σ m (3)

[0183]

[0184] Where m is the number of valves, N is the number of historical data points, and L′ i,j For the opening degree of the branch valve, This represents the average opening degree of the branch valves.

[0185] Optionally, the average data can be compared with the instantaneous ratio data, as shown in equation (5) below:

[0186]

[0187] Among them, L′ t,j To determine the opening degree of the branch valve in real time, ε represents the average opening degree of the branch valves. j This is an adjustable parameter.

[0188] Optionally, the average data is updated as shown in equation (6):

[0189]

[0190] in, L′ represents the average opening degree of the branch valves, N is the number of historical data points, and L′ is the average opening degree of the branch valves. t,1 This refers to the immediate opening degree of the branch valve.

[0191] Optionally, the standard deviation is updated as shown in equation (7):

[0192]

[0193] Where N is the number of historical data points, σ j L′ is the standard deviation. t,j To determine the opening degree of the branch valve in real time, This represents the average opening degree of the branch valves.

[0194] This invention proposes a leakage monitoring algorithm for open water systems. The algorithm targets open water systems where flow is controlled by valves, where the flow rate at the end of the water path is controlled by adjusting the valve opening. Based on commonly used data in such water systems, this invention establishes an open water system model. Under this model, through analysis of water system statistical data and combined with real-time data, an online leakage monitoring algorithm is designed for judging leakage conditions and determining leakage locations, achieving effective data utilization. Simulations verify the effectiveness of the method.

[0195] Figure 6 This is a schematic diagram of the structure of an electronic device 600 provided in an embodiment of the present invention. The electronic device 600 can vary considerably due to differences in configuration or performance. It may include one or more central processing units (CPUs) 601 and one or more memories 602. The memory 602 stores at least one instruction, which is loaded and executed by the processor 601 to implement the following open water system leakage monitoring method based on statistical characteristics:

[0196] S1. Obtain historical and real-time data of the open water system to be tested.

[0197] S2. Construct an open water system model based on historical data.

[0198] S3. Based on historical data, real-time data, open water system models, and online leak monitoring algorithms, obtain the leakage situation and location of the open water system.

[0199] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including instructions that can be executed by a processor in a terminal to perform the above-described statistical characteristic-based open water system leakage monitoring method. For example, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0200] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0201] 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 within the protection scope of the present invention.

Claims

1. A method for monitoring leaks in open water systems based on statistical characteristics, characterized in that, The method includes: S1. Acquire historical and real-time data of the open water system to be tested; wherein, the open water system is controlled by valves to control the flow rate, including the main water supply pipe and multiple outlets; S2. Construct an open water system model based on the historical data; S3. Based on the historical data, real-time data, open water system model, and online leakage monitoring algorithm, obtain the leakage status and leakage location of the open water system; The step S2, which involves constructing an open water system model based on the historical data, includes: S21. Obtain historical data for open water systems; The historical data includes the outlet pressure of the main water supply pipe of the open water system, the topology of the open water system, and the valve opening and flow rate of each outlet of the open water system. S22. Based on the historical data, construct a normal operation model and a leakage model for the open water system; The step S3, based on the historical data, real-time data, open water system model, and online leak monitoring algorithm, obtains the leakage status and location of the open water system, including: S31. Based on the historical data, calculate the average opening value of each valve in the open water system, and then calculate the ratio of each valve opening value to the average opening value of each valve, and form ratio data based on the ratio. S32. Take the average value of the ratio data to obtain the average value and standard deviation; S33. Based on the real-time data, calculate the real-time average value of the opening degree of each valve in the open water system, and then calculate the real-time ratio of each valve opening degree to the real-time average value of each valve opening, and form real-time ratio data based on the real-time ratio. S34. Compare the average data with the instantaneous ratio data, and determine whether the instantaneous data is normal or whether there is a leak in the open water system based on the comparison result. S35. If the real-time data is determined to be normal data, the real-time data is added to the historical data, and the average value and standard deviation of the historical data after the real-time data is added are updated. If it is determined that there is a leak in the open water system, the leak status and location of the open water system can be obtained based on the comparison results.

2. The method according to claim 1, characterized in that, The ratio data in S31 As shown in equation (1): （1） in, For the number of valves, This refers to the opening degree of the branch valve.

3. The method according to claim 1, characterized in that, The average data in S32 As shown in equation (2): （2） in, For the number of valves, For the opening degree of the branch valve, This represents the average opening degree of the branch valves. This represents the number of historical data entries.

4. The method according to claim 1, characterized in that, Standard deviation in S32 As shown in equations (3) and (4): （3） （4） in, For the number of valves, The number of historical data points. For the opening degree of the branch valve, This represents the average opening degree of the branch valves.

5. The method according to claim 1, characterized in that, In step S34, the average value data is compared with the instantaneous ratio data, as shown in equation (5) below: （5） in, To determine the opening degree of the branch valve in real time, This represents the average opening degree of the branch valves. It is an adjustable parameter. , This refers to the number of valves.

6. The method according to claim 1, characterized in that, The average data update in S35 is shown in equation (6) below: （6） in, This represents the average opening degree of the branch valves. The number of historical data points. This refers to the immediate opening degree of the branch valve.

7. The method according to claim 1, characterized in that, The standard deviation update in S35 is shown in equation (7) below: （7） in, The number of historical data points. Standard deviation, To determine the opening degree of the branch valve in real time, This represents the average opening degree of the branch valves.

8. A statistical feature-based open water system leakage monitoring device, wherein the statistical feature-based open water system leakage monitoring device is used to implement the statistical feature-based open water system leakage monitoring method as described in any one of claims 1-7, characterized in that, The device includes: The acquisition module is used to acquire historical and real-time data of the open water system to be tested; A construction module is used to build an open water system model based on the historical data; The output module is used to obtain the leakage status and leakage location of the open water system based on the historical data, real-time data, open water system model, and online leakage monitoring algorithm.

Images