Real-time Monitoring Method, System and Electronic Device for Operating Status of Water Supply Pipe Network

Abstract

The present invention provides a method for real-time monitoring of the operation status of a water supply network, including establishing a digital network topology structure model based on the obtained water supply network topology structure and pipeline parameters; collecting the measured parameters of the operation status of the water supply network within the control area of the water supply network, and preliminarily calculating relevant parameters; performing real-time network hydraulic simulation analysis and calculation based on the network topology structure model and the measured parameters to obtain the simulated parameters of the operation status of the water supply network; combining a water supply network data status anomaly detection algorithm to determine whether there is an anomaly in the status of the water supply network; when there is an anomaly, finding the cause of the anomaly based on a multi-modal factor comprehensive judgment formula. The present invention further discloses a corresponding system and electronic device. The present invention can accurately judge the water usage behavior of water usage units in the water supply network, the water leakage situation of the water supply network, and the status of data collection devices, and can also locate abnormal water usage nodes, network leakage points, and abnormal metering positions.

Description

Technical Field

[0001] The present invention belongs to the field of water supply network status monitoring and water conservation management control, and particularly relates to a real-time monitoring method, system and electronic device for the operation status of a water supply network. Background Art

[0002] The operation status of a water supply network (such as pressure, flow rate, flow velocity), the operation status of production equipment in water-using units, the water use and drainage conditions, and the measurement accuracy of water supply network metering and monitoring equipment (such as water meters) are the basis for judging whether the water supply network is abnormal and whether there are abnormal water use behaviors in the network system, and are also important bases for water conservancy supervision departments to conduct water balance tests and water conservation supervision of water-using units.

[0003] Abnormalities in the water supply network and abnormal water use behaviors mainly include phenomena such as network leakage, inaccurate water use metering, and abnormal changes in water use behaviors. For network leakage, at present, a single water meter metering method is mainly adopted in the water supply network for total amount accounting. If there is inaccurate water meter metering or abnormal water use behaviors, it is impossible to determine whether the problem is caused by damage to the water supply network itself or abnormal user behaviors only through the total amount accounting of water meter metering data, and it is even more impossible to determine the specific leakage location and leakage degree of the water supply network. Similarly, to determine abnormal water use behaviors or changes in the operation status of production equipment in water-using units, the water use or drainage conditions, it is often necessary to analyze the water consumption of the water-using unit in combination with the operation conditions. Once there is inaccurate water meter metering or leakage in the pipeline at the front end of the water-using unit, it is impossible to monitor the change in the water consumption of the water-using unit.

[0004] The existing technology mainly relies on single water meter data, lacks the ability of multi-source data fusion analysis, and is difficult to effectively distinguish phenomena such as network leakage, inaccurate water use metering, and abnormal water use behaviors, resulting in inaccurate leakage location, misjudgment of water use behaviors, and deviation in water balance tests. In addition, the existing technology has insufficient intelligence, cannot monitor the operation status of the water supply network in real time and give early warnings of abnormal behaviors, and is difficult to meet the requirements of refined management and water conservation supervision. Summary of the Invention

[0005] The present invention provides a real-time monitoring method, system and electronic device for the operation status of a water supply network based on multi-modal anomaly analysis. By detecting the measured parameters of the operation status of the water supply network such as pressure, flow rate, and flow velocity at the monitoring nodes of the water supply network, combining the energy equation and the continuity equation to obtain the simulated parameters of the operation status of the water supply network such as flow rate, pressure, and flow velocity, and when it is determined that there is data anomaly based on the data status anomaly detection algorithm for the water supply network, further combining the multi-modal evaluation method for the operation status of the water supply network to find the cause of the anomaly, so as to accurately judge the water use behaviors of water-using units in the water supply network, the leakage situation of the water supply network, and the situation of inaccurate water use metering, and also be able to locate the position where the abnormal data is located.

[0006] The first aspect of the present invention discloses a real-time monitoring method for the operation status of a water supply network, including the following steps:

[0007] Step 1: Based on the obtained topological structure and pipe parameters of the water supply network, establish a digital network topological structure model; the pipe parameters include the pipe length , pipe diameter , and pipe design flow modulus . Among them, is the number of the monitoring node of the water supply network, and are adjacent monitoring nodes;

[0008] Step 2: Collect the measured parameters of the operation status of the water supply network within the control area of the water supply network, and preliminarily calculate the measured flow production and consumption difference , measured pipe pressure drop , and calculated flow at the node ; the measured parameters include measured node pressure , measured node flow , and measured node water flow velocity ; is the collection time;

[0009] Step 3: Based on the network topological structure model, conduct real-time network hydraulic simulation analysis and calculation on the measured node pressure and measured node flow to obtain the simulation parameters of the operation status of the water supply network; the simulation parameters include simulated node pressure , simulated pipe pressure drop , simulated pipe flow , simulated pipe water flow velocity , simulated node water flow velocity , and pipe simulated flow modulus ;

[0010] Step 4: Based on the measured parameters and simulation parameters, combine the abnormal detection algorithm for the data status of the water supply network to determine whether there is an abnormality in the status of the water supply network, and enter Step 5 when it is determined that there is an abnormality;

[0011] Step 5: Based on a predefined multi-modal factor comprehensive judgment formula, find the cause of the abnormality and locate the abnormal position; the multi-modal factor comprehensive judgment formula has multiple modal evaluation factors related to the measured parameters and simulation parameters.

[0012] Among them, the pipe design flow modulus can be obtained from the "Water Supply and Drainage Design Manual" edited by the Southwest China Municipal Engineering Design and Research Institute and published by China Architecture and Building Press.

[0013] Furthermore, the following is also included in Step 2:

[0014] Store the measured data in the form of a one-dimensional vector and denote it as ;

[0015] The measured flow production and consumption difference is calculated by the following formula:

[0016]

[0017] In the formula, is the measured nodal flow at the access point of the municipal water supply network at time is the sum of the measured nodal flows at the branch pipes of the monitoring nodes at time is the total number of monitoring nodes of the water supply network;

[0018] The measured pipeline pressure drop is calculated by the following formula:

[0019]

[0020] In the formula, and respectively represent the measured nodal pressures at the adjacent monitoring nodes at time

[0021] The calculated flow rate at the node is calculated by the following formula:

[0022]

[0023] In the formula, represents the pipe diameter at the branch pipe of the monitoring node and is the measured nodal water flow velocity.

[0024] Furthermore, in step 3, real-time pipe network hydraulic simulation analysis and calculation are performed based on the energy equation and the pipe network continuity equation.

[0025] Furthermore, in step 3, the pipe simulation flow modulus is calculated based on the simulated pipe flow and the simulated pipe pressure drop The calculation formula is as follows:

[0026] .

[0027] Furthermore, step 4 specifically includes:

[0028] Calculate the average value of each parameter based on the node pressure, node flow rate, node water flow velocity in the measured parameters and historical simulation parameters, as well as the pipeline design flow modulus and historical pipeline simulation flow modulus. , , , ;

[0029] Calculate the deviation degree of each parameter and form a deviation degree vector corresponding to the monitoring node parameters. ， And a deviation degree vector corresponding to the pipeline parameters between adjacent monitoring nodes. , where , is the number of sampling times;

[0030] Form a deviation degree vector group from the deviation degrees of each parameter at different times at each monitoring node. X P Form a deviation degree vector group from the deviation degrees of each parameter at different times for the pipelines between each adjacent monitoring node. X C ;

[0031] The water supply network data status anomaly detection algorithm includes:

[0032] Calculate the distance between the deviation degree vector corresponding to the monitoring node or the pipeline between adjacent monitoring nodes at time and the deviation degree vectors at other times , and sort the calculation results in ascending order to obtain the distance sequence of ; ;

[0033] Calculate the z-nearest distance, z-distance neighborhood, reachable distance and reachable density of each component in the deviation degree vector in turn;

[0034] Combined with the reachable density and z-distance neighborhood of each vector in the deviation degree vector group or , calculate the anomaly factor of each deviation degree vector;

[0035] When the anomaly factor is greater than 1, it is determined that there is an anomaly in the water supply network state.

[0036] Furthermore, in step 5:

[0037] Obtain or calculate the average value of each parameter based on the node pressure, node flow rate, node water flow velocity in the measured parameters and historical simulation parameters, as well as the pipeline design flow modulus and historical pipeline simulation flow modulus. , , , ;

[0038] The modal evaluation factors include:

[0039] , where , , , in the formula, is the flow rate at the access point of the municipal water supply network at time is t the sum of the flow rates of all monitoring nodes at time is the number of sampling times, is the total number of monitoring nodes of the water supply network;

[0040] , where , , ;

[0041] , where , , , in the formula, is the water consumption of the water-using unit at the branch pipe of the monitoring node at time is the average value of the historical water consumption of the water-using unit at the branch pipe of the monitoring node, is the drainage coefficient, is the drainage volume of the water-using unit at the branch pipe of the monitoring node at time;

[0042] The multi-modal factor comprehensive judgment formula is:

[0043]

[0044] In the multi-modal factor comprehensive judgment formula, K 1 、K 2 、K 3 are modal coefficients determined by the characteristics of the water supply network system and the water-using characteristics of the water-using units;

[0045] The M, M 1 、M 2 、M 3 alarm thresholds are all dynamically adjustable; when M is greater than the alarm threshold, it indicates that there is an abnormality in the water supply network system, and it is necessary to further analyze M 1 、M 2 、M 3 to find the cause of the abnormality and locate the abnormal position.

[0046] Further, when M, M 1 、M 2 、M and 3 all have an alarm threshold of threshold A1;

[0047] Among them, M when 1 is greater than threshold A1, it includes three situations:

[0048] When M 11 is greater than threshold A1: ① M 11 is M the maximum value in 1 and M 12 、M 13 are all greater than threshold A2, it indicates that M 12 、M 23 there is a leak in the pipeline between the adjacent monitoring nodes corresponding to the maximum value; ② M 12 、M 13 are all less than threshold A2, it indicates that there is a distortion in water consumption measurement at the corresponding monitoring node, and the location with the problem is determined through a data anomaly detection algorithm;

[0049] When M 11 、M 13 are all greater than threshold A1, it indicates that M 13 there is a leak in the pipeline between the adjacent monitoring nodes corresponding to the maximum value;

[0050] When M 11 is less than threshold A3, it indicates that there is a distortion in water consumption measurement, and the location with the problem is determined through a data anomaly detection algorithm;

[0051] Among them, M when 2 is greater than threshold A1, it includes two situations:

[0052] When M 21 、M 22 are all greater than threshold A1, it indicates that M 22 there is a leak in the pipeline between the adjacent monitoring nodes corresponding to the maximum value;

[0053] When M 23 is greater than threshold A4, it indicates that M 23 there is a distortion in water consumption measurement at the monitoring node corresponding to the maximum value;

[0054] Among them, M When 3 is greater than the threshold A1, it includes two cases:

[0055] When M 31 is greater than the threshold A1, it indicates that M 31 there is abnormal water use in the corresponding water use unit greater than the threshold A1;

[0056] When M 32 is greater than the threshold A5 or less than the threshold A1, and M 33 is greater than the threshold A6, it indicates that M 33 the water metering at the monitoring node corresponding to the maximum value is distorted.

[0057] Furthermore, the threshold A1 is 1.1, A2 is 1.05, A3 is 1, A4 is 0.05, A5 is 0.9, and A6 is 0.1.

[0058] The second aspect of the present invention discloses a real-time monitoring system for the operation status of a water supply network, including a data acquisition device, a data analysis module, and a data center; the data acquisition device is arranged at the monitoring nodes of the water supply network and is used to collect the measured parameters of the operation status of the water supply network within the control area of the water supply network; the data analysis module is communicatively connected to the data acquisition device and is used to implement the real-time monitoring method for the operation status of the water supply network described in the first aspect of the present invention and upload the monitoring results to the data center.

[0059] The third aspect of the present invention discloses an electronic device, which is characterized in that it includes a processor and a memory; the processor is used to call the computer program stored in the memory and execute the real-time monitoring method for the operation status of the water supply network described in the first aspect of the present invention.

[0060] The present invention has the following beneficial effects:

[0061] (1) The real-time monitoring of the operation status of the water supply network provided by the present invention, by collecting real-time parameters such as the pressure, flow rate, and flow velocity of the water supply network, dynamically analyzes abnormal behaviors and abnormal water use behaviors such as network leakage, water meter distortion, and changes in water use conditions, and realizes real-time feedback and dynamic supervision. This method can improve the operation efficiency of the water supply network, evaluate the effectiveness of the previous traditional static water balance test under the current operation status of the network, provide big data support for the future operation and maintenance of the water supply network, and solve the problems of high hysteresis and insufficient accuracy of the traditional method.

[0062] (2) The water supply network data status anomaly detection algorithm proposed by the present invention can comprehensively reflect the temporal and spatial changes of the water supply status by integrating spatio-temporal characteristic parameters such as pressure fluctuations, flow rate changes, and unit water usage patterns in the water supply network system. This algorithm uses multi-modal data fusion technology to analyze influencing factors, avoids the interference of parameters with each other when multiple abnormal states occur simultaneously in problem judgment, accurately determines the modal abnormal state characteristics of the water supply network system, improves the timeliness of anomaly detection and the accuracy of positioning, and solves the problem of high misjudgment rate of traditional methods in complex scenarios.

[0063] (3) The present invention can dynamically set the alarm thresholds of each mode according to requirements and optimize and adjust the water supply network system in combination with the mode evaluation factor. This method can replace the traditional water balance test, realize the intelligent evaluation and management of the network operation status, provide a scientific basis for efficient water conservation, user supervision, and the improvement of the water price measurement system, and solves the problems of poor adaptability and insufficient flexibility of traditional methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a schematic diagram of the real-time monitoring system for the operation status of the water supply network;

[0065] Figure 1 Among them: C1 represents the first-level water meter; C2 represents the second-level water meter; C3 represents the water supply network; C4 represents the monitoring node of the water supply network; C5 represents the data transmission module; C6 represents the data analysis module; C7 represents the data center; C8 represents the water usage unit; C9 represents the branch pipe;

[0066] Figure 2 It is a topological structure diagram of the water supply network;

[0067] Figure 3 It is a schematic diagram of the analysis process of the operation status of the water supply network;

[0068] Figure 4 It is a schematic diagram of the determination process for finding the cause of anomalies by the multi-modal evaluation method. DETAILED DESCRIPTION OF THE INVENTION

[0069] To further understand the present invention, the technical solutions of the present invention will be further explained below in conjunction with specific embodiments.

[0070] Embodiment 1 of the present invention discloses a real-time monitoring system for the water-saving operation status of a water supply network based on multi-modal anomaly analysis (referred to as the "water-saving monitoring system" for short), which can be applied to the water supply network system of a certain water user. The water-saving monitoring system mainly includes a data acquisition device, a data analysis module C6, and a data center C7.

[0071] Such as Figure 1As shown in the figure, the water supply pipe network of the water - using unit is a loop - shaped water supply pipe network C3. There are several branch pipes C9 arranged on the loop - shaped pipe network. Each branch pipe includes at least one water - using unit C8, and these water - using units C8 usually refer to equipment or devices with relatively large water consumption. There are also several water supply pipe network monitoring nodes C4 arranged on the loop - shaped water supply pipe network C3. These water supply pipe network monitoring nodes C4 are usually set at the access points of the municipal water supply pipe network and at the points where the branch pipes are connected to the loop - shaped pipe network. It can be understood that the number of these water supply pipe network monitoring nodes C4 can also be greater than the number of branch - pipe access points, especially when the distance between adjacent branch pipes is relatively long. However, for the sake of easy understanding, a one - to - one correspondence relationship is adopted in this embodiment. Among them, the data acquisition device is integrated with a data transmission module C5, which is mainly used to collect in real - time the operation - state data such as flow rate and pressure at the pipe nodes, and transmit the collected operation - state data to the data analysis module C6 through the data transmission module C5. In specific applications, the data acquisition device can be a water meter with a built - in data transmission chip. For example, SCL61H - type, SCL - 61D2 - type water meters, MTW - X - DF4 - type ultrasonic water meters; or a water meter with an externally - attached data transmission chip can also be used. The externally - attached data transmission chip can adopt HZG - 4 data acquisition instrument, HZG - 9 series data acquisition instruments, and these data acquisition instruments have NB - loT and GPRS communication capabilities. Whether it is a water meter with a built - in data transmission chip or a water meter with an externally - attached data transmission chip mentioned above, both have the capabilities of monitoring flow rate, flow velocity, and pressure. According to different locations, water meters can be specifically divided into first - level water meters C1 (also called "master meters") installed at the access points of the municipal water supply pipe network and second - level water meters C2 installed at the water supply pipe network monitoring nodes C4. The first - level water meter records the total water consumption of the water - using unit, and a second - level water meter C2 is installed at each water supply pipe network monitoring node C4, and the second - level water meter records the water consumption of each water - using node.

[0072] The data analysis module C6 is also communicatively connected to the data center C7. The data analysis module C6 completes the abnormal analysis of the monitoring data and the determination of the pipe network operation state based on the received pipe network operation - state data, and the data center C7 stores the data and issues an alarm for abnormal water use in the pipe network.

[0073] As Figure 3 shown, Embodiment 2 of the present invention provides a real - time water - saving monitoring method for a water supply pipe network based on multi - modal abnormal analysis, which can be applied to the water - saving monitoring system described in Embodiment 1, and mainly includes the following steps:

[0074] Step 1, establish a pipe network topology structure.

[0075] Assume that the number of the access point of the municipal water supply pipe network (at the first - level water meter C1) is set to 0, and the numbers of the water supply pipe network branch monitoring nodes (at the second - level water meters C2) are set in sequence as , where is the total number of monitoring nodes in the water supply pipe network. For the convenience of description, in this patent, the municipal water supply pipe network access points and the monitoring nodes of the branch pipes of the water supply pipe network are collectively referred to as monitoring nodes (abbreviated as "nodes"). Therefore, the numbers of the monitoring nodes satisfy .

[0076] According to the pre-obtained and stored topological structure of the water supply pipe network (as Figure 2 shown), and the pipe length of the pipe between adjacent monitoring nodes (abbreviated as "pipe length "), the pipe diameter of the pipe between adjacent monitoring nodes (abbreviated as "pipe diameter "), the pipe design flow modulus of the pipe between adjacent monitoring nodes (abbreviated as "pipe design flow modulus ") and other parameters, a digital pipe network topological structure model is established, where is the number of adjacent monitoring nodes of the water supply pipe network monitoring, .

[0077] Step 2, data collection and preliminary analysis.

[0078] Within the control area of the water supply pipe network, the primary water meter C1 and the secondary water meter C2 respectively collect the measured parameters of the operating state of the water supply pipe network at each monitoring node with a preset sampling period (for example, every 3 minutes), and then transmit them to the data analysis module C6 in a wired or wireless manner through the data transmission chip in the water meter. Among them, the measured parameters of the operating state include the branch pipe water pressure at the branch pipe of the water supply pipe network monitoring node collected at the moment (abbreviated as "measured node pressure "), the water flow rate at the branch pipe of the monitoring node (abbreviated as "measured node flow "), the water flow velocity at the branch pipe of the monitoring node (abbreviated as "measured node water flow velocity ").

[0079] After receiving the data of the operating state of the water supply pipe network, the data analysis module C6 stores it in the form of a one-dimensional vector, that is, the vector . Taking the monitoring node with the number as an example, the stored vector data is recorded as . Based on the obtained vector , the data analysis module C6 makes a preliminary analysis of the pipe network related data as follows:

[0080] S21: Calculate the measured flow production and sales difference in the control area of the water supply pipe network at the moment (abbreviated as "measured flow production and sales difference ”), and the calculation formula is:

[0081] (1)

[0082] In the formula, is the measured nodal flow rate at the access point of the municipal water supply network at time (i.e., at the first-level water meter C1), is the sum of the measured nodal flow rates at the branch pipes of the monitored nodes at time 。

[0083] S22: Calculate the measured pipeline water pressure drop value (abbreviated as "measured pipeline pressure drop ") between the monitored nodes of the water supply pipeline at time

[0084] (2)

[0085] In the formula, and respectively represent the measured nodal pressures at the adjacent monitored nodes at time i.e., the second component in the vector ；

[0086] After the calculation, the one-dimensional vector composed of the measured pipeline pressure drops at different times can be expressed as .

[0087] S23: Calculate the measured pipeline flow rate (abbreviated as "calculated flow rate at the node ") at the branch pipes of the monitored nodes of the water supply pipeline at time ". Specifically, it can be calculated from parameters such as the measured nodal water flow velocity detected by a flow velocity meter and other detection devices. The calculation formula is:

[0088] (3)

[0089] In the formula, represents the pipe diameter at the branch pipe of node .

[0090] The one-dimensional vector (abbreviated as "vector ") composed of the calculated flow rates at the node at different times can be expressed as and ​

[0091] The data analysis related to the pipe network is carried out in the data analysis module C6. After the calculation is completed, the data analysis module C6 sends it to the data center C7, and the output data is stored in the form of a one-dimensional vector, that is, the vector .

[0092] Step 3, real-time pipe network hydraulic simulation analysis.

[0093] Based on the digital pipe network topology structure model constructed in Step 1 and the measured node pressure at the monitoring nodes of the water supply pipe network at the moment obtained in Step 2 , measured node flow and other parameters, real-time pipe network hydraulic simulation analysis is carried out by means of adjustment calculation, which specifically includes the following sub-steps:

[0094] S31: According to the continuity equation of the water supply pipe network (Formula 5), the flow distribution of the pipe network is carried out to obtain the initial flow of the pipes between adjacent monitoring nodes ;

[0095] S32: Based on the initial flow of the pipes between adjacent monitoring nodes obtained and the flow modulus of each pipe in the input vector in Step 2 , the pressure drop of each pipe is calculated respectively by (Formula 6) , and it is judged whether it satisfies the energy equation (Formula 4). If it satisfies, enter S35; if not, enter the next step;

[0096] S33: Calculate the corrected flow through the pipe correction flow equation (Formula 8);

[0097] S34: From the calculated corrected flow , add the corrected flow to the initial flow of the pipe to obtain the corrected water supply pipe flow (Formula 7);

[0098] S35: Repeat steps S32~S34 until the closed pressure closure difference of the looped pipe network is less than 0.001, that is, the looped pipe network satisfies the energy equation (Formula 4).

[0099] (Energy equation) (4)

[0100] (Continuity equation) (5)

[0101] (Pipe pressure drop equation) (6)

[0102] (7)

[0103] Among them, (8)

[0104] And after correction, (9)

[0105] Based on the above method, after the iteration ends, the simulation parameters of the operation state of the water supply network are obtained, including the t monitoring nodes at each moment i water pressure at the branch pipe (abbreviated as "simulated node pressure "), t the pipeline pressure drop between adjacent monitoring nodes at each moment (abbreviated as "simulated pipeline pressure drop "), the pipeline flow rate between adjacent monitoring nodes at each moment (abbreviated as "simulated pipeline flow rate "), the pipeline flow velocity between adjacent monitoring nodes at each moment (abbreviated as "simulated pipeline water flow velocity "). Similarly, based on this method, the pipeline water flow velocity at the branch pipe of the monitoring node at each moment (abbreviated as "simulated node water flow velocity ") can also be calculated.

[0106] Furthermore, the simulated pipeline flow rate, simulated pipeline pressure drop obtained from the real-time hydraulic simulation analysis of the water supply network can be used to calculate the flow modulus of the pipeline between adjacent monitoring nodes at each moment (abbreviated as "pipeline simulated flow modulus "), and the calculation formula is as follows:

[0107] (10)

[0108] Step 4: Judge whether there is an abnormality in the state of the water supply network according to the water supply network data state abnormality detection algorithm.

[0109] This step can specifically include the following parts:

[0110] First, according to the measured parameters collected, including the measured node pressure , measured node flow rate , measured node water flow velocity , and the pipeline design flow modulus The historical simulation parameters obtained from the simulation analysis, including the simulated node pressure, simulated pipeline flow rate, simulated pipeline water flow velocity, and pipeline simulated flow modulus, after removing abnormal data, the average values of each parameter are obtained respectively 、 、 、 。

[0111] Then, according to the simulated node pressure in the operation state simulation parameters of each monitoring node in the water supply network obtained from the foregoing steps 、simulated node water flow velocity , the calculated flow rate at the node , and the operation state simulation parameters of the pipeline between adjacent monitoring nodes, including simulated pipeline pressure drop 、simulated pipeline flow rate 、simulated pipeline water flow velocity 、pipeline simulated flow modulus , calculate the deviation degree of each parameter, and respectively form the deviation degree vector corresponding to the monitoring node parameters and the deviation degree vector corresponding to the pipeline parameters between adjacent monitoring nodes , where , is the number of sampling times.

[0112] Then, form a deviation degree vector group from the deviation degrees of each parameter of each monitoring node at different times X P , form a deviation degree vector group from the deviation degrees of each parameter of each pipeline at different times X C 。

[0113] ，

[0114] Finally, perform anomaly judgment according to the constructed data state anomaly detection algorithm. Here, taking the deviation degree vector corresponding to the monitoring node at the moment as an example for illustration, the included sub-steps are as follows:

[0115] S41: Taking the monitoring node as an example, calculate the distance between the deviation degree vector corresponding to the moment and the deviation degree vectors at other times , and sort the calculation results in ascending order to obtain the distance sequence of , where , and the calculation formula is as follows:

[0116] (11)

[0117] In the formula, and respectively represent the deviation degree vectors of the monitoring node at two different times; and respectively represent the and in the deviation degree vector group y the

[0118] S42: Calculate the z-nearest distance of each component in the deviation degree vector .

[0119] Taking the deviation degree vector of the monitoring node at time as an example, the z-th distance value in the distance sequence is called the z-nearest distance of

[0120] (12)

[0121] In the formula, is the z-th deviation degree vector closest when calculating the distance to the point.

[0122] S43: Based on the z-nearest distance of each vector in the obtained deviation degree vector group , calculate the z-distance neighborhood of each deviation degree vector.

[0123] Taking the deviation degree vector i of the monitoring node at time as an example, taking the vector as the center and the z-distance of

[0124] (13)

[0125] In the formula, is all deviation degree vectors whose distance to the vector is less than the z-nearest distance, represents the z-distance neighborhood of the vector .

[0126] S44: Based on the obtained deviation degree vector group For each vector in the z-distance neighborhood of each vector in the group, calculate the reachable distance of each deviation degree vector within the z-distance neighborhood. The calculation formula is as follows:

[0127] (14)

[0128] S45: Based on the reachable distance of each vector within the distance neighborhood of each vector in the obtained deviation degree vector group calculate the reachable density of each deviation degree vector. The calculation formula is as follows,

[0129] (15)

[0130] S46: Combine the reachable density of each vector in the deviation degree vector group and the z-distance neighborhood to calculate the outlier factor of each deviation degree vector. The calculation formula is as follows,

[0131] (16)

[0132] According to the above method, it is possible to process each monitoring node in the deviation degree vectors of each parameter at different times of the pipelines between adjacent monitoring nodes in

[0133] Using the above outlier detection algorithm, the measured parameters and simulated parameters of the operation status of the water supply network can be processed respectively to obtain the corresponding outlier factors. The data with outlier factors greater than 1 are set as outlier data, and then it is determined that there is an abnormality in the water supply system, and the next step is to comprehensively evaluate the status of the water supply network. Specifically, it includes the measured node pressure , the measured node flow , the measured node water flow velocity , the simulated node pressure , the simulated pipeline flow , the simulated pipeline water flow velocity , the pipeline simulated flow modulus and other parameters.

[0134] Step 5, find the cause of the abnormality according to the multimodal evaluation method for the operation status of the water supply network.

[0135] In this step, first, the modal evaluation factors (abbreviated as "factors") of the water supply network system need to be constructed. In the present invention, the factors can include M 1, M 2, M 3 these three.

[0136] , where, , , , where is the flow rate at the access point of the municipal water supply network (i.e., the main meter C1) at time is the sum of the flow rates of all monitoring nodes (i.e., the secondary meters C2) at time n represents the total number of data acquisitions, i.e., the sampling times. Among them, M 11 is the maximum value among the ratios of the sum of the water volumes of the main meter of the water supply network to the sum of the water volumes of all secondary meters in all sampling times; M 12 is the maximum value among the ratios of the average flow rate of the pipeline between adjacent monitoring nodes to the flow rate in the real-time simulation state in all sampling times; M 13 is the maximum value among the ratios of the designed flow rate modulus of the pipeline between adjacent monitoring nodes to the flow rate modulus in the real-time simulation state in all sampling times.

[0137] , where , , . Among them, M 21 is the maximum value among the ratios of the measured pressure of the monitoring nodes of the water supply network to the simulated pressure in the real-time state in all sampling times; M 22 is the maximum value among the ratios of the real-time simulated analysis pressure drop between adjacent monitoring nodes of the water supply network to the measured pressure drop in all sampling times; M 23 is the difference between the ratio of the measured flow rate of the monitoring nodes of the water supply network to the flow rate in the simulated analysis and 1 in all sampling times.

[0138] , where , , , where is the water consumption of the water-using unit at the branch pipe of the monitoring node at time is the drainage coefficient, is the drainage volume of the water-using unit at the branch pipe of the monitoring node at time. It can be understood that the data , and can all be collected from the water-using unit management system; is the water consumption quota of the water-using unit, and this data is determined by the data issued by the local water conservancy department. Among them, M 31is the maximum value among the ratios of the water consumption of the water-using unit to the water consumption quota of the water-using unit for all sampling times; M 32 is the maximum value of the ratio of the drainage volume converted from the water consumption of the water-using unit to the measured drainage volume for all sampling times; M 31 is the maximum value of the difference between the ratio of the drainage volume converted from the water consumption of the water-using unit to the measured drainage volume and 1 for all sampling times. It can be seen that factor M 3 is mainly related to the water consumption, drainage volume, and water consumption quota of the water-using unit, and is used to measure the water resource utilization efficiency or water resource management effect of the water-using unit, and to evaluate whether the water-using unit meets the water-saving goal or whether there is an unreasonable water use phenomenon.

[0139] Based on the defined multi-modal evaluation factors, a multi-modal factor comprehensive judgment formula for evaluating the operation status of the water supply pipe network system can be constructed, which is expressed as follows:

[0140] (17)

[0141] In the formula, K 1 、K 2 、K 3 are modal coefficients determined by the characteristics of the water supply pipe network system (such as the pipe network layout, the number of nodes, etc., the pipe network topology structure, the physical properties of the pipes such as material and roughness, and the hydraulic characteristics such as the pressure distribution, flow distribution, and resistance coefficient of the pipe network) and the water use characteristics of the water-using unit (such as the variation law of the daily water consumption and hourly water consumption of the water-using equipment, and the different water use patterns of the production and living units).

[0142] Among them, factors M, M 1 、M 2 、M 3 can all be adjusted and dynamically set the alarm threshold according to the corresponding requirements. For example, set M、 M 1 、M 2 、M 3 of the alarm threshold are all A1 , A1 is 1.1. When M is greater than 1.1, it indicates that there is an abnormality in the water supply pipe network system, and system troubleshooting and maintenance are required. The judgment logic can be combined with Figure 4 shown as follows:

[0143] (a) M When 1 is greater than 1.1, it is divided into three cases:

[0144] When M 11 is greater than 1.1: ① M 11is the maximum value in M1, indicating that the water supply pipe network does not satisfy the continuity equation. The sum of the flow rates of the secondary water meters is less than the flow rate of the primary water meter, and the water production and sales difference is greater than 10%; and M 12 、M 13 When both are greater than 1.05, it indicates that the simulated pipe flow rate is less than the average flow rate, and it is deduced that M 12 、M 23 There is leakage in the pipe between the adjacent monitoring nodes corresponding to the maximum value; ② M 12 、M 13 When both are less than 1.05, it indicates that the pipe leakage is within a reasonable range. However, at this time, the water production and sales difference of the water supply pipe network is greater than 10%, indicating that there is water use metering distortion at a monitoring node, that is, water meter metering distortion. Further, the water meters with problematic data readings can be determined by combining the water supply pipe network data status anomaly detection algorithm. That is, among the water meter detection data, the detection data anomaly factor calculated by the anomaly detection algorithm is greater than .

[0145] When M 11 、M 13 When both are greater than 1.1, it indicates that the water production and sales difference of the water supply pipe network is greater than 10%, M 13 The flow modulus change of the pipe between the two nodes corresponding to the maximum value is the largest, and there is leakage in the pipe between the two nodes.

[0146] When M 11 When it is less than 1, it indicates that the sum of the readings of the secondary meters in the water supply pipe network is greater than the reading of the primary meter (i.e., water meter metering distortion). Further, the water meters with problematic data readings can also be determined by combining the water supply pipe network data status anomaly detection algorithm.

[0147] (b)When M When 2 is greater than 1.1, it indicates that there are abnormalities in the flow distribution or pressure distribution of the water supply pipe network system, and further analysis is required. Usually, it is also divided into two situations:

[0148] When M 21 、M 22 When both are greater than 1.1, the node pressure corresponding to its maximum value is less than 90% of the average value and the pipeline pressure drop between the corresponding two nodes is greater than 110% of the average value, indicating that M 22 There is leakage in the pipe between the two nodes corresponding to the maximum value.

[0149] When M 23When it is greater than 0.05, M 23 The monitoring node corresponding to the maximum value i The water meter reading at this point is greater than 5% of the theoretical value, indicating that the water meter measurement is distorted.

[0150] (c) When M When 3 is greater than 1.1, it indicates that there are abnormal water - using problems in the water - using unit, which are usually divided into two cases:

[0151] When M 31 is greater than 1.1, it indicates that the water consumption of the water - using unit is greater than the average water consumption, and the water measurement is not distorted, indicating that M 31 the corresponding water - using unit greater than 1.1 has abnormal water use.

[0152] When M 32 is greater than 0.9 or less than 1.1, it indicates that the water supply and drainage do not match, the difference is greater than 10%, and M 33 is greater than 0.1, it indicates that M 33 the water measurement at the monitoring node corresponding to the maximum value is distorted.

[0153] In a specific application case, the topological structure model of the water supply network is as Figure 1 shown. According to the water - using characteristics of the water - using unit, K1 = 1 / 3, K2 = 1 / 3, K3 = 1 / 3 are taken. Given that the total - meter pressure P is 0.40 MPa, the water consumption of the water - using units corresponding to the monitoring nodes 1 - 6 of the water supply network are 0.01, 0.01, 0.02, 0.02, 0.01, 0.01 m 2 / s. It is assumed that there is a 10% leakage in the pipeline between monitoring nodes 3 and 4. For the sake of understanding, the abnormal phenomena of specific water - using units are not considered in this embodiment. M 31 、M 32 is taken as 1.00, and M 33 is taken as 0.00. The operation state of the water supply network is analyzed by the method described in the present invention.

[0154] The measured parameter values of the operation state of the water supply network obtained from step 2 are shown in Table 1:

[0155] Table 1 Monitoring parameter values of the operation state of the water supply network

[0156]

[0157] The simulated parameter values of the operation state of the water supply network obtained from step 3 are shown in Table 2:

[0158] Table 2 Simulated parameter values of the operation state of the water supply network

[0159]

[0160] The modal factor parameter values of the water supply pipe network operation status obtained from Step 5 are shown in Table 3 as follows:

[0161] Table 3 Modal factor parameter values of the pipe network operation status

[0162]

[0163] As can be seen from Table 3, M 11 = 1.10, M 12 = 1.22, M 13 = 1.57; M 21 = 1.14, M 22 = 1.56; M 31 = M 32 = 1.00. Therefore, M1 = 1.57, M2 = 1.56, M3 = 1.00; calculate M = + = 1.37. M = 1.37 > 1.10, from which it can be determined that there is an abnormality in the water supply pipe network system.

[0164] Further analysis shows that for M1, M 13 is the maximum value in M1, and its value is 1.57 > 1.10, corresponding to the second case in case (a), and the maximum value of M 13 among the monitoring nodes is the value corresponding to monitoring node 3. Thus, it is judged that there is a leak in the pipeline between monitoring nodes 3 and 4. For M2, M 22 is the maximum value in M2, and both M 21 and M 22 are greater than 1.10, corresponding to the first case in case (b), and the maximum values of M 21 and M 22 among the monitoring nodes are both monitoring node 3, indicating that there is a pipeline leak in the water supply pipe network system. Among them, the values of M 21 and M 22 for the pipeline between monitoring nodes 3 and 4 are 1.14 and 1.56 respectively, indicating that there is a leak in the pipeline between monitoring nodes 3 and 4, which is consistent with the assumption. Transmit the measured node pressures , simulated pipeline pressure drops , measured node flows , and measured node velocities of the abnormal data monitoring nodes, namely monitoring nodes 3 and 4, to data center C7 for data storage and early warning.

[0165] The multi-modal anomaly detection method of the present invention is illustrated by the above embodiments. According to the set modal evaluation factors, the specific problems existing in the pipe network can be accurately determined, and the key parameters of the pipe network operation status (such as the measured parameters corresponding to the abnormal data monitoring nodes) are uploaded to the data center and an alarm is issued.

[0166] The present invention proposes an analysis method for a multi-modal anomaly detection system of a water supply pipe network that fits multiple parameters such as working conditions, flow rates, flow velocities, and pressures according to the system water supply characteristics of different water supply pipe networks. It can effectively solve the problem of mutual interference between modes caused by situations such as pipe network leakage, inaccurate water consumption measurement, and abnormal changes in water consumption behavior, thereby determining the cause of the abnormality in the water supply pipe network system. The present invention can synchronously determine the synchronous changes in multi-modal parameters caused by changes such as pipe network leakage, inaccurate water consumption measurement, and abnormal water consumption behavior, and distinguish the specific modal spatio-temporal change characteristics under abnormal conditions of the water supply pipe network, so as to achieve simultaneous supervision of the water consumption behavior of water consumption units in the water supply pipe network, the leakage situation of the water supply pipe network, the status of water meters at metering nodes, and the accuracy of the data transmitted by detection equipment in the water balance test.

[0167] Furthermore, an electronic device is also disclosed in the embodiments of the present invention. It mainly includes a processor and a memory. The processor is mainly used to call the computer program stored in the memory and execute the real-time monitoring method for the operation status of the water supply pipe network provided in the above embodiments. Those skilled in the art can clearly understand that for the sake of convenience and brevity of description, the specific working process of the program code in the above-described electronic device and computer-readable storage medium can refer to the corresponding process in the foregoing method embodiments and will not be elaborated here.

[0168] Finally, it should be noted that although the embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the above specific embodiments and application fields. The above specific embodiments are merely illustrative and guiding, rather than restrictive. Under the inspiration of this specification, those of ordinary skill in the art can also make many forms without departing from the scope protected by the claims of the present invention, and all of these fall within the scope of protection of the present invention.

Claims

1. A real-time monitoring method for the operation status of a water supply pipe network, characterized in that, Including: Step 1: Based on the obtained topological structure and pipeline parameters of the water supply network, establish a digital topological structure model of the network. The pipeline parameters include the pipeline length , the pipeline diameter , the pipeline design flow modulus , where is the number of the monitoring node of the water supply network, and is the adjacent monitoring node; Step 2: Collect the measured parameters of the operation status of the water supply pipe network within the control area of the water supply pipe network, and preliminarily calculate the measured flow production and sales difference , the measured pipeline pressure drop , the calculated flow rate at the node ; The measured parameters include the measured node pressure , the measured node flow rate , the measured water flow velocity at the node ; is the collection time; Step 3: Based on the pipe network topological structure model, perform real-time pipe network hydraulic simulation analysis and calculation on the measured node pressure , the measured node flow to obtain the simulation parameters of the water supply pipe network operation state; the simulation parameters include the simulated node pressure , the simulated pipe pressure drop , the simulated pipe flow , the simulated pipe water flow velocity , the simulated node water flow velocity , the pipe simulated flow modulus ; Step 4: Based on the measured parameters and simulation parameters, combine with the water supply network data status anomaly detection algorithm to determine whether there is an anomaly in the water supply network status, and enter Step 5 when it is determined that there is an anomaly. Step 5: Based on a predefined multi-modal factor comprehensive judgment formula, find the cause of the anomaly and locate the anomaly position; the multi-modal factor comprehensive judgment formula has multiple modal evaluation factors related to the measured parameters and simulation parameters. In the said Step 5: Obtain or calculate the average value of each parameter based on the node pressure, node flow rate, node water flow velocity in the measured parameters and historical simulation parameters, as well as the pipeline design flow modulus and historical pipeline simulation flow modulus , , , ; The modal evaluation factors include: , wherein, , , , in the formula, is the flow rate at the access point of the municipal water supply network at time is the sum of the flow rates of all monitoring nodes at time is the number of sampling times, is the total number of monitoring nodes in the water supply network; , where, , , ; , wherein, , , , where, is the water consumption of the water-using unit at the branch pipe of the monitoring node monitored in real time, the average value of the water consumption of the water-using unit at the branch pipe of the historical monitoring node, is the drainage coefficient, is the drainage volume of the water-using unit at the branch pipe of the monitoring node monitored in real time; The multi-modal factor comprehensive judgment formula is: In the multi-modal factor comprehensive judgment formula, K 1 、K 2 、K 3 is a modal coefficient determined by the characteristics of the water supply pipe network system and the water usage characteristics of water-using units; The M, M 1 、M 2 、M alarm thresholds of 3 are all dynamically adjustable; when M it is greater than the alarm threshold, it indicates that there is an abnormality in the water supply pipe network system, and it is necessary to further analyze M 1 、M 2 、M 3 to find the cause of the abnormality and locate the abnormal position.

2. The real-time monitoring method for the operation state of a water supply pipe network according to claim 1, characterized in that In the said Step 2, it also includes: Store the measured parameters in the form of a one-dimensional vector and denote it as ; The measured flow production and sales difference The calculation formula is as follows: In the formula, is the measured nodal flow rate at the access point of the municipal water supply network at the moment, is the sum of the measured nodal flow rates at the branch pipes of the monitoring nodes at the moment, is the total number of the monitoring nodes of the water supply network; The measured pipeline pressure drop The calculation formula is as follows: In the formula, and respectively represent the measured node pressures at adjacent monitoring nodes at the moment; Calculate the flow rate at the node The calculation formula is as follows: In the formula, represents the monitoring node the pipe diameter at the branch pipe, is the measured water flow velocity of the node.

3. The real-time monitoring method for the operation status of a water supply pipe network according to claim 1, characterized in that In the said Step 3, based on the energy equation and the water supply network continuity equation, conduct real-time network hydraulic simulation analysis and calculation.

4. The real-time monitoring method for the operation status of a water supply pipe network according to claim 1, characterized in that In step 3, the pipe simulated flow modulus is calculated based on the simulated pipe flow and the simulated pipe pressure drop The calculation formula is as follows: 。 5. The real-time monitoring method for the operation status of a water supply pipe network according to claim 1, characterized in that The said Step 4 specifically includes: Calculate the average value of each parameter based on the node pressure, node flow rate, node water flow velocity in the measured parameters and historical simulation parameters, as well as the pipeline design flow modulus and historical pipeline simulation flow modulus , , , ; Calculate the deviation degrees of each parameter and form a deviation degree vector corresponding to the parameters of the monitoring node ， and a deviation degree vector corresponding to the pipeline parameters between adjacent monitoring nodes , where , is the number of sampling times; The deviation degrees of each parameter at each monitoring node at different times are combined to form a deviation degree vector group X P , and the deviation degrees of each parameter at different times in the pipeline between adjacent monitoring nodes are combined to form a deviation degree vector group X C ; The water supply network data status anomaly detection algorithm includes: Calculation monitoring node Or the pipeline between adjacent monitoring nodes at The deviation degree vector corresponding to the moment and the deviation degree vectors at other moments Distance , and sort the calculation results in ascending order to obtain Distance sequence of ; Calculate the z-nearest distance, z-distance neighborhood, reachable distance, and reachable density within the z-distance neighborhood for each component in the deviation degree vector in sequence. ​ Combined deviation degree vector group or Calculate the outlier factor of each deviation degree vector according to the reachable density and z-distance neighborhood of each vector in When the anomaly factor is greater than 1, it is determined that there is an anomaly in the water supply network status.

6. The real-time monitoring method for the operation status of the water supply network according to claim 1, characterized in that The M, M 1 、M 2 、M when the alarm thresholds of 3 are all the threshold A1; Among them, M When 1 is greater than the threshold value A1, it includes three cases: When M 11 is greater than the threshold A1: ① M 11 is M the maximum value in 1 and M 12 、M 13 both are greater than the threshold A2, it indicates that M 12 、M 23 there is a leak in the pipeline between adjacent monitoring nodes corresponding to the maximum value; ② M 12 、M 13 both are less than the threshold A2, it indicates that there is water consumption metering distortion at the corresponding monitoring node, and the location of the problem is determined through the data anomaly detection algorithm; When M 11 、M 13 are all greater than the threshold A1, it indicates that M 13 the maximum value corresponds to a leak in the pipeline between adjacent monitoring nodes; When M 11 When it is less than the threshold A3, it indicates that the water measurement is distorted, and the position where the problem exists is determined through the data anomaly detection algorithm; Among them, M When 2 is greater than the threshold value A1, it includes two cases: When M 21 、M 22 are all greater than the threshold A1, it indicates that M 22 the maximum value corresponds to a leak in the pipeline between adjacent monitoring nodes; When M 23 is greater than the threshold A4, it indicates that M 23 the water metering at the monitoring node corresponding to the maximum value is distorted; Among them, M When 3 is greater than the threshold value A1, it includes two cases: When M 31 is greater than the threshold A1, it indicates that M 31 the corresponding water usage unit greater than the threshold A1 has abnormal water usage; When M 32 greater than threshold A5 or less than threshold A1, and M 33 greater than threshold A6, it indicates that M 33 the water consumption measurement at the monitoring node corresponding to the maximum value is distorted.

7. The real-time monitoring method for the operation status of the water supply network according to claim 6, characterized in that The threshold values A1 is 1.1, A2 is 1.05, A3 is 1, A4 is 0.05, A5 is 0.9, and A6 is 0.

1.

8. A real-time monitoring system for the operation status of a water supply network, characterized in that, Including a data acquisition device, a data analysis module, and a data center; The data acquisition device is arranged at the water supply network monitoring nodes and is used to collect the measured parameters of the operation status of the water supply network within the control area of the water supply network. The data analysis module, which is communicatively connected to the data acquisition device, is used to implement the real-time monitoring method for the operation status of the water supply network according to any one of claims 1 to 7, and upload the monitoring results to the data center.

9. An electronic device, characterized in that, Including a processor and a memory; the processor is used to call the computer program stored in the memory and execute the real-time monitoring method for the operation status of the water supply network according to any one of claims 1 to 7.

Images