A pipe network operation monitoring system and method

Abstract

The application provides a pipe network operation monitoring system and method, which comprises a pressure detection device and an intelligent detection host; the intelligent detection host comprises an intelligent controller, and the intelligent controller comprises a fuzzy controller; adjacent valve wells of a pipe network are taken as a section, and the pressure detection device is arranged in the valve well at the head and tail of each section; the pressure detection device measures the pipe network operation pressure in real time and transmits the pipe network operation pressure to the intelligent controller; the intelligent controller processes the received pipe network operation pressure to obtain an exponential moving average value, obtains an average maximum value based on the exponential moving average value, and integrates the difference between the real-time measured pipe network operation pressure and the average maximum value to obtain a pressure integral value; the fuzzy controller takes the pressure integral value and the real-time measured pipe network operation pressure as inputs, performs fuzzy reasoning, and judges whether to shut down the valve or to issue an alarm.

Description

A pipeline network operation monitoring system and method Technical Field

[0001] This invention belongs to the field of pipeline operation monitoring technology, and particularly relates to a pipeline operation monitoring system and method. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Urban underground pipe networks house various pipelines, such as gas, heating, water supply, and rainwater. These pipelines are buried deep underground and cannot be inspected regularly. Once a pipeline leak occurs, it will cause significant damage and disrupt normal production and daily life. Ensuring the safe operation of these pipelines has become an urgent problem to be solved.

[0004] The existing technology patent number 201620639824.8 describes the use of Internet of Things sensors to collect pipeline pressure and flow, and when an abnormality occurs, the control center issues a valve shut-off command.

[0005] The existing technology patent number 201710071287.0 describes the use of a level gauge to analyze and determine whether there is water accumulation in the pipe gallery, and when water accumulation occurs, the pipeline valve is shut off using a PLC.

[0006] The aforementioned prior art has at least the following problems:

[0007] 1. Long-distance pipelines are located in remote urban areas where urban power cannot reach them, or urban power failures may cause the pipeline monitoring system to lack mains power for pressure detection and valve shut-off, which in turn may prevent the entire pipeline monitoring system from operating safely and stably.

[0008] 2. When monitoring pipeline operation, the large diameters of water supply, reclaimed water, and heating pipelines mean that relying solely on measuring pipeline pressure to determine valve closure pose a significant risk of interference. This interference primarily includes measurement errors, sensor malfunctions, and fluid noise. Even within acceptable pipeline interference levels, if the detected pressure exceeds the set value, initiating valve closure will lead to malfunction. Each valve closure requires time, and malfunctioning valve closures can severely impact municipal infrastructure in the area.

[0009] Furthermore, the upper limit of pressure within the pipeline is called the yield strength. Current pipeline pressure testing methods shut off valves once the pressure exceeds the yield strength. The limitation of this method is that:

[0010] A. Measurement errors, sensor malfunctions, and fluid noise can easily cause interference, leading to erroneous valve closures and unplanned pipeline shutdowns, causing unnecessary difficulties and economic losses for users.

[0011] B. Even if some measurement points exceed the set upper limit, it will not instantly damage the pipeline and cause material deformation. Only when the combined effect of pressure accumulation and time reaches a certain limit will the pipeline undergo irreversible deformation and burst.

[0012] 3. When implementing pipeline operation monitoring, the existing technology uses liquid level monitoring in the pipe gallery to determine whether a leak has occurred. This solution is a post-event monitoring, that is, an alarm is issued and the valve is shut off only when a pipe bursts in the pipe gallery and the water reaches a certain depth. This method is not real-time or predictable and cannot prevent leaks from occurring. Summary of the Invention

[0013] To overcome the shortcomings of the prior art, the present invention provides a pipeline operation monitoring system that can ensure the safe operation of pipelines.

[0014] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0015] Firstly, a pipeline network operation monitoring system is disclosed, comprising:

[0016] Pressure detection device and intelligent detection host;

[0017] The intelligent detection host includes an intelligent controller, and the intelligent controller includes a fuzzy controller;

[0018] The adjacent valve wells of the pipeline network are considered as a section, and pressure detection devices are installed in the valve wells at both the beginning and end of each section.

[0019] The pressure detection device measures the pipeline operating pressure in real time and transmits it to the intelligent controller. The intelligent controller processes the received pipeline operating pressure to obtain an exponential moving average value, and obtains an average maximum value based on the exponential moving average value. The pressure integral value is obtained by integrating the difference between the real-time measured pipeline operating pressure and the average maximum value.

[0020] The fuzzy controller uses the pressure integral value and the real-time measured pipeline operating pressure as input to perform fuzzy reasoning to determine whether to shut off the valve or issue an alarm.

[0021] As a further technical solution, a photovoltaic energy storage device is also included, which provides power to the pressure detection device and the intelligent detection host respectively.

[0022] As a further technical solution, the pressure detection device includes a pressure detection circuit, which includes a pressure sensor. The pressure sensor transmits the measured signal to a second amplifier and a fourth amplifier. The negative terminal of the second amplifier is connected to the positive terminal of a third amplifier. The outputs of the third amplifier and the fourth amplifier are respectively connected to the input terminal of a first amplifier. The first amplifier outputs the pressure difference signal of the pipeline.

[0023] As a further technical solution, the pressure detection device also includes a valve shut-off circuit, which uses a bidirectional thyristor to directly switch on and off the AC power supply circuit of the solenoid valve.

[0024] Secondly, a method for monitoring pipeline operation based on photovoltaic power supply is disclosed, including:

[0025] Real-time measurement of pipeline operating pressure;

[0026] The received pipeline operating pressure is processed to obtain the exponential moving average, and the average maximum value is obtained based on the exponential moving average. The pressure integral value is obtained by integrating the difference between the real-time measured pipeline operating pressure and the average maximum value.

[0027] The fuzzy controller uses the pressure integral value and the real-time measured pipeline operating pressure as inputs to perform fuzzy reasoning to determine whether to shut off the valve or issue an alarm.

[0028] As a further technical solution, the pressure integral value and the real-time measured pipeline operating pressure are used as inputs to the fuzzy controller for fuzzy inference. If the pipeline exceeds the tolerance range, the adjacent pipeline valves and upstream pumping stations are shut off to prevent accidents. If the pipeline pressure is lower than the normal range, an alarm is issued to remind the site to conduct inspections.

[0029] As a further technical solution, the received pipeline operating pressure is processed to obtain an exponential moving average value. The specific calculation method is as follows:

[0030] V(K)=α*V(K-1)+(1-α)*P(K) Formula (1)

[0031] Wherein, V(K): the most recent The average value at each sampling time;

[0032] P(K): The sampled value of the pipeline operating pressure measured at time K;

[0033] α: Adjustable hyperparameter.

[0034] As a further technical solution, it also includes: calculating the sampling results within a control period based on formula (1) to obtain the exponential moving average curve, and removing the error based on the curve.

[0035] As a further technical solution, for the sampling results within a control cycle, the bubble sort algorithm is used to calculate the maximum mean value and the minimum mean value;

[0036] After removing the maximum mean value and the minimum mean value in the array, the average values of the first five items and the last five items in the array are respectively taken as the maximum and minimum mean values.

[0037] As a further technical solution, the difference between the real-time measured pipeline operation pressure and the average maximum value is integrated to obtain the pressure integral value. The specific calculation formula is:

[0038] Calculation formula: S = ∫|P(K) - AV_max|; Formula (2)

[0039] In the formula, S is the pressure integral value;

[0040] P(K) is the pressure detection value;

[0041] AV_max is the maximum value calculated by the bubble sort.

[0042] As a further technical solution, the fuzzy controller takes the pressure integral value and the real-time measured pipeline operation pressure as inputs, conducts fuzzy reasoning, and judges whether to shut off the valve or give an alarm. Specifically:

[0043] During factory production, the ultimate pressure P_max and the action time T1 of the pipeline deformation are measured, then S_max = P_max * T1;

[0044] During normal pipeline operation, the annual normal minimum pressure value P_min of the pipeline is statistically calculated (this value appears at least 5 times in a year), and S_min = P_max * 1min is defined;

[0045] When the measured pressure P(K) exceeds the average maximum value calculated by the bubble sort method, the integral calculation is started. When the calculated value exceeds S_max, the adjacent section valve is shut off, and the upstream pumping station booster pump is shut off;

[0046] Assume S1 < S_max < S2, only the S2 area performs the shut-off;

[0047] When the measured pressure P(K) is less than the average minimum value calculated by the bubble sort method, the integral calculation is started. When the calculated value exceeds S_min, a warning is issued to remind that the pipeline outlet pressure in this section has dropped and maintenance inspection is required.

[0048] The above one or more technical solutions have the following beneficial effects:

[0049] To avoid malfunctions caused by simple pressure monitoring, this invention utilizes online self-learning to obtain the real-time average maximum and minimum values ​​of the pipeline in the current section, accurately calculating the pipeline's operating range. Different sections of underground pipelines have varying actual operating pressures due to differences in their distance from the booster pump station. Through online self-learning, the upper and lower limits of the pipeline's operating pressure in that section are plotted, providing a better understanding of the actual pipeline conditions.

[0050] This invention utilizes an exponential moving average algorithm to process data, assigning greater weight to more recent sampling times to better reflect current changes. This effectively eliminates error interference and smooths outliers.

[0051] This invention proposes a pressure integral algorithm that incorporates a pressure accumulation time method. Based on the inherent pressure-bearing characteristics of the pipeline, exceeding pressure limits at certain points will not lead to deformation. Irreversible deformation only occurs when the pressure accumulation time reaches its limit. The pressure accumulation time significantly improves the stability, accuracy, and speed of valve shut-off. Stability: It avoids erroneous shut-offs near the pressure upper limit that occur when only pressure is measured. Accuracy: It greatly improves valve shut-off accuracy, ensuring accurate valve closure before pipeline plastic deformation, thus guaranteeing pipeline reliability. Speed: The algorithm is fast, matching the pipeline fluid characteristics for rapid valve closure.

[0052] This invention improves the inspection efficiency of maintenance personnel. Underground pipelines are often long, sometimes exceeding ten kilometers. Relying solely on manual inspection requires a significant amount of manpower and resources, and it is impossible to accurately locate fault points. This invention can accurately locate pipeline faults and issue warning signals in advance.

[0053] For long-distance pipelines that may be located in suburban areas where urban power cannot reach, installing dedicated transformer substations for valve wells is expensive and involves difficult cable laying. This invention, a photovoltaic power supply device, effectively solves the power supply problem for pipeline sections far from urban areas, reducing costs and implementation difficulty.

[0054] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0055] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0056] Figure 1 is a diagram showing the components of a pipeline network for safe operation according to an embodiment of the present invention;

[0057] Figure 2 is a control strategy diagram for safe operation of the pipeline network according to an embodiment of the present invention;

[0058] Figure 3 is a schematic diagram of the pressure detection circuit according to an embodiment of the present invention;

[0059] Figure 4 is a schematic diagram of the valve shut-off circuit according to an embodiment of the present invention;

[0060] Figure 5 is a schematic diagram of the exponential moving average fitting curve of the embodiment of the present invention;

[0061] Figure 6 is a diagram of the subarray bubble sort structure according to an embodiment of the present invention;

[0062] Figure 7 is a sub-pressure accumulation time analysis diagram of an embodiment of the present invention;

[0063] Figures 8(a)-8(c) show the sub-fuzzy controller input P(k), fuzzy controller input S(k), and fuzzy controller output u, respectively, in an embodiment of the present invention.

[0064] Figure 9 is a schematic diagram of the photovoltaic power supply device according to an embodiment of the present invention;

[0065] Figure 10 is a schematic diagram of the long-distance pipeline detection layout according to an embodiment of the present invention. Detailed Implementation

[0066] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0067] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0068] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0069] Example 1

[0070] Referring to Figure 1, this embodiment discloses a pipeline network operation monitoring system, including: a pressure detection device, an intelligent detection host, and a photovoltaic energy storage device;

[0071] The photovoltaic energy storage equipment supplies power to both the pressure detection device and the intelligent detection host.

[0072] The intelligent detection host includes an intelligent controller, which in turn includes a fuzzy controller. Specifically, the fuzzy controller refers to a fuzzy inference engine built using an expert rule base. The intelligent controller's functions are: to receive data from the pressure detection device, analyze and process the data, and, using fuzzification, fuzzy inference, and defuzzification steps, output control commands, and, if necessary, shut off the valve.

[0073] The adjacent valve wells of the pipeline network are considered as a section, and pressure detection devices are installed in the valve wells at both the beginning and end of each section.

[0074] Pressure detection devices are installed in valve wells at both ends of each pipeline section. Sampling holes are set on the pipeline inside the valve wells, and pressure sensors are installed on the sampling holes. The pressure sensors measure the pipeline operating pressure in real time and transmit it to the intelligent controller. The intelligent controller processes the received pipeline operating pressure to obtain an exponential moving average, and obtains an average maximum value based on the exponential moving average. The pressure integral value is obtained by integrating the difference between the real-time measured pipeline operating pressure and the average maximum value.

[0075] The fuzzy controller uses the pressure integral value and the real-time measured pipeline operating pressure as inputs to perform fuzzy reasoning and determine whether to shut off the valve or issue an alarm.

[0076] In a specific implementation example, a valve well is installed every 2-3 kilometers in the municipal long-distance pipeline network. Pressure detection devices and intelligent controllers are installed in each valve well. After exiting the booster pump station, the pressure in the municipal long-distance pipeline gradually decreases due to fluid friction losses, as shown in Figure 10. The pressure value varies in each pipeline section. Therefore, the pressure of that section is measured in each valve well.

[0077] For example, when an abnormal pressure exceeds the limit in pipeline section 2, the intelligent controller will determine the cause and issue a valve closing command to shut down valve well 1 and valve well 2 as well as the upstream booster pump station, thereby reducing the scope of the accident and preventing secondary accidents in other pipeline sections.

[0078] When the system is running normally, the processor collects the pressure on the pipeline and inputs it into the fuzzy controller for analysis and judgment.

[0079] In this embodiment, as shown in Figure 9, the photovoltaic energy storage device consists of photovoltaic panels, a battery, and a photovoltaic controller. The photovoltaic panels are installed outside the valve well, while the battery and photovoltaic controller are installed inside the valve well. The electricity generated by the photovoltaic system is preferentially supplied to the monitoring system, and any excess electricity is stored in the battery. When there is no photovoltaic power, the battery supplies power to the detection device.

[0080] As shown in Figure 2, during the operation: pipeline detection: collect pressure data at representative locations on the pipeline.

[0081] Intelligent analysis and judgment: Based on the comparison of pressure and pressure-time integral values, fuzzy judgment is used for identification. If the pipeline exceeds its capacity, adjacent pipeline valves and upstream pumping stations are shut off to prevent accidents. If the pipeline pressure is below the normal range, an alarm is issued to remind on-site inspection.

[0082] Referring to Figure 3, the pressure detection circuit consists of: XDCR1, a pressure sensor used to measure pipeline pressure; U2B, a non-inverting amplifier; U2C, R7, and R8, forming a reference voltage circuit; and U2A, a differential amplifier used to acquire the pressure difference signal from the pipeline.

[0083] The improvement of this circuit lies in the design of a reference voltage. By using a differential amplifier, the pressure difference is compared with the reference voltage, thus accurately acquiring the pressure signal.

[0084] Referring to Figure 4, the valve shut-off circuit uses IC1 as an optocoupler. Signs represents the signal input. When the input is high, IC1 conducts, the bidirectional thyristor T1 conducts, and the circuit is turned on. When the input is low, IC1 is turned off, the bidirectional thyristor T1 is turned off, and the circuit is turned off.

[0085] The valve shut-off circuit is connected to the intelligent controller, which issues the valve shut-off command.

[0086] The improvements to this circuit are as follows: IC1 is optocoupler isolated, isolating the controller from field interference; a bidirectional thyristor T1 is used, allowing direct switching of the solenoid valve's AC power supply circuit; a C1R5 low-pass filter is added to eliminate noise; and a fuse F1 is added to protect the solenoid valve's AC power supply circuit.

[0087] In this implementation example, the intelligent controller can be a DSP controller or a microprocessor, etc.

[0088] Example 2

[0089] The purpose of this embodiment is to provide a method for monitoring pipeline operation based on photovoltaic power supply. The specific calculation steps are as follows:

[0090] (1) Define the control cycle as 1 minute and the sampling cycle as 1 second, and measure the pipeline operating pressure P(K) in real time;

[0091] (2) The exponential moving average value V(K) is obtained by processing P(K) using the exponential moving average algorithm;

[0092] (3) Use the bubble sort algorithm to calculate the average minimum value and the average maximum value;

[0093] (4) Calculate the pressure integral S(K);

[0094] (5) Input the pressure integral S(K) and pressure P(K) into the fuzzy controller, perform fuzzy reasoning, and determine whether to shut off the valve or issue an alarm.

[0095] Regarding the exponential moving average algorithm:

[0096] The index-weighted average is calculated as follows:

[0097] V(K)=α*V(K-1)+(1-α)*P(K) Formula (1)

[0098] V(K): the most recent The average value at each sampling time;

[0099] P(K): Pressure sample value at time K;

[0100] α: Adjustable hyperparameter.

[0101] Detailed calculation steps for exponential moving average:

[0102] Define the control period as 1 minute and the sampling period as 1 second. Then, according to formula (1), the sequential calculation results within one control period are:

[0103] At time K=0, V(0)=P(0)

[0104] At time K=1, V(1) = α*V(0) + (1-α)*P(1)

[0105] At time K=2, V(2)=α*V(1)+(1-α)*P(2)

[0106] At time K=3, V(3) = α*V(2) + (1-α)*P(3)

[0107] At time K=4, V(4) = α*V(3) + (1-α)*P(4)

[0108] And so on

[0109] At time K=56, V(56) = α*V(55) + (1-α)*P(56)

[0110] At time K=57, V(57)=α*V(56)+(1-α)*P(57)

[0111] At time K=58, V(58)=α*V(57)+(1-α)*P(58)

[0112] At time K=59, V(59)=α*V(58)+(1-α)*P(59)

[0113] Characteristics of exponential moving average:

[0114] The closer the sampling time, the greater the weight assigned, and the more it reflects the current changes.

[0115] It effectively eliminates error interference and smooths outliers.

[0116] As shown in Figure 5, the discrete point is the pressure sampling point P(K) at a certain moment, and the exponential moving average curve is the curve calculated by this algorithm. Through the exponential moving average algorithm, the sampled data is compared with AV_max below. If it exceeds 30% of AV_max, it is discarded. Thus, interferences A, B, and C can be effectively eliminated, thus eliminating the adverse effects of interference.

[0117] Bubble sort algorithm for finding maximum and minimum mean:

[0118] The idea behind the bubble sort algorithm is to compare and sort the data in the sampled array pairwise, swapping the largest values ​​one by one, until the data in the array is sorted from smallest to largest. The array result after bubble sort is shown in Figure 6.

[0119] Specifically, based on the defined control period of 1 minute and the sampling period of 1 second, the sampling array consists of 60 data points collected during the control period.

[0120] Find the maximum and minimum mean.

[0121] After removing the maximum and minimum values ​​from the array, the average of the first five and last five elements of the array is taken as the maximum and minimum average, respectively. The calculation method is as follows:

[0122] Remove the minimum value B(1), and calculate the average minimum value:

[0123]

[0124] Remove the maximum value B(60), and calculate the average maximum value.

[0125]

[0126] The program for the bubble sort algorithm is as follows:

[0127]

[0128]

[0129] Implementation using the pressure integral method:

[0130] The upper limit of pressure inside a pipeline is called the yield strength. Current pipeline pressure testing methods shut off valves once the pressure exceeds the yield strength. The limitation of this method is that:

[0131] Measurement errors, random errors, and noise interference can easily lead to inaccurate measurements. Error interference can cause valves to close malfunctions and pipelines to shut down unplanned, causing unnecessary difficulties and economic losses to users.

[0132] Even if several measurement points exceed the set upper limit value, the pipeline will not be instantly damaged, resulting in material deformation. Only when the pressure accumulation time acts doubly and reaches a certain limit value will irreversible deformation of the pipeline occur.

[0133] Therefore, the present invention proposes to use the pressure integration method to judge whether the pipeline pressure is deformed or leaked.

[0134] Calculation formula: S = ∫|P(K) - AV_max| Formula (4)

[0135] In the formula, S is the pressure integration calculation value;

[0136] P(K) is the pressure detection;

[0137] AV_max is the maximum value calculated by bubble sort;

[0138] At the time of factory shipment, the ultimate pressure P_max and the action time T1 of pipeline deformation are measured; then S_max = P_max * T1;

[0139] During the normal operation of the pipeline, the annual normal minimum pressure value P_min of the pipeline is statistically calculated (this value appears at least 5 times in a year), and S_min = P_max * 1min is defined;

[0140] When the measured pressure P(K) exceeds the average maximum value calculated by the bubble sort method, the integral calculation is started. When the calculated value exceeds S_max, the valves in the adjacent section are shut off, and the booster pump in the upstream pumping station is shut off

[0141] As shown in Figure 7, assume that S1 < S_max < S2, only the S2 area is shut off.

[0142] When the measured pressure P(K) is less than the average minimum value calculated by the bubble sort method, the integral calculation is started. When the calculated value exceeds S_min, a warning is issued to remind that the outlet pressure of the pipeline in this section has dropped and maintenance inspection is required.

[0143] The advantages of this algorithm are as follows:

[0144] By using online self-learning, the real-time average maximum value and real-time average minimum value of the pipeline in this section are obtained, and the operation range of the pipeline is accurately calculated.

[0145] The exponential sliding average algorithm is used to process the data. The closer the sampling time is, the greater the weight is given, and the more it reflects the current change situation. It effectively eliminates error interference and smooths the abnormal points.

[0146] By using fuzzy control, the valve control is realized.

[0147] MPPT based on fuzzy control mainly consists of three steps: fuzzification, fuzzy inference, and defuzzification.

[0148] Blur:

[0149] Based on the analysis of the pipeline network operation system, this invention selects the pressure detection value P(K) and the pressure-time integral S(K) as the inputs to the fuzzy controller. This differs significantly from existing fuzzy controllers, whose inputs affect control efficiency.

[0150] Define the subsets of pressure detection value P(K), pressure-time integral S(K), and control variable U as {small, medium, large}, with the universe of discourse [0,1].

[0151] Triangular membership functions are used. The membership function curve of the pressure detection value P(K) is shown in Figure 8(a), the membership function curve of the pressure-time integral S(K) is shown in Figure 8(b), and the membership function curve of the control quantity is shown in Figure 8(c).

[0152] Fuzzy reasoning

[0153] Fuzzy reasoning employs Madani reasoning.

[0154] (1) The inference algorithm adopted is:

[0155] η i =μ ei [e(k)]∧μ eci [ec(k)]=min(μ ei [e(k)]·μ eci [ec(k)])

[0156] η i For fuzzy inference calculations; μ ei [e(k)] represents the pressure membership value; μ eci [ec(k)] represents the membership value of the pressure integral; ∫ represents the minimum value operation.

[0157] Establishing a fuzzy rule base:

[0158] In designing fuzzy rules, this invention delves into the characteristics of pipeline network operation and conducts an in-depth analysis of the pipeline network's operational process. A fuzzy rule base based on actual operational conditions is established.

[0159] Example of a fuzzy rule base creation approach:

[0160] If both the pressure and the pressure integral are low, it indicates that the system is currently experiencing a pressure leak and an alarm should be triggered.

[0161] The corresponding rules are designed as follows:

[0162] IF P is small and S is small, then u is ON.

[0163] If both the pressure and the pressure integral are high, it indicates that the current system pressure exceeds the upper limit of the pressure integral, and the valve needs to be shut off.

[0164] The corresponding rules are designed as follows:

[0165] IF P is large and S is large, then u is OFF.

[0166] Based on the above design concept, the following 9 fuzzy rules were designed.

[0167] IF P is small and S is small, then u is ON.

[0168] IF P is small and S is medium, then u is ON.

[0169] IF P is small and S is large, then u is ON.

[0170] IF P is medium and S is small, then u is 0.

[0171] IF P is in and S is in, then u is 0.

[0172] IF P is medium and S is large, then u is 0.

[0173] IF P is large and S is small, then u is 0.

[0174] IF P is large and S is medium, then u is 0.

[0175] IF P is large and S is large, then u is OFF.

[0176] Defuzzing

[0177] The centroid method is used for defuzzification. The defuzzification calculation method is as follows:

[0178]

[0179] The overall workflow of the fuzzy algorithm is shown in Figures 8(a)-8(c).

[0180] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A pipeline network operation monitoring system, characterized in that, include: Pressure detection device and intelligent detection host; the intelligent detection host includes an intelligent controller, and the intelligent controller includes a fuzzy controller; Adjacent valve wells in the pipeline network are considered as a segment. Pressure detection devices are installed in the valve wells at both the beginning and end of each segment. These pressure detection devices measure the pipeline network operating pressure in real time and transmit the data to an intelligent controller. The intelligent controller processes the received operating pressure to obtain an exponential moving average, and then calculates the average maximum value based on this exponential moving average. Finally, it integrates the difference between the real-time measured operating pressure and the average maximum value to obtain the pressure integral value. The specific calculation method for obtaining the exponential moving average by processing the received operating pressure is: V(K) = α V(K-1)+(1-α) P(K) where V(K): the most recent The average value at each sampling time; P(K): the sampled value of the pipeline operating pressure measured at time K; α: an adjustable hyperparameter; the fuzzy controller uses the pressure integral value and the real-time measured pipeline operating pressure as input to perform fuzzy inference to determine whether to shut off the valve or issue an alarm, specifically: at the factory, the ultimate pressure P_max of pipeline deformation and the action time T1 are measured, then S_max = P_max T1; During normal pipeline operation, the annual minimum normal pressure value P_min is calculated, and S_min is defined as P_min = P_min. 1 min; when the measured pressure P(K) exceeds the average maximum value calculated by the bubble sort method, the integral calculation is started. When the calculated value exceeds S_max, the valves in the adjacent section are shut off, and the booster pump in the upstream pumping station is shut off; assuming S1 < S_max < S2, only the S2 area is shut off. When the measured pressure P(K) is less than the average minimum value calculated by the bubble sort method, the integral calculation is started. When the calculated value exceeds S_min, a warning is issued to remind that the outlet pressure of the pipeline in this section has dropped and maintenance inspection is required.

2. The pipeline network operation monitoring system as described in claim 1, characterized in that, The pressure detection device includes a pressure detection circuit, which includes a pressure sensor. The pressure sensor transmits the measured signal to a second amplifier and a fourth amplifier. The negative terminal of the second amplifier is connected to the positive terminal of a third amplifier. The outputs of the third amplifier and the fourth amplifier are respectively connected to the input terminal of a first amplifier. The first amplifier outputs the pressure difference signal of the pipeline.

3. The pipeline network operation monitoring system as described in claim 1, characterized in that, The pressure detection device also includes a valve shut-off circuit, which uses a bidirectional thyristor to directly switch on and off the AC power supply circuit of the solenoid valve.

4. A method for monitoring pipeline operation based on photovoltaic power supply, characterized in that, include: Real-time measurement of pipeline operating pressure; The received pipeline operating pressure is processed to obtain an exponential moving average, and an average maximum value is obtained based on the exponential moving average. The pressure integral value is obtained by integrating the difference between the real-time measured pipeline operating pressure and the average maximum value. The specific calculation method for obtaining the exponential moving average by processing the received pipeline operating pressure is: V(K) = α V(K-1)+(1-α) P(K) where V(K): the most recent The average value at each sampling time; P(K): the sampled value of the pipeline operating pressure measured at time K; α: an adjustable overparameter; using the pressure integral value and the real-time measured pipeline operating pressure as inputs to the fuzzy controller, fuzzy inference is performed to determine whether to shut off the valve or issue an alarm. Specifically: at the factory, the ultimate pressure P_max of pipeline deformation and the action time T1 are measured, then S_max = P_max T1; During normal pipeline operation, the annual minimum normal pressure value P_min is calculated, and S_min is defined as P_min = P_min. 1 min; when the measured pressure P(K) exceeds the average maximum value calculated by the bubble sort method, the integral calculation is started. When the calculated value exceeds S_max, the valves in the adjacent section are shut off, and the booster pump in the upstream pumping station is shut off. Assuming S1 < S_max < S2, only the S2 area is shut off. When the measured pressure P(K) is less than the average minimum value calculated by the bubble sort method, the integral calculation is started. When the calculated value exceeds S_min, a warning is issued to remind that the outlet pressure of the pipeline in this section has dropped and maintenance inspection is required.

5. A pipeline network operation monitoring method based on photovoltaic power supply as described in claim 4, characterized in that, The fuzzy controller uses the pressure integral value and the real-time measured pipeline operating pressure as inputs to perform fuzzy inference. If the pipeline exceeds the tolerance range, it will shut off the valves of adjacent pipelines and upstream pumping stations to prevent accidents. If the pipeline pressure is lower than the normal range, it will issue an alarm to remind the site to conduct inspections.

6. The pipeline operation monitoring method based on photovoltaic power supply as described in claim 4, characterized in that, Also includes: Based on Formula 1, the sampling results within a control period are calculated to obtain the exponential moving average curve, and the error is removed based on this curve.

7. The pipeline operation monitoring method based on photovoltaic power supply as described in claim 4, characterized in that, For the sampling results within a control period, the bubble sort algorithm is used to find the maximum and minimum mean. After removing the maximum and minimum mean in the array, the average of the first five and last five items in the array is taken as the maximum and minimum mean, respectively.

8. A pipeline network operation monitoring method based on photovoltaic power supply as described in claim 4, characterized in that, The pressure integral value is obtained by integrating the difference between the real-time measured operating pressure of the pipeline network and the average maximum value. The specific calculation formula is as follows: Calculation formula: In the formula, S is the pressure integral value; P(K) is the pressure detection value; and AV_max is the maximum value calculated by bubble sort.

Images