Intelligent pipe network pressure control method and system

Abstract

The invention discloses an intelligent pipe network pressure control method and system, and relates to the technical field of pipe network pressure control. According to the fitting result of the historical water supply data of the water supply network in the target area and the corresponding influence factor data, the function relation between the water supply pressure and the water consumption is obtained, correlation analysis is conducted on the influence factor data and the water consumption, and influence factor types with high correlation with the water consumption are screened out; constructing a neural network model with high-correlation influence factor data as input and corresponding water consumption data as output, obtaining a water supply pressure fitting value by using the trained neural network model and the function relationship, and adjusting the pressure of the water supply pipe network in the target area to obtain a water supply pressure fitting value. The cooperative influence effect of various factors on the water demand can be fully considered, so that the intelligence degree and the refinement degree of the control effect of the time-phased pressure control strategy are improved, and the intelligent control demand of the pipe network pressure at the present stage is met.

Description

Technical Field

[0001] The present invention relates to the technical field of pipe network pressure control, and particularly to an intelligent pipe network pressure control method and system. Background Art

[0002] With the acceleration of urbanization and the growth of population, the scale and complexity of the water supply pipe network system are constantly increasing. Traditional water supply pipe network pressure control methods mainly rely on manual regulation and simple pressure regulating devices, such as manual pressure regulating valves, timed pressure control devices, etc. To a certain extent, these methods can meet the basic water supply needs, but there are obvious deficiencies in refined management and intelligent control. In recent years, with the rapid development of sensor technology, automation control technology and communication technology, intelligent pipe network pressure control technology has gradually emerged. By installing monitoring devices such as pressure sensors and flow meters in the pipe network, real-time collection of pipe network operation data is carried out, and combined with an automation control system and intelligent algorithms, real-time monitoring and dynamic regulation of the pipe network pressure are realized. However, there are still some problems in the existing intelligent pressure control technology, which limit its wide application in actual water supply systems.

[0003] Currently, the main strategies for dynamic pressure control include the following: Time-division pressure control: Set different pressure target values according to the day-night variation of water demand. For example, maintain a higher pressure during the day and reduce the pressure at night; Zoning pressure control: Divide the pipe network into multiple independent metering areas (DMA), and implement independent pressure management in each area; Flow-pressure control: Dynamically adjust the pressure according to the flow rate change to ensure that the pipe network pressure matches the water demand; Intelligent algorithm control: Use intelligent algorithms (such as the FCM algorithm based on elevation neighborhood information) to adjust the opening degree of the control valve in real time and optimize the pipe network pressure distribution.

[0004] Among them, time-division pressure control is widely adopted as an effective control strategy. Although it can meet the water demand to a certain extent, the existing control strategies usually only consider the day-night variation of water demand, and adjust the pipe network pressure by setting fixed pressure target values in different time periods. And water demand is an important basis for the design and operation of the water supply system, which is affected by various factors, such as seasonal changes, climatic conditions, regional functions, etc. The time-division pressure control strategy in the prior art fails to fully consider the synergistic influence of other various factors on water demand, resulting in low intelligent level and insufficient refinement degree of the control effect. Therefore, an intelligent pipe network pressure control method and system are proposed. Summary of the Invention

[0005] The main object of the present invention is to provide an intelligent pipe network pressure control method and system, which can effectively solve the problems in the background technology.

[0006] To achieve the above object, the technical solution adopted by the present invention is as follows: An intelligent pipe network pressure control method, comprising: Obtaining historical water supply data and corresponding influencing factor data of a water supply pipe network in a target area, wherein the water supply data includes the water consumption in the target area and the water supply pressure of the water supply pipe network, the influencing factor data is data that affects the water consumption, fitting the water supply pressure and the water consumption, and obtaining the functional relationship between the water supply pressure and the water consumption; Performing correlation analysis on the influencing factor data and the water consumption, screening out the types of influencing factors with high correlation with the water consumption, constructing a neural network model with the influencing factor data with high correlation as the input and the corresponding water consumption data as the output, training the neural network model using the obtained data, and adjusting the model parameters of the neural network model until the expected value of its performance index is not lower than a set threshold, to obtain the trained neural network model; Obtaining real-time data of the influencing factors in the target area, using the real-time data of the influencing factors as the input, obtaining the predicted value of the water consumption using the trained neural network model, and substituting the obtained predicted value of the water consumption into the functional relationship to obtain the fitted value of the water supply pressure of the water supply pipe network in the target area; Adjusting the pressure of the water supply pipe network in the target area according to the obtained fitted value of the water supply pressure.

[0007] An intelligent pipe network pressure control system, comprising: A data acquisition module, configured to acquire historical water supply data and corresponding influencing factor data of a water supply pipe network in a target area, wherein the water supply data includes the water consumption in the target area and the water supply pressure of the water supply pipe network, and the influencing factor data is data that affects the water consumption; A data fitting module, configured to fit the water supply pressure and the water consumption, and obtain the functional relationship between the water supply pressure and the water consumption; A data analysis module, configured to perform correlation analysis on the influencing factor data and the water consumption, and screen out the types of influencing factors with high correlation with the water consumption; A neural network module, configured to construct a neural network model with the influencing factor data with high correlation as the input and the corresponding water consumption data as the output, train the neural network model using the obtained data, and adjust the model parameters of the neural network model until the expected value of its performance index is not lower than a set threshold, to obtain the trained neural network model; A real-time data acquisition module for acquiring real-time data of the influencing factors within a target area; A predicted pressure acquisition module for using the real-time data of the influencing factors as input, obtaining a predicted value of the water consumption by means of the trained neural network model, and substituting the obtained predicted value of the water consumption into the functional relationship to obtain a fitting value of the water supply pressure of the water supply network within the target area; A pipe network pressure regulation module for adjusting the water supply network pressure within the target area according to the obtained fitting value of the water supply pressure.

[0008] The system further includes a memory, a processor, and an electronic program stored in the memory and capable of running on the processor.

[0009] Furthermore, the screening process for the types of influencing factors highly correlated with the water consumption includes the following steps: Performing correlation analysis on the influencing factor data and the water consumption to obtain the correlation coefficient between the th influencing factor and the water consumption ; Normalizing the correlation coefficient to obtain the normalized correlation coefficient ; Arranging the normalized correlation coefficients in descending order, and selecting the first correlation coefficients with a sum value not less than 0.85 corresponding influencing factors as the types of influencing factors highly correlated with the water consumption.

[0010] Furthermore, the calculation method of the correlation coefficient is:

[0011] In the formula, is the th sampling value of the th influencing factor; is the th sampling value of the water consumption; is the total amount of sampling data; is the resolution coefficient, taking 0.5; is the two-level minimum difference of the sampling data; is the two-level maximum difference of the sampling data; where ; ; means first taking when = 1, 2, …, When the minimum value among all calculation results, and then when = 1, 2, …, the minimum value among all the minimum values obtained; It means first taking when = 1, 2, …, When the maximum value among all calculation results, and then when = 1, 2, …, the maximum value among all the maximum values obtained; is the type of influencing factor.

[0012] Furthermore, the number of neurons in the input layer of the neural network model is the same as the type of influencing factor having a high correlation with the water consumption, and the number of neurons in the output layer is 1.

[0013] Furthermore, the performance indicators of the neural network model are used to measure the difference between the predicted value and the sampled value of the water consumption, including one or more combinations of the first performance indicator, the second performance indicator, the third performance indicator, and the fourth performance indicator; the calculation methods of the performance indicators are respectively: The first performance indicator = ; The second performance indicator = ; The third performance indicator = ; The fourth performance indicator = ; Among them, is the th sampled value of the water consumption; is the predicted value of the water consumption obtained by using the neural network model; is the total amount of sampled data.

[0014] Furthermore, the calculation method of the expected value of the performance indicators of the neural network model is:

[0015] Among them, represents the expected value of the th performance indicator; represents the th calculation value of the th performance indicator; = 1, 2, 3, 4.

[0016] Further, the specific process of adjusting the pressure of the water supply pipe network in the target area according to the obtained fitting value of the water supply pressure is as follows: Obtain the fitting value of the water supply pressure and the pressure of the water supply pipe network in the current target area ; Determine the adjustment strategy according to the numerical relationship between the fitting value of the water supply pressure and the pressure of the water supply pipe network in the current target area , including: When , the first adjustment strategy; When , the second adjustment strategy; When , the third adjustment strategy; Among them, is the allowable threshold of the pipe network pressure; When adopting the first adjustment strategy, keep the pressure of the water supply pipe network in the current target area unchanged; When adopting the second adjustment strategy, gradually reduce the pressure of the water supply pipe network in the current target area with an adjustment step of λ to ; When adopting the third adjustment strategy, gradually increase the pressure of the water supply pipe network in the current target area with an adjustment step of λ to .

[0017] Further, the adjustment step satisfies the following relationship:

[0018] In the formula, is the maximum adjustment step, ; is to take and the minimum value of.

[0019] The present invention has the following beneficial effects Compared with the prior art, according to the fitting results of the historical water supply data and the corresponding influencing factor data of the water supply network in the target area, this solution obtains the functional relationship between the water supply pressure and the water consumption, conducts a correlation analysis between the influencing factor data and the water consumption, screens out the types of influencing factors with high correlation with the water consumption, constructs a neural network model with the influencing factor data with high correlation as the input and the corresponding water consumption data as the output, uses the trained neural network model to obtain the predicted value of the water consumption, and substitutes the obtained predicted value of the water consumption into the functional relationship to obtain the fitted value of the water supply pressure of the water supply network in the target area. Adjusting the pressure of the water supply network in the target area according to the obtained fitted value of the water supply pressure can fully consider the combined influence of various factors such as seasonal changes, climate conditions, and regional functions on the water demand, thereby improving the intelligent and refined degree of the control effect of the time-sharing pressure control strategy and meeting the intelligent control requirements of the current network pressure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic flow chart of an intelligent pipe network pressure control method of the present invention; Figure 2 is a structural block diagram of an intelligent pipe network pressure control system of the present invention; Figure 3 is a schematic structural diagram of one of the neural networks constructed in the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0021] The following further describes the present invention in conjunction with the specific embodiments. Among them, the drawings are only for illustrative purposes, showing only schematic diagrams, not physical diagrams, and should not be construed as a limitation to the present invention. In order to better illustrate the specific embodiments of the present invention, some components in the drawings will be omitted, enlarged or reduced, and do not represent the dimensions of the actual product.

[0022] The specific implementation process of the technical solution of the present invention includes the following steps: Step 1: Obtain the historical water supply data and the corresponding influencing factor data of the water supply network in the target area. Among them, the water supply data includes the water consumption in the target area and the water supply pressure of the water supply network, and the influencing factor data is the data that affects the water consumption; It should be noted that the water demand is an important basis for the design and operation of the water supply system, and it is affected by various factors. The following are the main influencing factors of the water demand: Time factor Day and night changes: Peak hours: Usually in the morning (6:00 - 8:00) and evening (18:00 - 20:00), the residential water demand is concentrated, mainly for washing, cooking, bathing, etc.

[0023] Low-demand period: During the night (0:00 - 6:00), water demand is relatively low, and the pipe network pressure can be reduced at this time to save energy.

[0024] Seasonal variations: Summer: With high temperatures, the water demand of residents increases, mainly for showers, air-conditioning cooling, irrigation, etc. In addition, the peak tourist season and outdoor activities in summer also increase water demand.

[0025] Winter: With low temperatures, the water demand of residents relatively decreases, but the heating system (such as heaters) may require a certain amount of water.

[0026] Geographical factors Terrain elevation: High-rise buildings or areas with higher terrain require higher water supply pressure to ensure that water can reach the top floors.

[0027] Areas with lower terrain may require less pressure, but it is necessary to prevent pipe network rupture caused by excessive water pressure.

[0028] Regional functions: Residential areas: Water demand is mainly concentrated in domestic water use, such as drinking, washing, and laundry.

[0029] Commercial areas: Water demand is concentrated during business hours, mainly for catering, cleaning, air-conditioning cooling, etc.

[0030] Industrial areas: Water demand may be very large and concentrated in water use time, mainly for production processes, cooling, cleaning, etc.

[0031] Agricultural areas: Water demand is mainly concentrated in the irrigation season and is greatly affected by seasons and weather.

[0032] Climatic factors Air temperature: High-temperature weather will increase the water demand of residents and businesses, mainly for showers, air-conditioning cooling, and irrigation.

[0033] Low-temperature weather may cause water pipes to freeze, and anti-freezing measures need to be taken. At the same time, residents may reduce outdoor water use.

[0034] Rainfall: When rainfall is sufficient, residents may reduce irrigation water, but heavy rain may increase the pressure on the drainage system.

[0035] In arid regions or during dry seasons, water demand may increase, especially for agricultural irrigation and landscape water use.

[0036] Humidity: In areas with high humidity, the water demand of residents and businesses may increase, mainly for air-conditioning dehumidification and cleaning.

[0037] By analyzing the water demand, the types of data to be collected can be determined. For categorical data, data encoding can be used for processing to facilitate subsequent data analysis.

[0038] Step 2: Fit the water supply pressure and water consumption to obtain the functional relationship between them.

[0039] It should be noted that in order to quantitatively analyze the relationship between water demand and water supply pressure, the following methods can be used: In a possible embodiment, a relationship model between water demand and water supply pressure can be established through experiments or actual monitoring data. For example, water consumption and water supply pressure data at different times and in different regions can be collected, and a mathematical model can be established through regression analysis: Q = a + b×P + c×P 2 ; where a, b, and c are regression coefficients, which can be solved by methods such as the least squares method.

[0040] In a possible embodiment, a hydraulic model (such as EPANET) can also be used to simulate and analyze the water supply network. By inputting parameters such as the topological structure, pipe diameter, and user water consumption of the network, the water demand under different pressures can be simulated. The simulation model can consider various factors such as network leakage and user behavior to provide more accurate analysis results.

[0041] In a possible embodiment, machine learning algorithms (such as neural networks, support vector machines, etc.) can also be used to model the relationship between water demand and water supply pressure. Through the training dataset, the algorithm can learn the complex non-linear relationship between water demand and water supply pressure and be used for prediction and optimization control strategies.

[0042] Step 3: Conduct correlation analysis between the influencing factor data and water consumption, and screen out the types of influencing factors that have a high correlation with water consumption; specifically: Conduct correlation analysis between the influencing factor data and water consumption to obtain the correlation coefficient between the th influencing factor and water consumption ; Normalize the correlation coefficient to obtain the normalized correlation coefficient ; Arrange the normalized correlation coefficients in descending order, and select the first correlation coefficients whose sum values are not less than 0.85

[0043] The corresponding influencing factors are used as the types of influencing factors that have a high correlation with water consumption.

[0044] In the formula, is the th sampling value of the th influencing factor; is the th sampling value of the water consumption; is the total amount of sampling data; is the discrimination coefficient, taking 0.5; is the two-level minimum difference of the sampling data; is the two-level maximum difference of the sampling data; where, ; ; means first taking the minimum value among all calculation results when = 1, 2,..., , and then taking the minimum value among all the obtained minimum values when = 1, 2,..., ; ; means first taking the maximum value among all calculation results when = 1, 2,..., , and then taking the maximum value among all the obtained maximum values when = 1, 2,..., ; ; is the type of influencing factors.

[0045] Step 4: Construct a neural network model with the influencing factor data of high correlation as the input and the corresponding water consumption data as the output, train the neural network model using the obtained data, and adjust the model parameters of the neural network model until the expected value of its performance index is not lower than the set threshold to obtain a trained neural network model; Among them, the number of neurons in the input layer of the neural network model is the same as the type of influencing factors with high correlation with the water consumption, and the number of neurons in the output layer is 1.

[0046] The performance index of the neural network model is used to measure the difference between the predicted value and the sampling value of the water consumption, including one or more combinations of the first performance index, the second performance index, the third performance index, and the fourth performance index; the calculation methods of the performance indexes are respectively: The first performance index = ; The second performance index = ; The third performance index = ; Fourth performance index = ; Wherein, is the th sampling value of water consumption; is the predicted value of water consumption obtained by using the neural network model; is the total amount of sampling data.

[0047] The calculation method of the expected value of the performance index of the neural network model is:

[0048] Wherein, represents the expected value of the th performance index; represents the th calculation value of the th performance index; = 1, 2, 3, 4.

[0049] In a possible embodiment, an artificial neural network model can be adopted, wherein the artificial neural network model has a three-layer structure of an input layer, an intermediate layer and an output layer as shown in Figure 3 , and the adjustable model parameter is the number of neurons in the intermediate layer.

[0050] Step 5: Obtain the real-time data of the influencing factors in the target area, use the real-time data of the influencing factors as the input, obtain the predicted value of water consumption by using the trained neural network model, and substitute the obtained predicted value of water consumption into the functional relationship to obtain the fitting value of the water supply pressure of the water supply network in the target area.

[0051] Step 6: Adjust the pressure of the water supply network in the target area according to the obtained fitting value of the water supply pressure. The specific process is as follows: Obtain the fitting value of the water supply pressure and the current water supply network pressure in the target area ; According to the fitting value of the water supply pressure and the current water supply network pressure in the target area to determine the adjustment strategy, including: When the first adjustment strategy; When the second adjustment strategy; When the third adjustment strategy; Wherein, is the allowable threshold of the network pressure; When the first adjustment strategy is adopted, the pressure of the water supply network within the current target area is maintained unchanged; When the second adjustment strategy is adopted, the pressure of the water supply network within the current target area is gradually decreased with an adjustment step of λ to ; When the third adjustment strategy is adopted, the pressure of the water supply network within the current target area is gradually increased with an adjustment step of λ to .

[0052] The adjustment step satisfies the following relationship:

[0053] wherein, is the maximum adjustment step, ; is to take and the minimum value therein.

[0054] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent pipe network pressure control method, characterized in that, Including: Obtain the historical water supply data of the water supply network in the target area and the corresponding influencing factor data. Among them, the water supply data includes the water consumption in the target area and the water supply pressure of the water supply network, and the influencing factor data is the data that affects the water consumption. Fit the water supply pressure and the water consumption to obtain the functional relationship between the water supply pressure and the water consumption; Conduct correlation analysis on the influencing factor data and the water consumption, screen out the types of influencing factors with high correlation with the water consumption, construct a neural network model with the influencing factor data with high correlation as the input and the corresponding water consumption data as the output, train the neural network model using the obtained data, and adjust the model parameters of the neural network model until the expected value of its performance index is not lower than the set threshold to obtain the trained neural network model; Obtain the real-time data of the influencing factors in the target area, use the real-time data of the influencing factors as the input, obtain the predicted value of the water consumption using the trained neural network model, and substitute the obtained predicted value of the water consumption into the functional relationship to obtain the fitted value of the water supply pressure of the water supply network in the target area; Adjust the water supply network pressure in the target area according to the obtained fitted value of the water supply pressure.

2. The intelligent pipe network pressure control method according to claim 1, characterized in that, The screening process for the types of influencing factors with high correlation with the water consumption includes the following steps: Perform a correlation analysis on the influence factor data and the water consumption to obtain the correlation coefficient between the th influence factor and the water consumption ; Normalize the correlation coefficient to obtain the normalized correlation coefficient ; The normalized correlation coefficients are arranged in descending order, and the first correlation coefficients with sum values not less than 0.85 are selected and the corresponding influencing factors are used as the influencing factor types highly correlated with the water consumption.

3. An intelligent pipe network pressure control method according to claim 1, characterized in that Coefficient of correlation The calculation method is as follows: ； In the formula, is the th sampling value of the th influencing factor; is the th sampling value of the water consumption; is the total amount of sampling data; is the resolution coefficient, taking 0.5; is the two - level minimum difference of the sampling data; is the two - level maximum difference of the sampling data; where ； ； It is expressed as first taking, when = 1, 2,..., at that time, the minimum value among all calculation results, and then taking, when = 1, 2,..., at that time, the minimum value among all the obtained minimum values; It is expressed as taking the maximum value among all calculation results when = 1, 2,..., at that time, and then taking the maximum value among all the maximum values obtained when = 1, 2,..., at that time; is the type of influencing factor.

4. According to an intelligent pipe network pressure control method described in claim 1, characterized in that The number of neurons in the input layer of the neural network model is the same as the types of influencing factors with high correlation with the water consumption, and the number of neurons in the output layer is 1.

5. According to an intelligent pipe network pressure control method described in claim 1, characterized in that The performance index of the neural network model is used to measure the difference between the predicted value and the sampled value of the water consumption, including one or a combination of the first performance index, the second performance index, the third performance index, and the fourth performance index; the calculation methods of the performance indexes are respectively: The first performance indicator ; Second performance indicator = ; Third performance indicator= ; Fourth performance indicator = ; Among them, is the th sampling value of water consumption; is the predicted value of water consumption obtained by using the neural network model; is the total amount of sampling data.

6. The intelligent pipe network pressure control method according to claim 5, characterized in that, The calculation method of the expected value of the performance index of the neural network model is: ； Among them, represents the expected value of the th performance indicator; represents the th calculated value of the th performance indicator; = 1, 2, 3, 4.

7. The intelligent pipe network pressure control method according to claim 1, characterized in that The specific process of adjusting the water supply network pressure in the target area according to the obtained fitted value of the water supply pressure is: Obtain the fitting value of the water supply pressure and the pressure of the water supply pipe network in the current target area ; According to the fitting value of the water supply pressure and the pressure of the water supply network in the current target area to determine the adjustment strategy, including: When the first adjustment strategy; When the second adjustment strategy; When the third adjustment strategy at that time; Among them, is the allowed pipe network pressure threshold; When adopting the first adjustment strategy, the pressure of the water supply network within the current target area is maintained unchanged; When adopting the second adjustment strategy, gradually reduce the pressure of the water supply network in the current target area with an adjustment step of λ to ; When adopting the third adjustment strategy, the water supply network pressure within the current target area is gradually increased with an adjustment step size of λ to .

8. The intelligent pipeline network pressure control method according to claim 7, characterized in that The adjustment step size satisfies the following relationship: ； In the formula, is the maximum adjustment step size, ; is to take and the minimum value among them.

9. An intelligent pipeline network pressure control system, characterized in that, Including: A data acquisition module, used to obtain the historical water supply data of the water supply network in the target area and the corresponding influencing factor data. Among them, the water supply data includes the water consumption in the target area and the water supply pressure of the water supply network, and the influencing factor data is the data that affects the water consumption; A data fitting module, used to fit the water supply pressure and the water consumption to obtain the functional relationship between the water supply pressure and the water consumption; A data analysis module, used to conduct correlation analysis on the influencing factor data and the water consumption, and screen out the types of influencing factors with high correlation with the water consumption; A neural network module, configured to build a neural network model with data of influencing factors having high correlation as input and corresponding water consumption data as output, train the neural network model using the acquired data, and adjust the model parameters of the neural network model until the expected value of its performance index is not lower than a set threshold, thereby obtaining the trained neural network model; A real-time data acquisition module, configured to acquire real-time data of the influencing factors in a target area; A predicted pressure acquisition module, configured to use the real-time data of the influencing factors as input, obtain a predicted value of the water consumption using the trained neural network model, and substitute the obtained predicted value of the water consumption into the functional relationship to obtain a fitted value of the water supply pressure of the water supply network in the target area; A pipe network pressure regulation module, configured to adjust the pressure of the water supply network in the target area according to the obtained fitted value of the water supply pressure; 10. An intelligent pipe network pressure control system according to claim 9, characterized in that, The system further includes a memory, a processor, and an electronic program stored in the memory and capable of running on the processor. Wherein, when the processor runs the electronic program, it can implement the steps of an intelligent pipe network pressure control method according to any one of claims 1-8.

Images