Intelligent peak regulation method for secondary water supply pump house

Abstract

The invention relates to an intelligent peak regulation method for a secondary water supply pump room. The method comprises the following steps: sampling water consumption in a set time period at a set time interval to obtain water consumption data; based on the water consumption data, constructing a water consumption regression model including water consumption trend estimation, seasonal component estimation and holiday effect estimation; the water consumption regression model is used for recognizing the water consumption peak, and then a secondary water supply and storage device is controlled to conduct water storage operation. The water supply peak can be effectively adjusted, and the water supply pressure of the waterworks is relieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent water supply systems, and in particular to an intelligent peak-shaving method for a secondary water supply pump room. Background Art

[0002] During peak water usage periods, water plants are prone to insufficient water supply capacity, resulting in water shortages in certain areas and reduced water supply network pressure; during low water usage periods, the water plant's excess water supply capacity cannot be fully utilized, resulting in increased water supply network pressure. Such frequent fluctuations in water pressure will lead to increased urban leakage and pipe burst rates, increased energy consumption of water plants and secondary water supply equipment, and reduced equipment service life.

[0003] In order to address the industry pain points of stabilizing water pressure and ensuring water supply, existing solutions include expanding and building new water plants, adding secondary water supply reservoirs and booster pumps, etc. However, these solutions have some obvious disadvantages. Expanding or building new water plants will put great pressure on finances. Adding secondary water supply reservoirs will increase the residence time of water bodies, easily breed bacteria and microorganisms, produce dirt precipitation, and cause secondary pollution. Therefore, there is a lack of an innovative method that can intelligently regulate the pressure of the water supply network to optimize the operating efficiency of the water supply system and reduce energy consumption and maintenance costs. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an intelligent peak-shaving method for a secondary water supply pump room, which can effectively adjust the water supply peak and alleviate the water supply pressure of a water plant.

[0005] The technical solution adopted by the present invention to solve the technical problem is: to provide an intelligent peak-shaving method for a secondary water supply pump room, characterized in that it includes the following steps:

[0006] The water consumption within a set time period is sampled at set time intervals to obtain water consumption data;

[0007] Based on the water consumption data, a water consumption regression model including water consumption trend estimation, seasonal component estimation, and holiday effect estimation is constructed;

[0008] The water consumption regression model is used to identify water consumption peaks and control the secondary water supply storage device to perform water storage operations.

[0009] Furthermore, based on the water consumption data, a water consumption regression model including water consumption trend estimation, seasonal component estimation, and holiday effect estimation is constructed, including:

[0010] Modeling the water consumption trend as a linear equation, and fitting a water consumption trend estimation function according to the water consumption data;

[0011] Modeling the seasonal component as a Fourier series, and fitting a seasonal component estimation function based on the water consumption data;

[0012] Model the holiday effect as a binary indicator function and obtain the holiday effect estimation function;

[0013] The water consumption regression model is constructed as the sum of the water quantity trend estimation function, the seasonal component estimation function, and the holiday effect estimation function, and Gaussian noise is introduced as a system error.

[0014] Furthermore, the water consumption trend estimation function is:

[0015] g(t)=kt+m

[0016] Where t is time, g(t) is the trend water consumption estimate at time t, k is the estimated growth rate, and m is the intercept estimate.

[0017] Furthermore, the seasonal component estimation function is

[0018]

[0019] Where t is time, s(t) is the seasonal component correction at time t, P is the period length, N is the order of the Fourier series, and a n and b n is the estimated value of Fourier coefficient.

[0020] Furthermore, the holiday effect estimation function is

[0021]

[0022] Where t is time, h(t) is the holiday effect correction at time t, and l i (t) represents the effect of the i-th holiday at time t, and H represents the total number of holidays.

[0023] Furthermore, the Gaussian noise follows a normal distribution N(0,σ 2 ).

[0024] Furthermore, the parameter σ of the Gaussian noise is obtained by maximum likelihood estimation.

[0025] Furthermore, the likelihood function L of the maximum likelihood estimation is:

[0026]

[0027] Among them, t i is the time point, y i is time point t i of water consumption.

[0028] Furthermore, the controlling of the secondary water supply reservoir to perform water storage operation includes:

[0029] Develop a water storage schedule based on the peak water usage;

[0030] The edge control device deployed at the secondary water supply storage device is used to control the opening of the water inlet valve according to the water storage schedule, so that the secondary water supply storage device increases the water storage capacity before the water consumption peak arrives.

[0031] Beneficial Effects

[0032] Due to the adoption of the above-mentioned technical scheme, the present invention has the following advantages and positive effects compared with the prior art: the present invention constructs a water consumption regression model by comprehensively considering water consumption trends, seasonal components, and holiday effects, thereby identifying water consumption peaks and formulating a personalized water storage schedule, and adjusting the water storage volume using edge control based on the water storage schedule, which can alleviate the problems of insufficient water supply during peak water consumption and excessive water supply during low water consumption periods, making the water plant run more smoothly, reducing the pressure on the water plant, and making the total cost more controllable; at the same time, it can also reduce the water pressure fluctuation of the water supply network, stabilize the water plant load, thereby extending the service life of the network and reducing the energy consumption of the water plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall within the scope limited by the appended claims of the application equally.

[0035] An embodiment of the present invention relates to an intelligent peak-shaving method based on machine learning and edge control.

[0036] This embodiment is deployed in a large water-using area that is supplied by a water plant. The water-using area includes residential areas, shopping malls, office buildings, and schools. Figure 1 As shown, the specific steps are as follows:

[0037] 1. Data Collection

[0038] Each secondary water supply tank in the area is connected to the IoT platform to collect the following data:

[0039] a. Water inlet flow: record the amount of water obtained by the secondary water supply tank from the water plant.

[0040] b. Water outflow: record the amount of water supplied by the secondary water supply tank to each user.

[0041] c. Liquid level data: monitor the water level in the secondary water supply tank.

[0042] 2. Data Preprocessing and Feature Extraction

[0043] The collected inlet flow, outlet flow and liquid level data are preprocessed, and the steps include data cleaning, missing value processing and outlier detection.

[0044] Features are extracted from the preprocessed data, including the temporal distribution of water consumption, daily and weekly water consumption patterns, etc.

[0045] 3. Machine Learning Model Training

[0046] The water consumption data of each secondary water supply tank is used to train the machine learning model, analyze the respective water use habits, and obtain the time characteristic data of peak and trough water use periods.

[0047] 4. Comparison of characteristic data and generation of water storage time

[0048] Automatically compare the peak and trough water usage of each secondary water supply tank to generate a personalized water storage schedule for each tank.

[0049] The schedule determines that each secondary water supply tank will store water during the time period outside the peak water supply period of the water plant to ensure sufficient water is replenished before the respective water consumption peak arrives.

[0050] 5. Edge control equipment deployment and automated control

[0051] Edge control devices are deployed at each secondary water supply tank. These devices are connected to the IoT platform to obtain water storage schedules in real time.

[0052] The edge control device controls the opening of the water inlet valve according to the water storage schedule, automatically performs water storage operations, and ensures that a certain amount of water is replenished before the peak water usage arrives.

[0053] Through real-time monitoring and adjustment of edge control equipment, the secondary water supply tank can effectively avoid the peak water supply period of the water plant, thereby alleviating the water supply pressure of the water plant.

[0054] The specific implementation method of step three is as follows:

[0055] 1. Data preparation: Format the hourly water consumption data for the past 30 days into time series data (t i ,y i ), where t i Indicates the time point, y i Indicates the corresponding water consumption.

[0056] 2. Linear trend estimation:

[0057] Assume a linear trend model g(t)=kt+mg(t)=kt+m, where k represents the growth rate and m represents the intercept.

[0058] The least squares method is used to estimate k and m, and the formula is as follows:

[0059]

[0060] in and are the means of time and water consumption, respectively.

[0061] 3. Seasonal component estimation:

[0062] Assume that the seasonal component s(t) can be expressed by Fourier series, the formula is:

[0063]

[0064] Where P represents the period length (e.g., the daily period is 24 hours), N represents the order of the Fourier series, and a n and b n are the Fourier coefficients.

[0065] Estimate the Fourier coefficients a by minimizing the residual sum of squares n and b n .

[0066] 4. Holiday effect estimation:

[0067] Assume that the holiday effect h(t) can be represented by a binary indicator function, the formula is:

[0068]

[0069] Among them l i (t) represents the effect of the i-th holiday at time t, and H represents the total number of holidays.

[0070] Estimate l by fitting the data during the holiday period i The value of (t).

[0071] 5. Error term estimation:

[0072] The error term ∈t represents the noise or unexplained fluctuations of the model and is assumed to follow a normal distribution N(0,σ2).

[0073] The value of σ is estimated by maximizing the log-likelihood function, which is:

[0074]

[0075] 6. Regression model synthesis:

[0076] Combine the components to build the overall model:

[0077] y(t)=g(t)+s(t)+h(t)+∈ t

[0078] Among them, g(t)=kt+mg(t)=kt+m is the linear trend component, s(t) is the seasonal component, h(t) is the holiday effect component, and ∈t is the error term.

[0079] Using the above method, the hourly water consumption data for the past 30 days is calculated and the parameters of each component are automatically estimated to generate a forecast of water consumption.

[0080] In summary, the present invention can effectively capture and predict the long-term changing trends, seasonal fluctuations and holiday effects of water consumption, and provide data support for water supply system optimization and peak-shaving strategies, thereby realizing intelligent peak-shaving, alleviating the water supply pressure of water plants, and optimizing the overall performance of the water supply system.

Claims

1. A secondary water supply pump room intelligent peak regulation method, characterized in that: The following steps are involved: The water consumption within a set time period is sampled at set time intervals to obtain water consumption data; Based on the water consumption data, a water consumption regression model including water consumption trend estimation, seasonal component estimation, and holiday effect estimation is constructed; The water consumption regression model is used to identify water consumption peaks and control the secondary water supply storage device to perform water storage operations.

2. The method according to claim 1, characterized in that The water consumption regression model including water consumption trend estimation, seasonal component estimation, and holiday effect estimation is constructed based on the water consumption data, including: Modeling the water consumption trend as a linear equation, and fitting a water consumption trend estimation function according to the water consumption data; Modeling the seasonal component as a Fourier series, and fitting a seasonal component estimation function based on the water consumption data; Model the holiday effect as a binary indicator function and obtain the holiday effect estimation function; The water consumption regression model is constructed as the sum of the water quantity trend estimation function, the seasonal component estimation function, and the holiday effect estimation function, and Gaussian noise is introduced as a system error.

3. The method according to claim 2, characterized in that The water consumption trend estimation function is: g(t)=kt+m Where t is time, g(t) is the trend water consumption estimate at time t, k is the estimated growth rate, and m is the intercept estimate.

4. The method according to claim 3, characterized in that The seasonal component estimation function is Where t is time, s(t) is the seasonal component correction at time t, P is the period length, N is the order of the Fourier series, and a n and b n is the estimated value of Fourier coefficient.

5. The method according to claim 4, characterized in that The holiday effect estimation function is Where t is time, h(t) is the holiday effect correction at time t, and l i (t) represents the effect of the i-th holiday at time t, and H represents the total number of holidays.

6. The method according to claim 5, characterized in that The Gaussian noise follows the normal distribution N(0,σ 2 ).

7. The method according to claim 6, characterized in that The parameter σ of the Gaussian noise is obtained by maximum likelihood estimation.

8. The method according to claim 7, characterized in that The likelihood function L of the maximum likelihood estimation is: Among them, t i is the time point, y i is time point t i of water consumption.

9. The method according to claim 1, characterized in that: The controlling of the secondary water supply reservoir to perform water storage operation comprises: Develop a water storage schedule based on the peak water usage; The edge control device deployed at the secondary water supply storage device is used to control the opening of the water inlet valve according to the water storage schedule, so that the secondary water supply storage device increases the water storage capacity before the water consumption peak arrives.

Images