An adaptive water pump variable pressure difference control method, device and system

Abstract

The present application relates to the technical field of water pump control of heating and cooling system, and discloses a self-adaptive water pump variable pressure difference control method, device and system. The method comprises the following steps: obtaining meteorological data in a future preset time, predicting user load of a target area heating and cooling system in the future preset time through a load prediction model; and substituting the user load in the future preset time into a dynamic simulation model to obtain a supply and return water pressure difference change curve of the target area heating and cooling system; and controlling the water pump operation frequency of the target area heating and cooling system based on the supply and return water pressure difference change curve. The water pump variable pressure difference control method disclosed by the present application provides an energy-saving optimization space for the current regional heating and cooling system pressure difference control, reduces the water pump conveying energy consumption under the system part load rate, and saves the operation cost of the system operator.

Description

Technical Field

[0001] This invention belongs to the field of water pump control technology for district heating and cooling systems, specifically relating to an adaptive water pump differential pressure control method, device, and system. Background Technology

[0002] With social development, building energy consumption continues to increase. Under the strategic background of my country's dual-carbon goals, building energy conservation and consumption reduction have become a hot issue. HVAC systems account for a large proportion of total building energy consumption, and within HVAC systems, the distribution system accounts for a large proportion. District cooling and heating systems are air conditioning cold and heat source systems that centrally prepare refrigerant at energy stations and transport it to users in different locations through pipelines. They have advantages such as high energy efficiency, high space utilization, reduced equipment redundancy, and low initial investment. However, due to the large system area, numerous users, and mutual influence among users, the control of its chilled water distribution system is difficult. Currently, the common variable flow control strategies for chilled water systems are generally of two types: temperature difference control and pressure difference control.

[0003] Temperature difference control adjusts the speed and frequency of the circulating water pump by detecting and judging the temperature difference between the supply and return water and the set temperature difference. Theoretically, it has good energy-saving effect, but it has high requirements for the pipeline network. At the same time, because the temperature sensor is a certain distance from the load change location, the control timeliness is poor, and problems such as hydraulic imbalance and failure to meet load requirements can easily occur during the adjustment process.

[0004] Differential pressure control regulates the speed of the circulating water pump by detecting whether the differential pressure control signal fed back from the differential pressure sensor meets the set differential pressure value. The advantage of differential pressure control is its low latency; the water pump can adjust relatively quickly according to changes in flow rate. However, when changes in user load and differential pressure are not synchronized, the differential pressure change cannot accurately represent the load change, limiting energy-saving potential.

[0005] Compared to constant differential pressure control, variable differential pressure control under partial load has better energy-saving potential. Several variable differential pressure control technologies already exist, such as calculating the supply and return water temperature difference based on outdoor temperature sensors, and using the integral of the error between the actual average supply and return water temperature difference and the calculated average temperature difference as the required calculated differential pressure value for the heating network. Alternatively, the load of the terminal equipment is predicted, and the instantaneous required flow rate is calculated based on a certain supply and return water temperature difference; then, based on a database of differential pressure and flow rate when the terminal valves are fully open, the maximum differential pressure value is selected as the real-time control value. However, existing variable differential pressure control technologies still have some shortcomings, such as control delay, insufficient control accuracy, difficulty in use, and limited applicability. Summary of the Invention

[0006] In view of the above-mentioned defects or improvement needs of existing technologies, this invention proposes an adaptive water pump differential pressure control method, which aims to provide an energy-saving differential pressure control scheme for district heating and cooling systems to reduce energy consumption of the transmission and distribution system.

[0007] To achieve the above objectives, in a first aspect of the present invention, an adaptive water pump differential pressure control method is provided, the method comprising:

[0008] Meteorological data for a future preset time period is obtained, and the user load of the heating and cooling system in the target area is predicted in the future preset time period through a load prediction model; the user load in the future preset time period is substituted into a dynamic simulation model to obtain the supply and return water pressure difference change curve of the heating and cooling system in the target area.

[0009] The operating frequency of the water pumps in the target area's heating and cooling system is controlled based on the supply and return water pressure difference curve.

[0010] The load prediction model is trained using historical operating data of the target area's heating and cooling system and corresponding historical meteorological data; the dynamic simulation model is built using system information of the target area's heating and cooling system and the historical operating data.

[0011] As a preferred embodiment of the present invention, the system information includes system architecture information and equipment information, and the historical operating data includes the node temperature, flow rate of each pipe section, pump status, flow rate of each user, and supply and return water temperature during the normal operation of the system within a historical preset time period.

[0012] The dynamic simulation model is constructed using the system information of the target area's heating and cooling system and the historical operating data, specifically as follows:

[0013] The system architecture information and equipment information are used to construct a simulation model framework for the heating and cooling system in the target area; the node temperature, the flow rate of each pipe segment, the pump status, and the flow rate of each user are used to determine the resistance of the heating and cooling pipe network segments and the characteristic parameters of the user valves.

[0014] Based on the aforementioned simulation model framework, the dynamic simulation model is built by inputting the resistance of the hot and cold water pipe network segments, the characteristic parameters of user valves, the flow rate of each user, and the supply and return water temperature data.

[0015] As a preferred embodiment of the present invention, the historical meteorological data and the meteorological data within the future preset time period are respectively the outdoor temperature, relative humidity and solar radiation within the historical time period and the future preset time period; the historical operation data includes the flow rate and supply and return water temperature of each user during the normal operation of the system within the historical preset time period;

[0016] The load forecasting model is trained using historical operating data of the heating and cooling system in the target area and corresponding historical meteorological data. Specifically:

[0017] The user load curve is determined by the flow rate and supply / return water temperature of each user.

[0018] Using the user load curve, historical meteorological data, and corresponding time labels as input, a load prediction model for the target area's heating and cooling system is constructed and trained.

[0019] As a preferred embodiment of the present invention, the training methods for the dynamic simulation model and the load prediction model include online training or offline training.

[0020] As a preferred embodiment of the present invention, before building the dynamic simulation model and the load prediction model, the system information, the historical operating data and the historical meteorological data are preprocessed; wherein the data preprocessing includes outlier correction, missing value imputation and hourly processing.

[0021] As a preferred embodiment of the present invention, the dynamic simulation model is constructed using Flomaster or MATLAB software.

[0022] As a preferred embodiment of the present invention, the step of controlling the pump operating frequency of the target area heating and cooling system based on the supply and return water pressure difference curve specifically involves:

[0023] The water pump is controlled in real time according to the supply and return water pressure difference curve; or,

[0024] The water pump is periodically controlled by varying the pressure difference according to the supply and return water pressure difference curve.

[0025] As a preferred embodiment of the present invention, when the pump is periodically controlled according to the supply and return water pressure difference curve, the interval control is performed with a cycle of 1 to 3 hours.

[0026] In another aspect of the present invention, an adaptive water pump differential pressure control device is provided, comprising:

[0027] The processing module is used to predict the user load of the target area heating and cooling system within a future preset time period through the load prediction model of the target area heating and cooling system, and substitute the user load within the future preset time period into the dynamic simulation model of the target area heating and cooling system to obtain the supply and return water pressure difference change curve of the target area heating and cooling system.

[0028] The execution module is used to control the operating frequency of the water pumps in the target area heating and cooling system based on the supply and return water pressure difference curve.

[0029] In another aspect of the present invention, an adaptive water pump differential pressure control system is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the adaptive water pump differential pressure control method according to any one of the first aspects of the present invention.

[0030] In summary, compared with the prior art, the above-described technical solutions conceived by this invention mainly possess the following technical advantages:

[0031] This invention analyzes historical operating data through modeling to guide differential pressure settings for the system over a future period, providing differential pressure control guidance for district heating and cooling systems and better meeting user needs. Especially for the more complex and critical hydraulic regulation of district heating and cooling systems, the method provided by this invention is easy to operate and does not require the establishment of a large-scale end-point pressure-flow correspondence database. Furthermore, compared to constant pressure regulation methods, the method provided by this invention offers higher accuracy, saves energy and operating costs; compared to sensor-based regulation, it has stronger timeliness, and the control effect can better meet user requirements.

[0032] Furthermore, the dynamic simulation model and load forecasting model are trained offline or online. The system information, historical operational data, historical meteorological data, and historical meteorological data for a predetermined future timeframe mentioned above are periodically updated, and the training of the dynamic simulation model and load forecasting model is periodically synchronized to update these updates, thereby improving the accuracy of the models and forecasts.

[0033] The water pump differential pressure regulation of the present invention is controlled according to the degree of differential pressure change, and can be adjusted in real time or in segments, such as once every 1 hour or once every 3 hours. This allows the water pump differential pressure regulation method of the present invention to be controlled according to the control accuracy and the actual situation of the system, so that the energy consumption and accuracy of water pump differential pressure regulation are balanced, making it more suitable for practical applications. Attached Figure Description

[0034] Figure 1 This is a flowchart illustrating the implementation of an adaptive water pump differential pressure control method according to an embodiment of the present invention.

[0035] Figure 2 This is an example of a district heating and cooling system framework according to an embodiment of the present invention;

[0036] Figure 3 This is an embodiment of the present invention. Figure 2 The diagram showing the fitting relationship between the user resistance coefficient and the total user load in partition A example;

[0037] Figure 4 This is a user load prediction result for a certain day, as exemplified in an embodiment of the present invention.

[0038] Figure 5 This is an embodiment of the present invention. Figure 2 The example shows the calculation results of the hourly control curves and piecewise control curves of the pressure differential in each zone on a certain day; among which... Figure 5 The species 'a' is Figure 2 The example shows the hourly control curves of the pressure difference in each zone on a given day. Figure 5 species b is Figure 2 The example shows the segmented control curves of the differential pressure on a given day for each zone;

[0039] Figure 6 This is an embodiment of the present invention. Figure 2 The power consumption comparison results between the example method and the traditional method are shown in the figure. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0041] In a first aspect, the present invention provides a method for controlling the differential pressure of a water pump based on load prediction. The specific implementation steps of this method are as follows: Figure 1 As shown, the specific embodiments can be described as follows:

[0042] Step 1: Build a dynamic simulation model and a load prediction model for the target area's heating and cooling system. Specifically:

[0043] System information and historical operating data of the heating and cooling system in the target area are collected, and a dynamic simulation model of the heating and cooling system in the target area is built based on the system information and historical operating data. For example, Flomaster or MATLAB software is used to build a dynamic simulation model of the system to fit the operating conditions of the target area system under historical dynamic loads, that is, the ideal supply and return water pressure difference change curve of the target area system during historical operation.

[0044] Historical meteorological data corresponding to the historical operating data is obtained, historical user load is calculated based on the historical operating data, and the historical user load, the historical meteorological data and the corresponding time label are used as input to establish and train the load prediction model of the heating and cooling system in the target area.

[0045] Step 2: Obtain meteorological data for a future preset time period, input the meteorological data and time label for the future preset time period into the load prediction model of the target area heating and cooling system, and predict the user load of the target area heating and cooling system for the future preset time period.

[0046] Step 3: Substitute the user load within the preset future time period into the dynamic simulation model of the target area heating and cooling system, and calculate the supply and return water pressure difference change curve of the target area heating and cooling system under ideal conditions.

[0047] Step 4: Control the operation of the water pumps in the target area heating and cooling system based on the supply and return water pressure difference change curve to achieve adaptive water pump differential pressure control of the target area heating and cooling system.

[0048] In embodiments of the present invention, the system information involved includes system architecture information and equipment information, such as the spatial layout and length of the system pipeline network, and the nameplate parameters and quantity of the equipment. A simulation model framework for the target district heating and cooling system is constructed based on the actual district heating and cooling system topology and equipment.

[0049] This involves historical operational data, such as the temperature of key nodes, flow rate of each pipe section, corresponding pump status, flow rate of each user, and supply and return water temperature during a period of normal system operation. For example, system information and historical operational data of a regional cooling system in a park are collected through an operation and maintenance monitoring system. Based on node temperature, flow rate of each pipe section, pump status, and flow rate data of each user, the resistance of the hot and cold water network sections and the characteristic parameters of user valves in the model are set. The user load curve is calculated based on the flow rate and supply and return water temperature data of each user. In particular, when fitting the resistance of the hot and cold water network sections in the model, since most user terminals are in a closed state when the load is low, the resistance fitting of the hot and cold water network sections only considers the range above 10% of the maximum load.

[0050] Based on the above simulation model framework for the target area heating and cooling system, input the resistance of the heating and cooling pipe network segments, the characteristic parameters of user valves, the flow rate and supply and return water temperature data of each user, and build a dynamic simulation model of the target area heating and cooling system.

[0051] The historical meteorological data and the meteorological data for a preset future time period are outdoor temperature, relative humidity, and solar radiation for the historical time period and the preset future time period, respectively. Historical meteorological parameters, including outdoor temperature, relative humidity, and solar radiation, are collected through small outdoor weather stations. Outdoor temperature, relative humidity, and solar radiation for a future period are obtained from the weather stations. The historical user load curve, historical meteorological data, and corresponding time labels are used as input to construct and train a load prediction model for the target area's heating and cooling system. The heating and cooling loads for the preset future time period are substituted into the dynamic simulation model of the target area's heating and cooling system to obtain the supply and return water pressure difference change curve of the target area's heating and cooling system.

[0052] The load of each user is based on the sum of the total heating and cooling loads of the equipment terminals in each zone within the framework of the heating and cooling system in the target area, and the total heating and cooling loads of the equipment terminals in each zone are obtained through the flow rate and temperature difference of the main pipes in each zone.

[0053] In embodiments of the present invention, the dynamic simulation model and load forecasting model can be used continuously after construction, without needing to be rebuilt every time differential pressure control is performed. Therefore, step 1 is not a necessary step for every differential pressure control operation. Furthermore, the training of the dynamic simulation model and load forecasting model can be offline or online. For example, the system information, historical operating data, historical meteorological data, and historical meteorological data within a preset future timeframe mentioned above are periodically updated, periodically synchronizing the training of the dynamic simulation model and load forecasting model to improve the accuracy of the model and forecasts.

[0054] In embodiments of the present invention, the historical operational data and historical meteorological data involved need to undergo data preprocessing after acquisition to form a usable dataset. Specifically, outlier correction, missing value imputation, and hourly processing can improve the quality of the dataset and the accuracy of the prediction model.

[0055] For example, after data preprocessing, some temperature sensor data is calibrated, and some outlier values ​​are deleted. For instance, historical operational data granularity is 1 minute, and historical meteorological data granularity is 10 minutes; considering actual control needs, it is uniformly processed to 1 hour, ultimately forming a dataset usable for subsequent analysis and modeling. For example, historical data can span several years or several months, and data acquisition granularity can be 1 hour, 10 minutes, or 1 minute; considering actual control accuracy and computational complexity, a granularity of 1 hour is recommended; for raw data granularity within 1 hour, it can be processed to 1 hour as needed. Missing value imputation strategies in data preprocessing can include interpolation imputation combined with recorded values ​​or other methods.

[0056] In embodiments of the present invention, modeling is performed, for example, using Flomaster or MATLAB software. The constructed model is adjusted according to actual control needs, and the model should be able to output the actual hydraulic conditions of the pipeline network under specified user loads and pump operating states.

[0057] The user load set in the dynamic simulation model of the target area heating and cooling system is either an independent user load or a collective load of several related independent users. The user load can be a single air conditioning terminal, or a centralized representation of all users in a building or area. When users are represented centrally, due to the large number of terminals, the overall valve characteristics cannot be directly expressed by theoretical formulas; however, their relationship with the total load can be expressed by formula fitting.

[0058] For example, such as Figure 2 As shown, this project has six user zones, with each pair of zones sharing a common supply and return water header. The model treats each zone as a single integrated user. When setting the model parameters, the user valve resistance characteristics are related to the load magnitude, and this correlation can be expressed using a fitting formula. This can be achieved by first calculating the resistance coefficient K for all operating conditions in the existing data, and then fitting it with the load q. Figure 3 The figure shows the fitting relationship of partition A in this model, where K = 40645q. -0.684 Considering that most user terminals are in a closed state when the load is low, the formula fitting only considers the range of more than 10% of the maximum load. Figure 3 The drag coefficient is calculated in Flomaster software using the following formula to determine the user's pressure loss:

[0059]

[0060] Where ΔP is the pressure loss in Pa, and A is the flow area in m². 2 ρ is the fluid density, in kg / m³ 3 F represents flow rate, in meters (m³). 3 / s; K is the drag coefficient.

[0061] Verified, based on Figure 2 The model built in the case study can fit the hydraulic conditions under different working conditions. When the corresponding user load and pump operation status are set, the simulated hydraulic conditions are consistent with the actual situation.

[0062] For example, in an embodiment of the present invention, the load forecasting model can be selected according to actual conditions, and the load forecasting model constructed above can be used to predict user load, such as using random forest or long short-term memory recurrent neural network to predict user load, inputting meteorological data for a preset future time, and outputting the user load for the corresponding future time.

[0063] In embodiments of the present invention, the scale of user load prediction should be determined according to actual control needs. For situations with multiple user groups / multiple user areas, one or more prediction methods can be selected for training based on actual computational load and prediction accuracy to improve the accuracy of user load prediction.

[0064] For example, such as Figure 4 This provides the forecast results for a specific day across six zones. For example, six load forecasting models were established for users in six zones; the input parameters are ambient temperature, dew point temperature, radiation, working days, hours, and load from one week ago; the output is the user load for a future period.

[0065] In an embodiment of the present invention, the user load within a preset future time period is incorporated into the system dynamic simulation model to calculate the supply and return water pressure difference change curve under ideal conditions. The ideal condition is the minimum pressure difference required to meet the maximum supply and return water temperature difference. For example, if the system is designed with a 5°C temperature difference, then the supply and return water temperature difference should be kept exactly at 5°C as much as possible, and the pressure difference at this time is the minimum pressure difference.

[0066] Ideally, the minimum pressure difference required to meet the maximum supply and return water temperature difference is considered; this example is designed with a 7°C temperature difference. For example, ... Figure 5 'a' in the context is based on Figure 2 The supply and return water pressure difference change curves obtained from the model are hourly control curves. In reality, due to the frequency conversion limitation of the water pump operation, the lower pressure difference set value cannot always be achieved. The actual operating limit of the water pump should be taken as the standard.

[0067] In embodiments of the present invention, in actual control, the water pump differential pressure regulation is controlled according to the degree of differential pressure change, which can be real-time adjustment or segmented adjustment, for example, segmented adjustment once every 1 hour or once every 3 hours. The actual control curve can be fitted into a piecewise function according to the control accuracy and actual system conditions for easy control.

[0068] The differential pressure curve is fitted into a piecewise function according to the actual control accuracy requirements, and the variable differential pressure control of the water pump is carried out based on this differential pressure function.

[0069] The obtained differential pressure curve is fitted into a piecewise function, and the variable differential pressure control of the water pump is performed based on this differential pressure function. For example, Figure 5 b in the text is based on Figure 2 The model yields a supply and return water pressure difference curve that varies every 3 hours, which is a segmented control curve for the pressure difference. The control value for each segment is the hourly maximum value within that time period.

[0070] In embodiments of the present invention, such as Figure 6 As shown, it is based on Figure 2Compared with the traditional method, the segmented differential pressure control of the model can save 69.59% of power consumption compared with the traditional user constant differential pressure control, and 74.04% of power consumption compared with the dry pipe constant differential pressure control.

[0071] According to another aspect of the present invention, an adaptive water pump differential pressure control device is provided, comprising:

[0072] The processing module is used to predict the user load of the target area heating and cooling system within a future preset time period through the load prediction model of the target area heating and cooling system, and substitute the user load within the future preset time period into the dynamic simulation model of the target area heating and cooling system to obtain the supply and return water pressure difference change curve of the target area heating and cooling system.

[0073] The execution module is used to control the operating frequency of the water pumps in the target area heating and cooling system based on the supply and return water pressure difference curve.

[0074] In embodiments of the present invention, the adaptive water pump differential pressure control device further includes:

[0075] The data acquisition module is used to acquire system information, historical operating data and historical meteorological data corresponding to the heating and cooling system in the target area, as well as meteorological data within a preset future time period.

[0076] The module is used to build a dynamic simulation model of the target area heating and cooling system based on the system information and the historical operating data, and to build a load prediction model of the target area heating and cooling system based on the historical operating data and the historical meteorological data.

[0077] According to another aspect of the present invention, an adaptive water pump differential pressure control system is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0078] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0079] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive water pump differential pressure control method, characterized in that, The method includes: Meteorological data for a future preset time period is acquired, and the user load of the heating and cooling system in the target area is predicted for the future preset time period through a load prediction model; the user load for the future preset time period is substituted into a dynamic simulation model to obtain the supply and return water pressure difference change curve of the heating and cooling system in the target area; the operating frequency of the water pumps of the heating and cooling system in the target area is controlled based on the supply and return water pressure difference change curve. The load prediction model is trained using historical operating data of the target area's heating and cooling system and corresponding historical meteorological data. The dynamic simulation model is built using system information of the target area's heating and cooling system and the historical operating data. The system information includes system architecture information and equipment information. The historical operating data includes the node temperature, flow rate of each pipe segment, pump status, flow rate of each user, and supply and return water temperature during the system's normal operation within a preset historical time period. The dynamic simulation model is built using the system information and historical operating data of the target area's heating and cooling system. Specifically, it involves: constructing a simulation model framework for the target area's heating and cooling system using the system architecture information and equipment information; determining the resistance of the heating and cooling pipe network segments and the characteristic parameters of user valves using the node temperature, flow rate of each pipe segment, pump status, and flow rate of each user; and building the dynamic simulation model by inputting the resistance of the heating and cooling pipe network segments, the characteristic parameters of user valves, the flow rate of each user, and the supply and return water temperature data based on the simulation model framework. The historical meteorological data and the meteorological data for the future preset time period are respectively the outdoor temperature, relative humidity, and solar radiation for the historical time period and the future preset time period; the historical operating data includes the flow rate and supply and return water temperature of each user during the normal operation of the system in the historical preset time period; the load prediction model is trained and formed by the historical operating data of the target area heating and cooling system and the historical meteorological data corresponding to the historical operating data, specifically: determining the user load curve through the flow rate and supply and return water temperature of each user; using the user load curve, historical meteorological data, and corresponding time labels as input, constructing and training the load prediction model of the target area heating and cooling system.

2. The adaptive water pump differential pressure control method according to claim 1, characterized in that, The training methods for the dynamic simulation model and the load prediction model include online training or offline training.

3. The adaptive water pump differential pressure control method according to claim 1, characterized in that, Before building the dynamic simulation model and the load forecasting model, the system information, the historical operating data and the historical meteorological data are preprocessed. The data preprocessing described therein includes outlier correction, missing value imputation, and hourly processing.

4. The adaptive water pump differential pressure control method according to claim 1, characterized in that, The dynamic simulation model is built using Flomaster or MATLAB software.

5. The adaptive water pump differential pressure control method according to claim 1, characterized in that, The method of controlling the pump operating frequency of the target area heating and cooling system based on the supply and return water pressure difference curve is as follows: The water pump is controlled in real time according to the supply and return water pressure difference curve; or, The water pump is periodically controlled by varying the pressure difference according to the supply and return water pressure difference curve.

6. The adaptive water pump differential pressure control method according to claim 5, characterized in that, When the pump is periodically controlled according to the supply and return water pressure difference curve, the interval control is carried out with a cycle of 1 to 3 hours.

7. An adaptive water pump differential pressure control device, characterized in that, include: The processing module is used to predict the user load of the target area heating and cooling system within a future preset time through a load prediction model, and substitute the user load within the future preset time into the dynamic simulation model of the target area heating and cooling system to obtain the supply and return water pressure difference change curve of the target area heating and cooling system. The execution module is used to control the operating frequency of the water pumps in the target area heating and cooling system based on the supply and return water pressure difference change curve; The load prediction model is trained using historical operating data of the target area's heating and cooling system and corresponding historical meteorological data. The dynamic simulation model is built using system information of the target area's heating and cooling system and the historical operating data. The system information includes system architecture information and equipment information. The historical operating data includes the node temperature, flow rate of each pipe segment, pump status, flow rate of each user, and supply and return water temperature during the system's normal operation within a preset historical time period. The dynamic simulation model is built using the system information and historical operating data of the target area's heating and cooling system. Specifically, it involves: constructing a simulation model framework for the target area's heating and cooling system using the system architecture information and equipment information; determining the resistance of the heating and cooling pipe network segments and the characteristic parameters of user valves using the node temperature, flow rate of each pipe segment, pump status, and flow rate of each user; and building the dynamic simulation model by inputting the resistance of the heating and cooling pipe network segments, the characteristic parameters of user valves, the flow rate of each user, and the supply and return water temperature data based on the simulation model framework. The historical meteorological data and the meteorological data for the future preset time period are respectively the outdoor temperature, relative humidity, and solar radiation for the historical time period and the future preset time period; the historical operating data includes the flow rate and supply and return water temperature of each user during the normal operation of the system in the historical preset time period; the load prediction model is trained and formed by the historical operating data of the target area heating and cooling system and the historical meteorological data corresponding to the historical operating data, specifically: determining the user load curve through the flow rate and supply and return water temperature of each user; using the user load curve, historical meteorological data, and corresponding time labels as input, constructing and training the load prediction model of the target area heating and cooling system.

8. An adaptive water pump differential pressure control system, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the adaptive water pump differential pressure control method according to any one of claims 1 to 6.

Images