Water supply system and method suitable for high-rise building

Abstract

The invention discloses a water supply system and method suitable for a high-rise building, the water supply system suitable for the high-rise building comprises a data acquisition layer, a processing analysis layer and a decision control layer, a plurality of variable frequency water supply pumps are arranged in the water supply system, and each floor is provided with a corresponding variable frequency water supply pump; the data acquisition layer is connected with a plurality of water volume sensors, and the water volume sensors are used for collecting water volume data of the current high-rise building in real time and summarizing the water volume data to the data acquisition layer; the processing analysis layer is used for processing the water consumption data in real time and generating a water consumption curve; the processing analysis layer stores historical water consumption data, constructs a water consumption prediction model of each floor in combination with the water consumption data and the water consumption curve, and predicts the water consumption demand of each floor; and the decision control layer is used for generating a water supply plan of each floor according to the water consumption prediction model, and adjusting the output power of the variable frequency water supply pump corresponding to each floor according to the water supply plan. Through the intelligent hierarchical control strategy, accurate matching and intelligent control of water supply requirements of all floors of the high-rise building are achieved, and the stability and reliability of the water supply system are improved.

Description

Technical Field

[0001] The present application relates to the technical field of building water supply, and in particular to a water supply system and method applicable to high-rise buildings. Background Art

[0002] Currently, water supply for high-rise buildings mainly relies on fixed-power water supply pumps. Although they can usually meet average water demand, their water supply capacity is often insufficient when faced with special water use conditions, such as in summer or during peak water use periods of the day.

[0003] The existing solution is to add backup pumps when water consumption increases, but these methods are either costly or still difficult to meet demand when water consumption surges. From the above description, it can be seen that the existing water supply system lacks flexibility and cannot adjust the water supply capacity in real time according to actual water consumption, resulting in energy waste or insufficient water supply.

[0004] With the increasing number of high-rise buildings, the traditional water supply system control method can no longer meet the differentiated water demand of each floor. The existing technology usually uses a single water supply pump with a simple control system to achieve water supply for the entire building. This method cannot be accurately adjusted according to the actual water demand of each floor, which easily leads to problems such as low water supply efficiency, high energy consumption and large water pressure fluctuations.

[0005] Therefore, how to accurately meet the water supply demand and achieve high efficiency and energy saving in the water supply process when high-rise buildings face special water use conditions is an urgent problem to be solved in this field. Summary of the invention

[0006] The present application provides a water supply system and method suitable for high-rise buildings, which is suitable for high-rise residential buildings, office buildings and other high-rise buildings. By real-time monitoring and analysis of water use data, the power output of the water supply pump is automatically adjusted to meet the water supply demand at different times, while achieving energy saving and consumption reduction. The purpose is to achieve accurate matching of water supply demand and efficient energy saving through intelligent means.

[0007] In the first aspect, the present application provides a water supply system suitable for high-rise buildings, which adopts the following technical means: A water supply system suitable for high-rise buildings, comprising a data acquisition layer, a processing and analysis layer, and a decision control layer, wherein the water supply system is provided with a plurality of variable frequency water supply pumps, and each floor has a corresponding variable frequency water supply pump; The data collection layer is connected to a plurality of water volume sensors, which are used to collect the current water consumption data of the high-rise building in real time and summarize it to the data collection layer; The processing and analysis layer is used to process the water consumption data in real time and generate a water consumption curve; the processing and analysis layer stores historical water consumption data, and combines the water consumption data and the water consumption curve to build a water consumption prediction model for each floor, and predict the water demand for each floor; The decision control layer is used to generate a water supply plan for each of the floors according to the water consumption prediction model, and to adjust the output power of the variable frequency water supply pump corresponding to each of the floors according to the water supply plan.

[0008] By adopting the above technical solution, through real-time monitoring and analysis, it is ensured that the water supply system builds a water consumption prediction model according to the analysis results, predicts the water consumption and generates a water supply plan, and automatically adjusts the power output of the variable frequency water supply pump according to the prediction results, so as to realize water supply on demand and effectively reduce energy consumption.

[0009] Preferably, the data collection layer integrates a meteorological data interface, a holiday data interface and a resident data interface. The meteorological data interface is used to collect seasonal temperature and humidity change data, the holiday information interface is used to collect current date and holiday information, and the resident data interface is used to collect the number of residents on each floor.

[0010] By adopting the above technical solution, interfaces for multiple influencing factors are integrated in the data collection layer to collect factors that may affect water consumption, which helps to more accurately predict water demand and achieve precise water supply.

[0011] In the second aspect, the present application provides a water supply method applicable to high-rise buildings, which adopts the following technical means: A water supply system suitable for high-rise buildings based on any one of the above items specifically comprises the following steps: Acquire the water consumption data of each floor in the current high-rise building from a water quantity sensor in real time, wherein there are a plurality of water quantity sensors, which are respectively arranged at the water supply nodes on each floor; Acquire historical water consumption data of each of the floors, and generate a water consumption curve by combining the water consumption data with the historical water consumption data; Calling the influencing factor interface to obtain influencing factor information, and establishing a water consumption prediction model based on the historical water consumption data, the water consumption curve and the influencing factor information in combination with a preset prediction algorithm; A water supply plan for the current floor is generated based on the water consumption prediction model. Each floor has a corresponding variable frequency water supply pump, and the power output of the corresponding variable frequency water supply pump is adjusted according to the water supply plan.

[0012] By adopting the above technical solution, it is possible to monitor water consumption in real time, generate a representative water consumption curve based on historical water consumption data, and accurately predict the water consumption of high-rise buildings based on the collected information on influencing factors. The variable frequency water supply pump can be adjusted according to the prediction results to achieve precise water supply, avoid insufficient water supply during peak hours and energy waste during off-peak hours, and achieve the goal of efficient and energy-saving water supply.

[0013] Preferably, the acquiring of the historical water usage data of each floor, combining the water consumption data with the historical water usage data, and generating a water usage curve specifically comprises the following steps: Obtain historical water consumption data for each of the floors, the historical water consumption data including the water consumption at each preset time point, sort and organize the historical water consumption data in chronological order to generate a historical water consumption curve, integrate the water consumption data with the historical water consumption curve to generate a water consumption curve, and update the water consumption curve in real time each time new water consumption data is collected.

[0014] By adopting the above technical solution, new real-time data is continuously collected over time, and the water consumption curve is updated in real time. The water consumption curve can dynamically display the water consumption from the past to the present and even in the future, and can make a certain degree of prediction of future water consumption based on historical trends, which is very beneficial for the accurate prediction of subsequent water consumption.

[0015] Preferably, the calling of the influencing factor interface to obtain the influencing factor information specifically includes the following steps: The influencing factor information includes meteorological data, holiday data and resident data. The meteorological data interface is called to obtain the meteorological data, and the meteorological data includes the external temperature, humidity and precipitation; the holiday data interface is called to obtain the holiday data of the specified time period, and the holiday data includes the holiday name and holiday period; the resident data interface is called to obtain the resident data, and the resident data includes the number of residents on each floor, and the number of residents is sorted into time series data; and the water consumption data, the meteorological data and the resident data are combined for analysis. The meteorological data, the holiday data and the resident data are integrated according to the time dimension to form an influencing factor data set, and the correlation between each factor and water consumption is analyzed using statistical methods to determine which factors have a significant impact on water consumption.

[0016] By adopting the above technical solution, meteorological data, holiday data and resident data are retrieved from the outside to construct a water use prediction model, providing strong support for water resource management and scheduling, and achieving more accurate water use prediction.

[0017] Preferably, the water consumption prediction model is established based on the historical water consumption data, the water consumption curve and the influencing factor information in combination with a preset prediction algorithm, and specifically includes the following steps: Extracting construction features from the historical water consumption data, the construction features including a timestamp and the water consumption data, and constructing a feature vector in combination with the influencing factor data set; A water consumption prediction model is constructed according to a preset time series analysis algorithm, the characteristic vector is used as the input of the water consumption prediction model, and the historical water consumption data is used as a training set to train the preset water consumption prediction model; a predicted water consumption curve is generated according to the water consumption prediction model.

[0018] By adopting the above technical solutions, historical water consumption data and external influencing factors work together on the floor water consumption prediction model, providing strong support for the optimized scheduling and efficient operation of the water supply system.

[0019] Preferably, after the water consumption prediction model is established in combination with the preset prediction algorithm, the following steps are also included: Performing a smoothness test on the predicted water use curve, wherein the smoothness test includes determining whether the predicted water use curve is smooth in time series. If the predicted water use curve does not have abnormal jumps and fluctuations, the smoothness test is deemed to have passed. Performing a consistency check on the predicted water usage curve and the water usage curve, wherein the consistency check includes determining whether the time nodes of the predicted water usage curve and the water usage curve correspond one-to-one, and if the time nodes of the two correspond one-to-one, the consistency check passes; If any one of the smoothness test and the consistency test fails, the water consumption prediction model is re-optimized based on the historical water consumption data until the smoothness test and the consistency test pass.

[0020] By adopting the above technical solution and the above verification logic, it can be ensured that the predicted water consumption curve generated by the predicted water consumption model maintains a certain consistency and continuity with the water consumption curve, thereby improving the rationality and reliability of the prediction results.

[0021] Preferably, generating the water supply plan for the current floor according to the water consumption prediction model specifically includes the following steps: Analyze the height of each of the floors, determine all the floors that can be controlled by each of the variable frequency water supply pumps according to the variable frequency water supply pumps corresponding to each of the floors, and perform weighted averaging on the predicted water consumption curves of all the floors controlled by the same variable frequency water supply pump to obtain a water supply curve for each of the variable frequency water supply pumps; The total water consumption of each floor within a preset period is calculated respectively, and the weight of each predicted water consumption curve is determined according to the total water consumption of the corresponding floor.

[0022] By adopting the above technical solution, a unique water supply curve is generated for each variable frequency water supply pump, which can not only meet the water demand of each floor but also achieve energy saving. It can be continuously adjusted as the water demand changes, and is flexible, adaptable and sustainable.

[0023] Preferably, after obtaining the water supply curve of each variable frequency water supply pump, the method further includes the following steps: Real-time monitoring of the water supply data of the water supply curve corresponding to each of the variable frequency water supply pumps, and dynamic adjustment of the power output of the variable frequency water supply pump; By using a time series analysis method, the water consumption peak and valley periods of the water supply curve corresponding to each variable frequency water supply pump are identified, an average threshold range is set, and the period when the water supply data is higher than the maximum value of the average threshold range is marked as a peak period, and the period when the water supply data is lower than the minimum value of the average threshold range is marked as a valley period; During the peak period of the water supply curve, the output power of the current variable frequency water supply pump is increased, and during the valley period of the water supply curve, the output power of the current variable frequency water supply pump is reduced.

[0024] By adopting the above technical solution, the peak and valley periods of water consumption in the water supply curve can be identified, and the output power of the variable frequency water supply pump can be adjusted according to the peak and valley periods. This can accurately match the water demand in different periods, which can not only meet the water demand in peak periods, but also achieve energy saving in valley periods and reduce equipment wear.

[0025] In summary, this application at least includes the following beneficial effects: 1. This application realizes accurate matching and intelligent control of water supply demand on each floor of high-rise buildings through intelligent stratified control strategy, and improves the stability and reliability of the water supply system.

[0026] 2. This application uses a prediction model to predict water consumption, providing a scientific basis for the formulation of water supply plans and improving the accuracy and foresight of water supply.

[0027] 3. This application realizes precise control of water supply to each floor through intelligent adjustment of variable frequency water pumps, effectively reduces energy consumption and waste of water resources, and brings higher economic and social benefits to high-rise building property managers. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is an architectural diagram of a water supply system applicable to high-rise buildings according to an embodiment of the present application; Figure 2 It is a flow chart of a water supply method applicable to high-rise buildings according to an embodiment of the present application. DETAILED DESCRIPTION

[0029] The present application provides a water supply system and method suitable for high-rise buildings. In order to make the purpose, technical solution and advantages of the present application clearer, the implementation method of the present application will be further described in detail below.

[0030] The following is a further detailed description of an embodiment of a water supply system applicable to high-rise buildings of the present application in conjunction with the drawings of the specification.

[0031] A water supply system suitable for high-rise buildings, comprising a data acquisition layer, a processing and analysis layer, and a decision control layer. The water supply system is provided with a plurality of variable frequency water supply pumps, and each floor has a corresponding variable frequency water supply pump; Generally speaking, each variable frequency water supply pump controls several adjacent floors. In high-rise buildings, higher floors often require greater water pressure. If the same water supply pump is used on lower floors, the water pressure on higher floors will be insufficient. The number of floors associated with a variable frequency water supply pump can be designed according to actual needs. I will not go into details here.

[0032] The data collection layer is connected to multiple water volume sensors, which are used to collect the current water consumption data of high-rise buildings in real time and summarize them to the data collection layer; The data collection layer also integrates meteorological data interface, holiday data interface and resident data interface. The meteorological data interface is used to collect seasonal temperature and humidity change data, the holiday information interface is used to collect current date and holiday information, and the resident data interface is used to collect the number of residents on each floor.

[0033] In other specific implementable methods, interfaces of other factors that can affect water use can be integrated into the data collection layer as required to establish a targeted prediction model for water use prediction.

[0034] The processing and analysis layer is used to process water consumption data in real time and generate water consumption curves. The processing and analysis layer stores historical water consumption data, and combines water consumption data and water consumption curves to build a water consumption prediction model for each floor to predict the water demand of each floor. The decision control layer is used to generate a water supply plan for each floor based on the water consumption prediction model, and to adjust the output power of the variable frequency water supply pump corresponding to each floor according to the water supply plan.

[0035] In another practicable specific solution, a water supply method applicable to high-rise buildings is also disclosed.

[0036] A water supply method applicable to high-rise buildings, based on any one of the above-mentioned water supply systems applicable to high-rise buildings, specifically comprises the following steps: S1. Real-time acquisition of water consumption data of each floor in the current high-rise building from a water quantity sensor. There are multiple water quantity sensors, which are respectively arranged at the water supply nodes on each floor.

[0037] The water supply nodes include areas with high water consumption such as kitchens and bathrooms. In this embodiment, the water supply node is set in the bathroom, and can also be set in multiple places. When the water consumption data is counted, it needs to be totaled.

[0038] S2. Obtain the historical water consumption data of each floor, and generate a water consumption curve by combining the water consumption data with the historical water consumption data, which specifically includes the following steps: S21. Obtain historical water consumption data for each floor. The historical water consumption data includes the water consumption at each preset time point. In this embodiment, statistics are on the water consumption of each floor in the past year. It can also be set to half a year or several months.

[0039] S22. Sort and organize the historical water consumption data in chronological order to generate a historical water consumption curve. This curve chart shows the changing trend and cyclical characteristics of water consumption in the past year.

[0040] S23, integrating the water consumption data with the historical water consumption curve to generate a water consumption curve. The integration process is to add points of the real-time water consumption data to the historical curve chart to reflect the current water consumption situation.

[0041] As time goes by, new real-time data is continuously collected, and the water consumption curve is updated in real time every time new water consumption data is collected. In this way, the water consumption curve can change dynamically, showing the past and present water consumption curves, providing more effective data for subsequent predictions, and the collection time can also be set independently. In this embodiment, the collection is performed once every hour.

[0042] S3. Call the influencing factor interface to obtain influencing factor information, and establish a water consumption prediction model based on historical water consumption data, water consumption curves, and influencing factor information in combination with a preset prediction algorithm, specifically including the following steps: The influencing factor information includes meteorological data, holiday data and resident data.

[0043] S31, calling the meteorological data interface to obtain meteorological data, in this embodiment, specifically obtaining the required meteorological data through the HTTP request GET method, and providing parameters such as geographic location and time range when requesting. Meteorological data includes external temperature, humidity and precipitation, and the obtained meteorological data is cleaned, converted and stored.

[0044] S32, calling the holiday data interface to obtain holiday data for a specified time period. In this embodiment, the public API service of the government open data platform is used to obtain holiday information for a specified time period, usually a year or a month, through an HTTP request. The holiday data includes the name of the holiday and the holiday period. The holiday data is organized into a list format that is easy to query and use, which is convenient for subsequent query and use.

[0045] S33. Call the resident data interface to obtain resident data. The resident data interface is an internal or customized interface. The number of residents on each floor is obtained through HTTP requests. The number of residents is organized into time series data so that it can be combined with water consumption and meteorological data for analysis.

[0046] S34. Integrate meteorological data, holiday data, and resident data according to the time dimension to form a data set containing influencing factors.

[0047] S35. Extract and construct features from historical water use data, where the constructed features include timestamps and water consumption data, and construct feature vectors in combination with the influencing factor data set.

[0048] S36. Construct a water consumption prediction model based on a preset time series analysis algorithm, use the feature vector as the input of the water consumption prediction model, use the historical water consumption data as a training set, and train the preset water consumption prediction model; generate a predicted water consumption curve based on the water consumption prediction model.

[0049] S37. After the water consumption prediction model is generated, the accuracy and reliability of the water consumption prediction model need to be evaluated, which specifically includes the following steps.

[0050] S371. Perform a smoothness test on the predicted water consumption curve. The smoothness test includes determining whether the predicted water consumption curve is smooth in the time series, which reflects the natural law of changes in water consumption.

[0051] If the predicted water usage curve does not show abnormal jumps and fluctuations, the smoothness test is deemed to have passed; otherwise, the smoothness test has failed.

[0052] S372. Perform a consistency check on the predicted water usage curve and the water usage curve. The consistency check includes determining whether the time nodes of the predicted water usage curve and the water usage curve correspond one to one, thereby ensuring the continuity of the time series.

[0053] If the time nodes of the two correspond one to one, the consistency check passes; otherwise, the consistency check fails.

[0054] S38. If any one of the smoothness test and the consistency test fails, the water consumption prediction model is optimized again based on the historical water consumption data. The optimization process requires continuous adjustment of the parameters of the water consumption prediction model until the smoothness test and the consistency test pass.

[0055] In another specific implementation, the water consumption prediction model can also be verified by cross-validation, specifically, the historical water consumption data is divided into K subsets, K-1 subsets are used as training sets each time, and the remaining subset is used as a validation set. This process is repeated K times, and the average error of each validation is calculated.

[0056] If the average error obtained in each verification is relatively stable and does not fluctuate much, then the water consumption prediction model has good stability. Set an error range. If the difference between all average errors exceeds this error range, the model needs to be retrained.

[0057] S4. Generate a water supply plan for the current floor based on the water consumption prediction model. Each floor has a corresponding variable frequency water supply pump. The power output of the corresponding variable frequency water supply pump is adjusted according to the water supply plan. Specifically, the following steps are included: S41. Analyze the height of each floor, determine all the floors that can be controlled by each variable frequency water supply pump based on the variable frequency water supply pump corresponding to each floor, and perform weighted average on the predicted water consumption curves of all floors controlled by the same variable frequency water supply pump to obtain the water supply curve of each variable frequency water supply pump.

[0058] For example, in a specific implementable method, floors 15 to 20 are controlled by the same variable frequency water supply pump. The predicted water usage curves of the six floors from 15 to 20 are weighted averaged to obtain the water supply curve of the variable frequency water supply pump. This water supply curve should be able to reflect the overall changes in water demand from floors 15 to 20.

[0059] S42. Calculate the total water consumption of each floor within a preset period, and determine the weight of each predicted water consumption curve according to the total water consumption of the corresponding floor.

[0060] The preset period refers to a certain period of time in the past. In a specific implementable method, the preset period is one month, that is, the total water consumption of floors 15 to 20 in the past month is calculated. This total water consumption can be obtained based on historical water consumption data, and the respective weights of floors 15 to 20 are allocated according to the total water consumption when performing weighted averaging.

[0061] Specifically, add up the total water consumption of floors 15 to 20 in one month to get the water consumption of all floors in that month. Then the weight of each floor is: floor weight = water consumption of the floor in that month / water consumption of all floors in that month.

[0062] S43, real-time monitoring of the water supply data of the water supply curve corresponding to each variable frequency water supply pump, and dynamically adjusting the power output of the variable frequency water supply pump; Through the time series analysis method, the water consumption peak and valley periods of the water supply curve corresponding to each variable frequency water supply pump are identified, and the average threshold range is set. The period when the water supply data is higher than the maximum value of the average threshold range is marked as the peak period, and the period when the water supply data is lower than the minimum value of the average threshold range is marked as the valley period.

[0063] During the peak period of the water supply curve, the output power of the current variable frequency water supply pump is increased, and during the trough period of the water supply curve, the output power of the current variable frequency water supply pump is reduced.

[0064] In a specific implementation manner, each water supply plan includes the output power and water supply volume of a variable frequency water supply pump in each time period.

[0065] In another specific implementation, when real-time monitoring data shows a sudden increase in water consumption, the water supply system immediately responds and increases the power of the water supply pump to ensure stable water supply.

[0066] If the water consumption continues to be lower than the lowest value of the average threshold range for a period of time, the water supply system will gradually reduce the power of the water supply pump or switch to intermittent water supply mode.

[0067] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A water supply system suitable for high-rise buildings, characterized in that: The water supply system includes a data collection layer, a processing and analysis layer, and a decision control layer. The water supply system is provided with a plurality of variable frequency water supply pumps, and each floor has a corresponding variable frequency water supply pump; The data collection layer is connected to a plurality of water volume sensors, which are used to collect the current water consumption data of the high-rise building in real time and summarize it to the data collection layer; The processing and analysis layer is used to process the water consumption data in real time and generate a water consumption curve; the processing and analysis layer stores historical water consumption data, and combines the water consumption data and the water consumption curve to build a water consumption prediction model for each floor, and predict the water demand for each floor; The decision control layer is used to generate a water supply plan for each of the floors according to the water consumption prediction model, and to adjust the output power of the variable frequency water supply pump corresponding to each of the floors according to the water supply plan.

2. A water supply system suitable for high-rise buildings according to claim 1, characterized in that: The data collection layer integrates a meteorological data interface, a holiday data interface and a resident data interface. The meteorological data interface is used to collect seasonal temperature and humidity change data, the holiday information interface is used to collect current date and holiday information, and the resident data interface is used to collect the number of residents on each floor.

3. A water supply method suitable for high-rise buildings, based on a water supply system suitable for high-rise buildings according to any one of claims 1-2, characterized in that: The specific steps include: Acquire the water consumption data of each floor in the current high-rise building from a water quantity sensor in real time, wherein there are a plurality of water quantity sensors, which are respectively arranged at the water supply nodes on each floor; Acquire historical water consumption data of each of the floors, and generate a water consumption curve by combining the water consumption data with the historical water consumption data; Calling the influencing factor interface to obtain influencing factor information, and establishing a water consumption prediction model based on the historical water consumption data, the water consumption curve and the influencing factor information in combination with a preset prediction algorithm; A water supply plan for the current floor is generated based on the water consumption prediction model. Each floor has a corresponding variable frequency water supply pump, and the power output of the corresponding variable frequency water supply pump is adjusted according to the water supply plan.

4. A water supply method suitable for high-rise buildings according to claim 3, characterized in that: The obtaining of the historical water consumption data of each floor, combining the water consumption data with the historical water consumption data, and generating a water consumption curve specifically includes the following steps: Obtain historical water consumption data for each of the floors, the historical water consumption data including the water consumption at each preset time point, sort and organize the historical water consumption data in chronological order to generate a historical water consumption curve, integrate the water consumption data with the historical water consumption curve to generate a water consumption curve, and update the water consumption curve in real time each time new water consumption data is collected.

5. A water supply method suitable for high-rise buildings according to claim 3, characterized in that: The invoking the influencing factor interface to obtain influencing factor information specifically includes the following steps: The influencing factor information includes meteorological data, holiday data and resident data. The meteorological data interface is called to obtain the meteorological data, and the meteorological data includes the temperature, humidity and precipitation of the outside world; the holiday data interface is called to obtain the holiday data of the specified time period, and the holiday data includes the holiday name and holiday period; the resident data interface is called to obtain the resident data, and the resident data includes the number of residents on each floor, and the number of residents is sorted into time series data; The meteorological data, the holiday data and the resident data are integrated according to the time dimension to form an influencing factor data set.

6. A water supply method suitable for high-rise buildings according to claim 3, characterized in that: The water consumption prediction model is established based on the historical water consumption data, the water consumption curve and the influencing factor information in combination with a preset prediction algorithm, specifically including the following steps: Extracting construction features from the historical water consumption data, the construction features including a timestamp and the water consumption data, and constructing a feature vector in combination with the influencing factor data set; A water consumption prediction model is constructed according to a preset time series analysis algorithm, the characteristic vector is used as the input of the water consumption prediction model, and the historical water consumption data is used as a training set to train the preset water consumption prediction model; a predicted water consumption curve is generated according to the water consumption prediction model.

7. A water supply method suitable for high-rise buildings according to claim 6, characterized in that: After the water consumption prediction model is established by combining the preset prediction algorithm, the following steps are also included: Performing a smoothness test on the predicted water use curve, wherein the smoothness test includes determining whether the predicted water use curve is smooth in time series. If the predicted water use curve does not have abnormal jumps and fluctuations, the smoothness test is deemed to have passed. Performing a consistency check on the predicted water usage curve and the water usage curve, wherein the consistency check includes determining whether the time nodes of the predicted water usage curve and the water usage curve correspond one-to-one, and if the time nodes of the two correspond one-to-one, the consistency check passes; If any one of the smoothness test and the consistency test fails, the water consumption prediction model is re-optimized based on the historical water consumption data until the smoothness test and the consistency test pass.

8. A water supply method suitable for high-rise buildings according to claim 7, characterized in that: Generating the water supply plan for the current floor according to the water consumption prediction model specifically includes the following steps: Analyze the height of each of the floors, determine all the floors that can be controlled by each of the variable frequency water supply pumps according to the variable frequency water supply pumps corresponding to each of the floors, and perform weighted averaging on the predicted water consumption curves of all the floors controlled by the same variable frequency water supply pump to obtain a water supply curve for each of the variable frequency water supply pumps; The total water consumption of each floor within a preset period is calculated respectively, and the weight of each predicted water consumption curve is determined according to the total water consumption of the corresponding floor.

9. A water supply method suitable for high-rise buildings according to claim 8, characterized in that: After obtaining the water supply curve of each variable frequency water supply pump, the following steps are also included: Real-time monitoring of the water supply data of the water supply curve corresponding to each of the variable frequency water supply pumps, and dynamic adjustment of the power output of the variable frequency water supply pump; By using a time series analysis method, the water consumption peak and valley periods of the water supply curve corresponding to each variable frequency water supply pump are identified, an average threshold range is set, and the period when the water supply data is higher than the maximum value of the average threshold range is marked as a peak period, and the period when the water supply data is lower than the minimum value of the average threshold range is marked as a valley period; During the peak period of the water supply curve, the output power of the current variable frequency water supply pump is increased, and during the valley period of the water supply curve, the output power of the current variable frequency water supply pump is reduced.

Images