Partitioned multi-stage serial pressurized conveying system and balance regulation platform

Abstract

The application discloses a partition type multi-stage serial pressurized conveying system and a balance regulation platform, relates to the technical field of water supply equipment, and comprises a steady flow tank, a basic pressure module and a serial pressurizing module which are sequentially connected with the steady flow pump, wherein the balance regulation platform comprises a pressure data acquisition unit, a working condition analysis unit, a partition adjustment unit and a remote control unit; the application adopts relay type serial operation of multiple water pumps, the water outlet of the previous stage water pump after pressurization is used as the water inlet of the next stage, the water pressure is gradually increased, meanwhile, the first water supply unit of serial water supply bears all water supply flow and terminal water head loss, the water pump of the next stage only bears the actual water supply height difference, the lift of the water pump of the next stage and unnecessary reserved amount are reduced, the power of the water pump of the next stage is reduced, meanwhile, the balance regulation platform is used for monitoring the municipal pipe network pressure in real time, dynamic regulation of the rotating speed is realized, the pipe network pressure is ensured to be always in a non-negative pressure state, and the influence on water use of surrounding users or damage to the pipe network is avoided.

Description

Technical Field

[0001] This invention relates to the field of water supply equipment technology, and in particular to a zoned multi-stage series pressurized delivery system and a balance control platform. Background Technology

[0002] In traditional building secondary pressurized water supply systems, a single pump or parallel pump sets are typically used to supply water to all areas of the building (including low, middle, and high zones). In order to ensure sufficient pressure at the most unfavorable point in the water supply area (usually the highest and farthest user), a high pressure must be set at the pump outlet. However, this pressure is seriously excessive for users who are close to the pump room and on lower floors.

[0003] Therefore, this overpressure energy is entirely converted from electrical energy consumption, but it is meaningless to nearby users and may even damage their water equipment. This energy is ultimately wasted through the pressure reducing valve or faucet in the user's home, making it the biggest energy black hole in the system.

[0004] Meanwhile, traditional zoning (e.g., low-zone, medium-zone, high-zone) is based on physical zoning during building design, and the piping is fixed once laid. However, actual water demand is dynamic; for example, peak commercial water demand in low-zone and peak residential water demand in high-zone may not coincide. Therefore, fixed zoning prevents the system from optimizing its scheduling based on real-time, changing water demand patterns. This can lead to situations where demand in low-zone is minimal, but pumps are still operating at high pressure to meet the needs of a few users, or peak water demand across zones causes a momentary pressure shortage in one zone while pumps in other zones operate at low load and cannot provide coordinated support.

[0005] Therefore, there is an urgent need for a zoned, multi-stage series pressurization and delivery system and a balance control platform that can fundamentally solve the contradiction between "overpressure at the near end and underpressure at the far end", achieve on-demand pressure supply, and improve the system's energy efficiency. Summary of the Invention

[0006] The purpose of this invention is to reduce the head and unnecessary margin of the subsequent pumps by using multiple pumps in a relay-type series operation, thereby reducing the power of the subsequent pumps. At the same time, the municipal pipeline pressure is monitored in real time by a balance control platform, and the pump speed is dynamically adjusted when pumping water to ensure that the pipeline pressure is always in a negative pressure state, thus avoiding affecting the water use of surrounding users or damaging the pipeline.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: a partitioned multi-stage series pressurization and conveying system, comprising a stabilizing tank and a basic pressure module and a series booster module connected in sequence to a stabilizing pump. The top of the stabilizing tank is provided with a water inlet, and the bottom surface of the stabilizing tank is connected to a main water inlet pipe. The output end of the main water inlet pipe is connected to the basic pressure module, and the output end of the basic pressure module is connected to the series booster module.

[0008] The basic pressure module includes a primary water pump. The main inlet pipe is internally connected to several evenly distributed branch pipes. The output end of each branch pipe is connected to the primary water pump through an eccentric reducing hose. A primary butterfly valve is installed between the output ends of the primary water pump and the branch pipes.

[0009] The series booster module includes a secondary water pump and a tertiary water pump. The output end of the primary water pump is connected to a check valve. The check valve and the secondary water pump are connected together to a low-zone pressurized water outlet pipeline. The internal part of the low-zone pressurized water outlet pipeline is connected to a secondary butterfly valve. The output end of the secondary water pump is also connected to a check valve. The check valve and the tertiary water pump are connected together to a middle-zone pressurized water outlet pipeline. The internal part of the middle-zone pressurized water outlet pipeline is connected to a tertiary butterfly valve. The output end of the tertiary water pump is connected to a high-zone pressurized water outlet pipeline.

[0010] Furthermore, the basic pressure module is equipped with three sets of primary water pumps, and the output end of each primary water pump is a low-zone discharge port. The low-zone discharge port is connected to the low-zone pressurized water outlet pipeline, and the low-zone pressurized water outlet pipeline is connected to the low-pressure water supply network.

[0011] Furthermore, the series booster module is equipped with two sets of secondary water pumps and two sets of tertiary water pumps. The output end of each secondary water pump is a middle zone outlet, which is connected to the middle zone pressurized water outlet pipeline, which is connected to the middle zone water supply network. The output end of each tertiary water pump is a high zone outlet, which is connected to the high zone pressurized water outlet pipeline, which is connected to the high pressure water supply network.

[0012] Furthermore, the primary butterfly valve, the primary water pump, and the check valve connected to the primary water pump are all connected to the first frequency converter. The secondary butterfly valve, the secondary water pump, and the check valve connected to the secondary water pump are all connected to the second frequency converter. At the same time, the second frequency converter is also connected to the tertiary butterfly valve, the tertiary water pump, and the check valve connected to the tertiary water pump.

[0013] This invention also provides a balance control platform, comprising a pressure data acquisition unit, a working condition analysis unit, a zoned adjustment unit, and a remote control unit, wherein:

[0014] The pressure data acquisition unit is used to acquire data of the building area where the multi-stage series pressurization and delivery system is located, and divides the low-pressure demand area and the high-pressure demand area. Several low-pressure most unfavorable points are selected in the low-pressure demand area, and several high-pressure most unfavorable points are selected in the high-pressure demand area. The pressure values ​​P1 of the low-pressure most unfavorable points and Ph of the high-pressure most unfavorable points are acquired by pressure sensors and sent to the pressure analysis unit respectively.

[0015] The operating condition analysis unit is used to obtain historical water consumption data within the building area where the multi-stage series pressurized conveying system is located, and to establish a water consumption prediction model based on deep learning algorithms. It also obtains water consumption prediction data within the building area where the multi-stage series pressurized conveying system is located, and sets target pressure values ​​for low-pressure areas and high-pressure areas based on the water consumption prediction data.

[0016] The zoned adjustment unit includes a low-pressure reference state adjustment module and a core operating condition adjustment module. The low-pressure reference state adjustment module is used to acquire a pressure value P1 and compare it with a preset low-pressure zone target pressure value. When the pressure difference is greater than a preset pressure difference threshold, a first control signal is generated and a pressure adjustment value is output to the remote control unit based on a fuzzy PID algorithm.

[0017] The core operating condition adjustment module is used to acquire the pressure value Ph and compare it with the preset target pressure value of the high pressure zone. When the pressure difference is greater than the preset pressure difference threshold, a second control signal is generated and the pressure adjustment value is output to the remote control unit based on the fuzzy PID algorithm.

[0018] The remote control unit is used to acquire and process the first control signal and the second control signal. The first control signal is sent to the first frequency converter to adjust the operating frequency of the basic pressure module so that the pressure value P1 at the most unfavorable point of low pressure is close to the preset target pressure value of low pressure zone. At the same time, the second control signal is sent to the second frequency converter to adjust the operating frequency of the series booster module so that the pressure value Ph at the most unfavorable point of high pressure is close to the preset target pressure value of high pressure zone.

[0019] Furthermore, the specific process for selecting the most unfavorable low-pressure point and the most unfavorable high-pressure point is as follows:

[0020] S101. Obtain the building area data where the multi-stage series pressurized conveying system is located, and divide the low-pressure demand area and high-pressure demand area according to the distribution status of the low-pressure area pipeline network and the high-pressure area pipeline network. Input all pipe diameter, length, material, fittings and elevation information of the low-pressure demand area and the high-pressure demand area into EPANET respectively.

[0021] S102. Setting the operating conditions of the pipeline network operation model based on EPANET simulation: Set the maximum water consumption condition, i.e., the flow rate per second during the morning peak water consumption period calculated according to the specifications;

[0022] S103. The pipeline operation model simulates the pressure loss of water flow in the pipeline network under the flow rate per second during the morning peak water usage period, and outputs the simulation results. The simulation results are displayed in the form of a cloud map, which accurately shows the node with the lowest pressure in the pipeline network, i.e. the most unfavorable point, and the low-pressure most unfavorable point and the high-pressure most unfavorable point are obtained respectively.

[0023] Furthermore, the specific process for obtaining water use forecast data is as follows:

[0024] S201. Obtain historical water usage data within the building area where the multi-stage series pressurized transmission system is located. The historical water usage data includes the pipeline flow rate, terminal pressure, and pump frequency corresponding to the lowest and highest pressure points. Integrate multiple sets of historical water usage data as training samples. Divide the generated training samples into training set, validation set, and test set according to a ratio of 7:2:1. Construct a water usage prediction model based on a deep learning algorithm.

[0025] S202. Download the weight file and load it onto the corresponding network to initialize the migration network parameters;

[0026] S203. Modify the last fully connected layer of the network, keep the input unchanged, set the output to the water use prediction data, initialize the weights of the last layer, use the gradient descent algorithm for learning, and use fixed step size decay to optimize the training parameters. Retrain the entire network to obtain the water use prediction model.

[0027] S204. During the training process, small batches of samples are randomly and non-repeatedly drawn from the training set for training. One training cycle is completed after all training samples are drawn. The training is completed after several cycles to obtain the water use prediction model.

[0028] S205. Deploy the trained water use prediction model in the operating condition analysis unit, and input the real-time water use data collected in real time into the water use prediction model to obtain water use prediction data.

[0029] Furthermore, the specific process for setting target pressure values ​​for low-pressure areas and high-pressure areas based on water usage forecast data is as follows:

[0030] S301. Obtain historical water usage data within the building area where the multi-stage series pressurized conveying system is located, set a time axis with a 24-hour cycle, mark historical water usage data on the time axis and draw water usage fluctuation curves, and delineate water usage fluctuation nodes based on preset curvature values.

[0031] S302. Based on the preset pre-adjustment cycle, define the pre-adjustment node before the water usage fluctuation node, and obtain the water usage data fluctuation value corresponding to the pre-adjustment node. Simultaneously, water use prediction data at water use fluctuation nodes are obtained. After removing the dimensions from the data, the target pressure value is calculated using the following formula. : ,in Base pressure value, This is a preset proportionality coefficient;

[0032] S303. Perform the above calculations for low-pressure demand areas and high-pressure demand areas respectively to obtain the corresponding target pressure values ​​for low-pressure areas and high-pressure areas.

[0033] Furthermore, the specific process of outputting the pressure regulation value based on the fuzzy PID algorithm is as follows:

[0034] S401. Obtain the pressure value P1 and the preset target pressure value Pt for the low-pressure zone, and calculate the pressure difference. ;

[0035] S402. Calculate the rate of change of deviation ec(t): ec(t) = e(t) - e(t-1), where t is the current time. When ec(t) shows an increasing trend, the pressure continues to rise.

[0036] S403. Based on the membership function, define a fuzzy set: eec={NB,NM,NS,ZO,PS,PM,PB}, fuzzify e, and pre-set an expert rule base;

[0037] The expert rule base includes proportional parameter tuning rules, integral parameter tuning rules, and differential parameter tuning rules, among which:

[0038] Proportional parameter tuning rule: The principle behind this rule determines the response speed. When the deviation is large, a large proportional ratio is used to achieve a fast response, and when the deviation is small, a small proportional ratio is used to prevent overshoot.

[0039] Integral parameter tuning rules: The idea behind these rules is to eliminate steady-state errors. When the deviation is large, the integral is suppressed to prevent overshoot; when the deviation is small, the residual error is eliminated.

[0040] Differential parameter tuning rules: The idea behind these rules is to predict future changes. When the deviation is large, braking is applied; when the deviation is small, over-suppression is prevented.

[0041] S404. Substitute the fuzzy set into all the rule bases mentioned above, and calculate the trigger strength of each rule according to the AND logic to obtain a fuzzy output result.

[0042] S405. Calculate the centroid of the fuzzy output shape based on the output result, obtain the abscissa value corresponding to the centroid, and obtain the final pressure adjustment value. The pressure adjustment value includes the proportional adjustment value ΔKp, the integral adjustment value ΔKi, and the derivative adjustment value ΔKd.

[0043] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0044] 1. This zoned multi-stage series pressurized water delivery system uses multiple water pumps in a relay-type series operation. The water pressure is gradually increased by the stage after being pressurized by the previous stage pump. It is suitable for ultra-high-rise buildings or long-distance water delivery. At the same time, the first water supply unit in the series supply bears all the water supply flow and terminal head loss. The subsequent water pumps only bear the actual water supply height difference, which reduces the head of the subsequent water pumps and unnecessary reserve margin, and reduces the power of the subsequent water pumps.

[0045] 2. This balanced control platform, through the first closed loop, ensures that the outlet pressure of the basic pressure module is determined solely by the end-user demand in the low-zone, fundamentally eliminating the significant energy waste caused by pressure-reducing valves. Simultaneously, through the second closed loop, the pressure in the high-zone is independently regulated by the series-connected booster module, solving the problem of unstable pressure in the high-zone caused by fluctuations in water consumption in the low-zone in traditional systems. Furthermore, the two independent end-user pressure closed loops ensure that pressure regulation in the high and low zones does not interfere with each other, enabling real-time adjustment based entirely on the actual pressure feedback from the most unfavorable point. It automatically adapts to changes in water usage patterns at different times and seasons, requiring no manual intervention and always operating at the lowest energy consumption point, achieving on-demand pressure supply and improving the system's energy efficiency. Attached Figure Description

[0046] Figure 1 A schematic diagram of the overall external structure of the present invention is shown;

[0047] Figure 2 This invention is shown as a schematic diagram of its overall external structure from another angle.

[0048] Figure 3 A schematic diagram of the balance control platform structure of the present invention is shown;

[0049] Legend: 1. Stabilizing tank; 2. Inlet; 3. Main inlet pipe; 4. Primary water pump; 5. Branch pipe; 6. High-zone pressurized outlet pipe; 7. Primary butterfly valve; 8. Check valve; 9. Secondary water pump; 10. Low-zone pressurized outlet pipe; 11. Secondary butterfly valve; 12. Medium-zone pressurized outlet pipe; 13. Tertiary butterfly valve; 14. Tertiary water pump. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Example 1: As Figures 1-2 As shown, a partitioned multi-stage series pressurized conveying system includes a flow stabilizing tank 1 and a basic pressure module and a series booster module connected in sequence with a flow stabilizing pump. The top of the flow stabilizing tank 1 is provided with a water inlet 2, and the bottom surface of the flow stabilizing tank 1 is connected with a water inlet main pipe 3. The output end of the water inlet main pipe 3 is connected to the basic pressure module, and the output end of the basic pressure module is connected to the series booster module.

[0052] The basic pressure module includes a primary water pump 4, and the internal connection of the main inlet pipe 3 is a number of evenly distributed branch pipes 5. The output end of each branch pipe 5 is connected to the primary water pump 4 through an eccentric reducing hose. A primary butterfly valve 7 is installed between the output ends of the primary water pump 4 and the branch pipes 5.

[0053] The series booster module includes a secondary water pump 9 and a tertiary water pump 14. The output end of the primary water pump 4 is connected to a check valve 8. The check valve 8 and the secondary water pump 9 are connected together to a low-zone pressurized water outlet pipeline 10. The internal part of the low-zone pressurized water outlet pipeline 10 is connected to a secondary butterfly valve 11. The output end of the secondary water pump 9 is also connected to a check valve 8. The check valve 8 and the tertiary water pump 14 are connected together to a middle-zone pressurized water outlet pipeline 12. The internal part of the middle-zone pressurized water outlet pipeline 12 is connected to a tertiary butterfly valve 13. The output end of the tertiary water pump 14 is connected to a high-zone pressurized water outlet pipeline 6.

[0054] The basic pressure module is equipped with three sets of primary water pumps 4. The output end of each primary water pump 4 is a low zone outlet, which is connected to the low zone pressurized water outlet pipeline 10. The low zone pressurized water outlet pipeline 10 is connected to the low pressure water supply network.

[0055] The series booster module is equipped with two sets of secondary water pumps 9 and two sets of tertiary water pumps 14. The output end of each secondary water pump 9 is the middle zone outlet, which is connected to the middle zone pressurized water outlet pipeline 12. The middle zone pressurized water outlet pipeline 12 is connected to the middle zone water supply network. The output end of each tertiary water pump 14 is the high zone outlet, which is connected to the high zone pressurized water outlet pipeline 6. The high zone pressurized water outlet pipeline 6 is connected to the high pressure water supply network.

[0056] The first-stage butterfly valve 7, the first-stage water pump 4, and the check valve 8 connected to the first-stage water pump 4 are all connected to the first frequency converter. The second-stage butterfly valve 11, the second-stage water pump 9, and the check valve 8 connected to the second-stage water pump 9 are all connected to the second frequency converter. At the same time, the second frequency converter is also connected to the third-stage butterfly valve 13, the third-stage water pump 14, and the check valve 8 connected to the third-stage water pump 14.

[0057] Example 2:

[0058] like Figure 3 As shown, a balance control platform includes a pressure data acquisition unit, a working condition analysis unit, a zoned adjustment unit, and a remote control unit, wherein:

[0059] The pressure data acquisition unit is used to acquire data of the building area where the multi-stage series pressurization and delivery system is located, and divides the low-pressure demand area and the high-pressure demand area. Several low-pressure most unfavorable points are selected in the low-pressure demand area, and several high-pressure most unfavorable points are selected in the high-pressure demand area. The pressure values ​​P1 of the low-pressure most unfavorable points and Ph of the high-pressure most unfavorable points are acquired by pressure sensors and sent to the pressure analysis unit respectively.

[0060] The specific process for selecting the most unfavorable points for low-pressure and high-pressure operation is as follows:

[0061] S101. Obtain the building area data where the multi-stage series pressurized conveying system is located, and divide the low-pressure demand area and high-pressure demand area according to the distribution status of the low-pressure area pipeline network and the high-pressure area pipeline network. Input all pipe diameter, length, material, fittings and elevation information of the low-pressure demand area and the high-pressure demand area into EPANET respectively.

[0062] S102. Setting the operating conditions of the pipeline network operation model based on EPANET simulation: Set the maximum water consumption condition, i.e., the flow rate per second during the morning peak water consumption period calculated according to the specifications;

[0063] S103. The pipeline operation model simulates the pressure loss of water flow in the pipeline network under the flow rate per second during the morning peak water usage period, and outputs the simulation results. The simulation results are displayed in the form of a cloud map, which accurately shows the node with the lowest pressure in the pipeline network, i.e. the most unfavorable point, and the low-pressure most unfavorable point and the high-pressure most unfavorable point are obtained respectively.

[0064] The operating condition analysis unit is used to obtain historical water consumption data within the building area where the multi-stage series pressurized conveying system is located, and to establish a water consumption prediction model based on deep learning algorithms. It also obtains water consumption prediction data within the building area where the multi-stage series pressurized conveying system is located, and sets target pressure values ​​for low-pressure areas and high-pressure areas based on the water consumption prediction data.

[0065] The specific process for obtaining water use forecast data is as follows:

[0066] S201. Obtain historical water usage data within the building area where the multi-stage series pressurized transmission system is located. The historical water usage data includes the pipeline flow rate, terminal pressure, and pump frequency at the lowest and highest pressure points. Integrate multiple sets of historical water usage data as training samples. Divide the generated training samples into training set, validation set, and test set according to a ratio of 7:2:1. Construct a water usage prediction model based on deep learning algorithms.

[0067] S202. Download the weight file and load it onto the corresponding network to initialize the migration network parameters;

[0068] S203. Modify the last fully connected layer of the network, keep the input unchanged, set the output to the water use prediction data, initialize the weights of the last layer, use the gradient descent algorithm for learning, and use fixed step size decay to optimize the training parameters. Retrain the entire network to obtain the water use prediction model.

[0069] S204. During the training process, small batches of samples are randomly and non-repeatedly drawn from the training set for training. One training cycle is completed after all training samples are drawn. The training is completed after several cycles to obtain the water use prediction model.

[0070] S205. Deploy the trained water use prediction model in the operating condition analysis unit, and input the real-time water use data collected in real time into the water use prediction model to obtain water use prediction data.

[0071] The specific process of setting target pressure values ​​for low-pressure areas and high-pressure areas based on water usage forecast data is as follows: S301, obtain historical water usage data within the building area where the multi-stage series pressurization and transmission system is located, set a time axis with a 24-hour cycle, mark historical water usage data on the time axis and draw water usage fluctuation curves, and delineate water usage fluctuation nodes based on preset curvature values.

[0072] S302. Based on the preset pre-adjustment cycle, define the pre-adjustment node before the water usage fluctuation node, and obtain the water usage data fluctuation value corresponding to the pre-adjustment node. Simultaneously, water use prediction data at water use fluctuation nodes are obtained. After removing the dimensions from the data, the target pressure value is calculated using the following formula. : ,in Base pressure value, This is a preset proportionality coefficient;

[0073] S303. Perform the above calculations for low-pressure demand areas and high-pressure demand areas respectively to obtain the corresponding target pressure values ​​for low-pressure areas and high-pressure areas.

[0074] Specifically, the water usage prediction model of the operating condition analysis unit predicts that the pipeline flow rate corresponding to the most unfavorable low-pressure point and the most unfavorable high-pressure point will remain at extremely low values ​​from 2:00 AM to 4:00 AM.

[0075] At 2:00 AM, Pl and Ph will be lowered from their daytime standard values ​​to their nighttime energy-saving values.

[0076] The target pressure value is automatically adjusted back to the standard value at 4:30 a.m., before the morning peak water usage begins.

[0077] At exactly 8:00 AM, the pipeline flow corresponding to the most unfavorable low-pressure point and the most unfavorable high-pressure point will experience a steep increase.

[0078] Before 8:00 AM, adjust the Pl of the basic pressure module from the standard value during the day to the compensation value, that is, to reach the target pressure value;

[0079] At the same time, the series booster module is slowly adjusted to the compensation value, that is, the target pressure value is reached;

[0080] When the peak flow arrives precisely at 8:00 AM, the pipeline pressure P1 will start to drop from a relatively high value and eventually fall back to near the standard value. By raising the target in advance, the impending pressure drop is actively offset, which greatly improves the system pressure stability when water consumption changes drastically and eliminates the instantaneous occurrence of remote underpressure.

[0081] The zoned regulation unit includes a low-pressure reference state regulation module and a core operating condition regulation module. The low-pressure reference state regulation module is used to acquire the pressure value P1 and compare it with the preset target pressure value of the low-pressure zone. When the pressure difference is greater than the preset pressure difference threshold, a first control signal is generated and the pressure regulation value is output to the remote control unit based on the fuzzy PID algorithm.

[0082] The core operating condition adjustment module is used to acquire the pressure value Ph and compare it with the preset target pressure value of the high pressure zone. When the pressure difference is greater than the preset pressure difference threshold, a second control signal is generated and the pressure adjustment value is output to the remote control unit based on the fuzzy PID algorithm.

[0083] The specific process of outputting the pressure regulation value based on the fuzzy PID algorithm is as follows:

[0084] S401. Obtain the pressure value P1 and the preset target pressure value Pt for the low-pressure zone, and calculate the pressure difference. ;

[0085] S402. Calculate the rate of change of deviation ec(t): ec(t) = e(t) - e(t-1), where t is the current time. When ec(t) shows an increasing trend, the pressure continues to rise.

[0086] S403. Based on the membership function, define a fuzzy set: eec={NB,NM,NS,ZO,PS,PM,PB}, fuzzify e, and pre-set an expert rule base;

[0087] The expert rule base includes proportional parameter tuning rules, integral parameter tuning rules, and differential parameter tuning rules, among which:

[0088] Proportional parameter tuning rule: The principle behind this rule determines the response speed. When the deviation is large, a large proportional ratio is used to achieve a fast response, and when the deviation is small, a small proportional ratio is used to prevent overshoot.

[0089] Integral parameter tuning rules: The idea behind these rules is to eliminate steady-state errors. When the deviation is large, the integral is suppressed to prevent overshoot; when the deviation is small, the residual error is eliminated.

[0090] Differential parameter tuning rules: The idea behind these rules is to predict future changes. When the deviation is large, braking is applied; when the deviation is small, over-suppression is prevented.

[0091] S404. Substitute the fuzzy set into all the rule bases mentioned above, and calculate the trigger strength of each rule according to the AND logic to obtain a fuzzy output result.

[0092] S405. Calculate the centroid of the fuzzy output shape based on the output result, obtain the abscissa value corresponding to the centroid, and obtain the final pressure adjustment value. The pressure adjustment value includes the proportional adjustment value ΔKp, the integral adjustment value ΔKi, and the derivative adjustment value ΔKd.

[0093] The remote control unit is used to acquire and process the first control signal and the second control signal. The first control signal is sent to the first frequency converter to adjust the operating frequency of the basic pressure module so that the pressure value P1 at the most unfavorable point of low pressure is close to the preset target pressure value of low pressure zone. At the same time, the second control signal is sent to the second frequency converter to adjust the operating frequency of the series booster module so that the pressure value Ph at the most unfavorable point of high pressure is close to the preset target pressure value of high pressure zone.

[0094] This invention utilizes a relay-style series operation of multiple water pumps. The water pressurized by the previous pump serves as the inlet for the next pump, progressively increasing the water pressure. This is suitable for ultra-high-rise buildings or long-distance water delivery. Simultaneously, the first water supply unit in the series supply bears the entire water flow and terminal head loss, while subsequent pumps only bear the actual water supply height difference. This reduces the head and unnecessary reserve margin of subsequent pumps, thus minimizing their power consumption. Furthermore, a balance control platform monitors the municipal pipeline pressure in real time and dynamically adjusts the pump speed during pumping to ensure that the pipeline pressure is never negative, preventing impact on surrounding users' water supply or damage to the pipeline network.

[0095] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value.

[0096] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0097] In the two embodiments provided in this application, it should be understood that the disclosed apparatus and system can be implemented in other ways; for example, the apparatus embodiments described above are merely illustrative, for example, the division of modules is merely a logical functional division, and there may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed; another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the apparatus or module can be electrical, mechanical or other forms.

[0098] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A balanced regulatory platform, characterized in that, The pressure data acquisition unit, the working condition analysis unit, the partition adjustment unit and the remote control unit are included, wherein: The pressure data acquisition unit is used for acquiring the building area data of the multi-stage serial pressurizing pump group, dividing the low-pressure demand area and the high-pressure demand area, selecting a plurality of low-pressure most unfavorable points in the low-pressure demand area, selecting a plurality of high-pressure most unfavorable points in the high-pressure demand area, acquiring the pressure value Pl of the low-pressure most unfavorable point and the pressure value Ph of the high-pressure most unfavorable point through the pressure sensor, and sending them to the pressure analysis unit respectively; The working condition analysis unit is used for acquiring the historical water consumption data in the building area of the multi-stage serial pressurizing pump group, establishing a water consumption prediction model based on a deep learning algorithm, acquiring water consumption prediction data in the building area of the multi-stage serial pressurizing pump group, and setting the low-pressure area target pressure value and the high-pressure area target pressure value based on the water consumption prediction data; The specific process of obtaining the water consumption prediction data is as follows: S201, acquire the historical water consumption data in the building area of the multi-stage serial pressurizing pump group, the historical water consumption data including the corresponding pipe network flow, end pressure and pump group frequency of the low-pressure most unfavorable point and the high-pressure most unfavorable point, integrate a plurality of sets of historical water consumption data as training samples, divide the generated training samples into a training set, a validation set and a test set according to a ratio of 7:2:1, and construct a water consumption prediction model based on a deep learning algorithm; S202, download the weight file and load it onto the corresponding network for initializing the migration network parameters; S203, modify the last fully connected layer of the network, keep the input unchanged, set the output as the water consumption prediction data, perform weight initialization on the last layer, use the gradient descent algorithm for learning, and use fixed step attenuation to optimize the training parameters, retrain the entire network, and obtain the water consumption prediction model; S204, during the training process, randomly extract small batches of samples from the training set for training, and after all the training samples are extracted, it is a training period, iterate for several cycles to complete the training, and obtain the water consumption prediction model; S205, deploy the trained water consumption prediction model in the working condition analysis unit, input the real-time water consumption data collected in real time into the water consumption prediction model, and obtain the water consumption prediction data; The specific process of setting the low-pressure area target pressure value and the high-pressure area target pressure value based on the water consumption prediction data is as follows: S301, acquire the historical water consumption data in the building area of the multi-stage serial pressurizing pump group, set a time axis with a period of 24 hours, mark the historical water consumption data on the time axis and draw a water consumption fluctuation curve, and divide the water consumption fluctuation nodes based on a preset curvature value; S302, demarcate a pre-adjustment node before a water consumption fluctuation node according to a preset pre-adjustment period, and obtain a water consumption data fluctuation value corresponding to the pre-adjustment node , meanwhile, obtain water consumption prediction data at the water consumption fluctuation node , after dimensionless processing of the data, calculate a target pressure value according to the following formula : , wherein is a basic pressure value, is a preset proportional coefficient S303, the above calculation is performed for the low-pressure demand area and the high-pressure demand area respectively to obtain the corresponding low-pressure area target pressure value and high-pressure area target pressure value; The partition adjustment unit includes a low-pressure reference state adjustment module and a core working condition adjustment module, the low-pressure reference state adjustment module is used for acquiring the pressure value Pl and comparing it with the preset low-pressure area target pressure value, when the pressure difference value is greater than the preset pressure difference threshold value, a first control signal is generated and a pressure adjustment value is output to the remote control unit based on the fuzzy PID algorithm; The core working condition adjusting module is configured to obtain a pressure value Ph and compare the pressure value Ph with a preset target pressure value of a high pressure area, and generate a second control signal and output a pressure adjusting value to the remote control unit based on a fuzzy PID algorithm when a pressure difference value is greater than a preset pressure difference threshold value; The remote control unit is configured to obtain and process the first control signal and the second control signal, and send the first control signal to the first frequency converter to adjust the operating frequency of the basic pressure module so that the pressure value Pl of the low pressure most unfavorable point is close to the preset target pressure value of the low pressure area, and send the second control signal to the second frequency converter to adjust the operating frequency of the series supercharging module so that the pressure value Ph of the high pressure most unfavorable point is close to the preset target pressure value of the high pressure area.

2. The balanced regulatory platform of claim 1, wherein, The specific process of selecting the low pressure most unfavorable point and the high pressure most unfavorable point is as follows: S101, obtain the building area data of the multi-stage series supercharging pump group, and divide the low pressure demand area and the high pressure demand area according to the distribution state of the low pressure area pipe network and the high pressure area pipe network, and input all the pipe diameters, lengths, materials, pipe fittings and elevation information of the low pressure demand area and the high pressure demand area into EPANET respectively; S102, set the working condition based on the pipe network operation model simulated by EPANET: set the design maximum water use condition, that is, the second flow rate at the morning water peak period calculated according to the specification; S103, the pipe network operation model simulates the pressure loss of water flow in the pipe network under the second flow rate at the morning water peak period to output the simulation result, which is in the form of a cloud chart, accurately showing the node with the lowest pressure in the pipe network, that is, the most unfavorable point, and obtaining the low pressure most unfavorable point and the high pressure most unfavorable point respectively.

3. The balanced regulatory platform of claim 1, wherein, The specific process of outputting the pressure adjusting value based on the fuzzy PID algorithm is as follows: S401、acquire the pressure value Pl and the preset low pressure zone target pressure value Pt, and calculate the pressure difference value ; S402, calculate the change rate ec(t) of the deviation: ec(t)=e(t)-e(t-1), where t is the current time, and when ec(t) is in an increasing trend, the pressure continues to rise; S403, define the fuzzy set eec based on the membership function: eec={NB,NM,NS,ZO,PS,PM,PB}, fuzzify e, and pre-set the expert rule base; The expert rule base includes proportional tuning rules, integral tuning rules and differential tuning rules, wherein: The proportional tuning rule: its rule idea determines the response speed, a large deviation requires a large proportion to achieve fast response, and a small deviation requires a small proportion to prevent overshoot; The integral tuning rule: its rule idea is used to eliminate steady-state error, a large deviation requires inhibition of integration to prevent overshoot, and a small deviation is used to eliminate residual error; The differential tuning rule: its rule idea is used to predict future changes, a large deviation requires braking, and a small deviation prevents excessive inhibition; S404, substitute the fuzzy set into all the above rule bases, and calculate the triggering strength of each rule according to the logic to obtain a fuzzy output result; S405, calculate the centroid of the fuzzy output shape based on the output result, obtain the horizontal coordinate value corresponding to the centroid, and obtain the final pressure adjusting value, which includes the proportional adjusting value ΔKp, the integral adjusting value ΔKi and the differential adjusting value ΔKd.

Images