Joint optimization analysis method and system based on multiple water pumps and storage medium

Abstract

The invention discloses a joint optimization analysis method and system based on multiple water pumps and a storage medium, and the method comprises the steps: obtaining the operation frequency, shaft power and lift data of a target water pump group during stable operation through a flow self-inspection module, and calculating the total water flow based on the operation frequency, shaft power and lift data; through a lift self-checking module, the lift required by the system is calculated based on the total water flow, the operation frequency and the shaft power, and the required lift is obtained; and through the optimization control module, when the lift and the water flow of system operation are both required values of the system, it is judged that the system is in a stable state, and when the system is in the stable state, an optimal optimization configuration scheme of the current water pump set is generated based on the optimization control module.

Description

Technical Field

[0001] This invention relates to the field of water pump monitoring data analysis, and more specifically, to a method, system, and storage medium for joint optimization analysis based on multiple water pumps. Background Technology

[0002] Multi-pump systems are widely used in many industrial and civil sectors, such as water supply, drainage, and fire fighting. However, traditional multi-pump control methods often lack intelligence and energy efficiency, leading to energy waste and low pump efficiency.

[0003] In existing technologies, energy-saving control of water pumps mainly relies on values ​​such as temperature, pressure, temperature difference, or pressure difference as control benchmarks to perform PID control on the pump frequency, thereby achieving energy-saving effects. However, this primarily involves the individual control of a single pump, and the energy-saving effect is based solely on the frequency conversion control of that single pump. It fails to leverage the optimized control of multiple pumps in parallel within the entire pump group, resulting in limited energy-saving potential. Furthermore, it cannot directly determine the current water flow rate of the system, relying only on auxiliary equipment such as flow meters for measurement, which cannot accurately guarantee whether the system's operating flow rate matches the actual required flow rate. Additionally, existing technologies lack effective overall early warning systems for multi-pump systems, hindering the application of water pumps in various scenarios. Summary of the Invention

[0004] This invention overcomes the shortcomings of the prior art and proposes a joint optimization analysis method, system and storage medium based on multiple water pumps.

[0005] The first aspect of this invention provides a joint optimization method based on multiple water pumps, comprising:

[0006] The self-test module obtains the operating frequency, shaft power, and head data of the target pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data.

[0007] The head self-test module calculates the required head of the system based on the total water flow, operating frequency, and shaft power to obtain the required head.

[0008] By optimizing the control module, when the system's operating head and water flow rate are both the system's required values, it is determined to be in a stable state. In the stable state, the parameters of the target pump group are obtained, including: the number of pumps in parallel is n, the required head of the system is H, the required flow rate is Q, the pump operating frequency is f, and the pump shaft power is P.

[0009] When the number of pumps in the target pump group is 1, the frequency of a single pump is f1 if it meets the head H and Q. Based on Q and f1, and combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency, the shaft power of the pump at this time is calculated to be P1.

[0010] When the number of pumps in the target pump group is 2, the frequency of each pump is f2 when the head H and Q / 2 are met. Based on Q / 2 and f2, combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of the single pump is P2 and the total shaft power of the pump is 2*P2.

[0011] When the number of pumps in the target pump set is n, the frequency of a single pump is fn when the head H and Q / n are met. Based on Q / n and fn, and combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of a single pump is Pn and the total shaft power of the pump is n*Pn.

[0012] Based on the above calculation results, the values ​​of P1, 2*P2, ..., n*Pn are sorted to select the data with the smallest value and marked as k*Pk. The optimal number of pumps in operation is k, and the optimal total shaft power of the pumps is k*Pk.

[0013] Based on the optimal number of operating water pumps and the optimal total shaft power of water pumps, an optimized configuration scheme is generated.

[0014] A prediction model based on LSTM is constructed. The model is trained using historical monitoring data of the target pump group. The prediction model is then used to make predictions based on the monitoring data of the target pump group in the current period, so as to obtain the predicted pump monitoring data for the next period.

[0015] Each pump is evaluated based on the health operation data of the target pump group, and a monitoring and early warning range is generated. The early warning pump information is generated by combining the predicted pump monitoring data. Based on the early warning pump information and the optimized control module, multiple early warning configuration schemes for the next cycle are generated.

[0016] In this solution, the flow self-test module acquires the operating frequency, shaft power, and head data of the target pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data. Specifically:

[0017] When the target pump set is running stably, the total water flow of the current water system is calculated based on the operating frequency and shaft power of the target pump set, combined with the pump flow-shaft power curve or the mathematical formula of the curve at the current frequency.

[0018] Based on the total water flow, the real-time head of the current system is calculated according to the pump flow-head curve or the mathematical formula of the curve at the current frequency.

[0019] In this solution, the step of acquiring the operating frequency, shaft power, and head data of the target pump set during stable operation through the flow self-test module, and calculating the total water flow based on the operating frequency, shaft power, and head data, further includes:

[0020] When the target pump set is running stably, the head data is obtained based on the real-time pressure data of the target pump set. Combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency of the target pump set, the total water flow of the current water system is calculated.

[0021] In this solution, the required head is calculated by the head self-test module based on the total water flow, operating frequency, and shaft power. Specifically, the required head is obtained as follows:

[0022] By using preset flow meters and preset pump sensor control reference points, and through PID control, when the actual flow rate of the target pump group reaches the system's required flow rate, the required head of the water system is calculated based on the pump flow-head curve or the mathematical formula of the curve at the current operating frequency.

[0023] In this solution, the step of calculating the required head of the system based on the total water flow, operating frequency, and shaft power using a head self-checking module to obtain the required head also includes:

[0024] If the required water flow rate and the shaft power of the pump are known, the current frequency of the target pump set can be calculated based on the pump flow rate-shaft power curve or the mathematical formula of the curve at different operating frequencies. Then, the required head of the water system can be calculated by combining the pump flow rate-head curve or the mathematical formula of the curve at the current frequency.

[0025] In this scheme, the construction of an LSTM-based prediction model involves training the model using historical monitoring data of the target pump group, and then using the prediction model to predict the pump monitoring data for the next period by comparing it with the current period's monitoring data of the target pump group. This process includes:

[0026] Construct a prediction model based on LSTM;

[0027] Obtain historical monitoring data of the target water pump set, including water pressure, power, and flow rate data;

[0028] Based on the time dimension, historical monitoring data is serialized to form a time series dataset;

[0029] The time series dataset is divided into a training set and a test set according to a preset ratio. The training set and the test set are then imported into the prediction model for training, and the trained prediction model is obtained.

[0030] In this scheme, the prediction of the next cycle's predicted pump monitoring data is obtained by using a prediction model and the monitoring data of the target pump group in the current cycle. Specifically:

[0031] Acquire the monitoring data of the target pump group in the current cycle and mark it as the current monitoring data;

[0032] Based on the time dimension, the current monitoring data is serialized and imported into the prediction model for data prediction, generating the predicted pump monitoring data for the next cycle.

[0033] In this scheme, each pump is evaluated based on the health operation data of the target pump group, and a monitoring and early warning range is generated. Early warning pump information is generated by combining the predicted pump monitoring data. Based on the early warning pump information and the optimized control module, multiple early warning configuration schemes for the next cycle are generated, specifically as follows:

[0034] Acquire the health operation data of the target water pump set, including the water pump set's operating time, equipment life assessment, and maintenance record data;

[0035] Based on the health operation data, the monitoring operation parameter range of each pump in the pump group is evaluated, and the monitoring and early warning range information of the pump group is generated.

[0036] Taking one pump in the target pump group as the unit of analysis, the time-dimensional predictive monitoring numerical fluctuation analysis of the pump is carried out based on the predicted pump monitoring data to obtain the monitoring fluctuation curve and the duration of extreme values.

[0037] Based on the monitoring fluctuation curve, extreme value duration and monitoring warning range information, a warning analysis is performed, and water pumps exceeding the warning range are marked as warning water pumps, thus forming warning water pump information;

[0038] Let the number of warning pumps be N. The target pump group is combined into multiple groups of simulated pump groups. Each simulated pump group includes all non-warning pumps and 0 to N warning pumps. The optimization control module configures the multiple simulated pump groups in the optimal way to generate multiple warning configuration schemes.

[0039] A second aspect of the present invention also provides a joint optimization system based on multiple water pumps, the system comprising: a memory and a processor, wherein the memory includes a joint optimization program based on multiple water pumps, and the joint optimization program based on multiple water pumps, when executed by the processor, performs the following steps:

[0040] The self-test module obtains the operating frequency, shaft power, and head data of the target pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data.

[0041] The head self-test module calculates the required head of the system based on the total water flow, operating frequency, and shaft power to obtain the required head.

[0042] By optimizing the control module, when the system's operating head and water flow rate are both the system's required values, it is determined to be in a stable state. In the stable state, the parameters of the target pump group are obtained, including: the number of pumps in parallel is n, the required head of the system is H, the required flow rate is Q, the pump operating frequency is f, and the pump shaft power is P.

[0043] When the number of pumps in the target pump group is 1, the frequency of a single pump is f1 if it meets the head H and Q. Based on Q and f1, and combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency, the shaft power of the pump at this time is calculated to be P1.

[0044] When the number of pumps in the target pump group is 2, the frequency of each pump is f2 when the head H and Q / 2 are met. Based on Q / 2 and f2, combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of the single pump is P2 and the total shaft power of the pump is 2*P2.

[0045] When the number of pumps in the target pump set is n, the frequency of a single pump is fn when the head H and Q / n are met. Based on Q / n and fn, and combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of a single pump is Pn and the total shaft power of the pump is n*Pn.

[0046] Based on the above calculation results, the values ​​of P1, 2*P2, ..., n*Pn are sorted to select the data with the smallest value and marked as k*Pk. The optimal number of pumps in operation is k, and the optimal total shaft power of the pumps is k*Pk.

[0047] Based on the optimal number of operating water pumps and the optimal total shaft power of water pumps, an optimized configuration scheme is generated.

[0048] A prediction model based on LSTM is constructed. The model is trained using historical monitoring data of the target pump group. The prediction model is then used to make predictions based on the monitoring data of the target pump group in the current period, so as to obtain the predicted pump monitoring data for the next period.

[0049] Each pump is evaluated based on the health operation data of the target pump group, and a monitoring and early warning range is generated. The early warning pump information is generated by combining the predicted pump monitoring data. Based on the early warning pump information and the optimized control module, multiple early warning configuration schemes for the next cycle are generated.

[0050] A third aspect of the present invention also provides a computer-readable storage medium comprising a multi-pump-based joint optimization program, wherein when executed by a processor, the multi-pump-based joint optimization program implements the steps of the multi-pump-based joint optimization method as described in any of the preceding claims.

[0051] This invention discloses a joint optimization analysis method, system, and storage medium based on multiple water pumps. Through a flow self-checking module, it acquires the operating frequency, shaft power, and head data of the target water pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data. Through a head self-checking module, it calculates the required head of the system based on the total water flow, operating frequency, and shaft power, obtaining the required head. Through an optimization control module, when both the system's operating head and water flow are the required values, it is determined to be in a stable state. In the stable state, the optimization control module generates the optimal configuration scheme for the current water pump set. Attached Figure Description

[0052] Figure 1 A schematic diagram of a module of a multi-pump joint optimization method according to the present invention is shown;

[0053] Figure 2 A block diagram of a multi-pump joint optimization system according to the present invention is shown. Detailed Implementation

[0054] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0055] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0056] Figure 1 A schematic diagram of a module of a multi-pump joint optimization method according to the present invention is shown.

[0057] like Figure 1 As shown, the first aspect of the present invention provides a joint optimization method based on multiple water pumps, comprising:

[0058] The self-test module obtains the operating frequency, shaft power, and head data of the target pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data.

[0059] The head self-test module calculates the required head of the system based on the total water flow, operating frequency, and shaft power to obtain the required head.

[0060] By optimizing the control module, when the system's operating head and water flow rate are both the system's required values, it is determined to be in a stable state. In the stable state, the parameters of the target pump group are obtained, including: the number of pumps in parallel is n, the required head of the system is H, the required flow rate is Q, the pump operating frequency is f, and the pump shaft power is P.

[0061] When the number of pumps in the target pump group is 1, the frequency of a single pump is f1 if it meets the head H and Q. Based on Q and f1, and combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency, the shaft power of the pump at this time is calculated to be P1.

[0062] When the number of pumps in the target pump group is 2, if a single pump meets the requirements of head H and Q / 2 (in the case of pumps in parallel, the head is the same and the flow rate is superimposed, simplifying the calculation, at this time all pumps are treated as pumps with the same parameters), the frequency is f2. Based on Q / 2 and f2, combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of a single pump is calculated to be P2, and the total shaft power of the pump is 2*P2.

[0063] When the number of pumps in the target pump group is n, if a single pump meets the requirements of head H and Q / n (in the case of pumps in parallel, the head is the same and the flow rate is superimposed, simplifying the calculation, at this time all pumps are treated as pumps with the same parameters), the frequency is fn. Based on Q / n and fn, combined with the pump flow rate-head curve or the mathematical formula of the curve at the current frequency, the shaft power of a single pump is calculated to be Pn, and the total shaft power of the pump is n*Pn.

[0064] Based on the above calculation results, the values ​​of P1, 2*P2, ..., n*Pn are sorted to select the data with the smallest value and marked as k*Pk. The optimal number of pumps in operation is k, and the optimal total shaft power of the pumps is k*Pk.

[0065] Based on the optimal number of operating water pumps and the optimal total shaft power of water pumps, an optimized configuration scheme is generated.

[0066] A prediction model based on LSTM is constructed. The model is trained using historical monitoring data of the target pump group. The prediction model is then used to make predictions based on the monitoring data of the target pump group in the current period, so as to obtain the predicted pump monitoring data for the next period.

[0067] Each pump is evaluated based on the health operation data of the target pump group, and a monitoring and early warning range is generated. The early warning pump information is generated by combining the predicted pump monitoring data. Based on the early warning pump information and the optimized control module, multiple early warning configuration schemes for the next cycle are generated.

[0068] It should be noted that in the optimal number of operating pumps (k) and the optimal total shaft power (k*Pk), when the water system is operating stably according to its needs, the current head and flow rate of the water system are the system's required values. The actual dynamic optimization control of the pump set must meet the current system's required flow rate and head, that is, to minimize the power consumption of the pump set (i.e., minimize the total shaft power) while meeting the current system's required flow rate and head. The method of this invention can dynamically calculate the scheme with the lowest power consumption. The target pump set is considered to be in a stable operating state after the equipment has been running for a period of time. The required head (H) can be obtained by calculating the required head using this invention, and the required flow rate (Q) is a preset value, determined according to actual needs. The optimized configuration scheme includes information such as the selected number of operating pumps and the set operating frequency.

[0069] According to an embodiment of the present invention, the step of acquiring the operating frequency, shaft power, and head data of the target water pump set during stable operation through the flow self-test module, and calculating the total water flow based on the operating frequency, shaft power, and head data, specifically involves:

[0070] When the target pump set is running stably, the total water flow of the current water system is calculated based on the operating frequency and shaft power of the target pump set, combined with the pump flow-shaft power curve or the mathematical formula of the curve at the current frequency.

[0071] Based on the total water flow, the real-time head of the current system is calculated according to the pump flow-head curve or the mathematical formula of the curve at the current frequency.

[0072] It should be noted that the flow self-checking module is used to calculate the total water flow data in real time and to calculate whether the total water flow meets the current required flow rate Q, where Q is a preset value. In the mathematical formula of the water pump flow-shaft power curve or curve, the mathematical formula of the curve is the mathematical expression of the water pump flow-shaft power curve, and the two calculation methods are consistent. In the mathematical formula of the curve, each relationship curve appearing in this embodiment has a corresponding mathematical expression.

[0073] According to an embodiment of the present invention, the step of acquiring the operating frequency, shaft power, and head data of the target water pump set during stable operation through the flow self-test module, and calculating the total water flow based on the operating frequency, shaft power, and head data, further includes:

[0074] When the target pump set is running stably, the head data is obtained based on the real-time pressure data of the target pump set (the pressure difference before and after the pump is the head of the current system). Combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency of the target pump set, the total water flow of the current water system is calculated.

[0075] It should be noted that in this embodiment, the calculation of the total water flow includes the two methods mentioned above, and the specific method to be used depends on the actual data acquired and the application.

[0076] According to an embodiment of the present invention, the step of calculating the required head of the system based on the total water flow, operating frequency, and shaft power using the head self-test module to obtain the required head is specifically as follows:

[0077] By using preset flow meters and preset pump sensor control reference points, and through PID control, when the actual flow rate of the target pump group reaches the system's required flow rate, the required head of the water system is calculated based on the pump flow-head curve or the mathematical formula of the curve at the current operating frequency.

[0078] It should be noted that the water system refers to the entire target water pump set.

[0079] According to an embodiment of the present invention, the step of calculating the required head of the system based on the total water flow, operating frequency, and shaft power using a head self-test module to obtain the required head further includes:

[0080] If the required water flow rate and the shaft power of the pump are known, the current frequency of the target pump set can be calculated based on the pump flow rate-shaft power curve or the mathematical formula of the curve at different operating frequencies. Then, the required head of the water system can be calculated by combining the pump flow rate-head curve or the mathematical formula of the curve at the current frequency.

[0081] It should be noted that there are two methods for calculating the required head in this embodiment, and the specific method to be used depends on the actual situation.

[0082] According to an embodiment of the present invention, the step of constructing an LSTM-based prediction model, training the model using historical monitoring data of the target pump group, and predicting the next period's pump monitoring data by comparing the prediction model with the current period's monitoring data of the target pump group, includes:

[0083] Construct a prediction model based on LSTM;

[0084] Obtain historical monitoring data of the target water pump set, including water pressure, power, and flow rate data;

[0085] Based on the time dimension, historical monitoring data is serialized to form a time series dataset;

[0086] The time series dataset is divided into a training set and a test set according to a preset ratio. The training set and the test set are then imported into the prediction model for training, and the trained prediction model is obtained.

[0087] It should be noted that the historical monitoring data refers to monitoring data from a recent period, which is representative and can effectively reflect the current operating status of the water pump set, and is used as training data.

[0088] According to an embodiment of the present invention, the step of predicting the pump monitoring data for the next period by using a prediction model and the monitoring data of the target pump group in the current period specifically involves:

[0089] Acquire the monitoring data of the target pump group in the current cycle and mark it as the current monitoring data;

[0090] Based on the time dimension, the current monitoring data is serialized and imported into the prediction model for data prediction, generating the predicted pump monitoring data for the next cycle.

[0091] It should be noted that the current monitoring data includes the pressure, power, flow rate, temperature, vibration, and current data of the water pump set. These monitoring data often exhibit periodic changes over time and have a certain degree of predictability. This invention uses an LSTM model for data prediction, effectively performing predictive analysis on the data.

[0092] According to an embodiment of the present invention, the evaluation of each pump based on the health operation data of the target pump group and the generation of a monitoring and early warning range, the generation of early warning pump information by combining the predicted pump monitoring data, and the generation of multiple early warning configuration schemes for the next cycle based on the early warning pump information and the optimized control module, specifically:

[0093] Acquire the health operation data of the target water pump set, including the water pump set's operating time, equipment life assessment, and maintenance record data;

[0094] Based on the health operation data, the monitoring operation parameter range of each pump in the pump group is evaluated, and the monitoring and early warning range information of the pump group is generated.

[0095] Taking one pump in the target pump group as the unit of analysis, the time-dimensional predictive monitoring numerical fluctuation analysis of the pump is carried out based on the predicted pump monitoring data to obtain the monitoring fluctuation curve and the duration of extreme values.

[0096] Based on the monitoring fluctuation curve, extreme value duration and monitoring warning range information, a warning analysis is performed, and water pumps exceeding the warning range are marked as warning water pumps, thus forming warning water pump information;

[0097] Let the number of warning pumps be N. The target pump group is combined into multiple groups of simulated pump groups. Each simulated pump group includes all non-warning pumps and 0 to N warning pumps. The optimization control module configures the multiple simulated pump groups in the optimal way to generate multiple warning configuration schemes.

[0098] It should be noted that the health operation data includes data corresponding to each water pump. The monitoring and early warning range includes the range of various monitoring data indicators and the early warning duration threshold for each water pump. Due to the different health statuses of each water pump, the corresponding monitoring and early warning ranges are different and determined by the health status. For example, water pumps with good health status have a larger early warning range, better operational tolerance, and small monitoring fluctuations in a short period of time will not affect the overall operation; their early warning duration thresholds are also larger. Conversely, water pumps with poor health status have a smaller early warning range, requiring them to be kept within a relatively strict data fluctuation range to prevent anomalies; their early warning duration thresholds are also smaller. The monitoring fluctuation curve is a numerical fluctuation curve formed by analyzing and predicting data, which can well reflect the predicted monitoring value fluctuation trend. The extreme value duration is the duration within the interval corresponding to the maximum and minimum values ​​of the monitoring data, used to analyze the duration when the early warning range is exceeded, and comparing the duration with the early warning duration threshold. Water pumps exceeding the early warning range are marked as early warning water pumps, including comparisons of monitoring data indicator ranges and early warning duration thresholds. The number of warning pumps is set to N. Multiple pumps are combined to form multiple simulated pump groups. Specifically, combinations of pumps that may encounter emergency situations are generated based on the warning pumps. For example, when the non-warning pump is one pump A and the warning pumps are two pumps B1 and B2, the multiple simulated pump groups include four combinations: (A), (A, B1), (A, B2), and (A, B1, B2). These generate four warning schemes. When a warning pump malfunctions, dynamic pump operation can be controlled through the corresponding warning scheme to ensure the real-time safe, efficient, and low-energy-consumption operation of the pump group.

[0099] Based on the above calculation results, the values ​​of P1, 2*P2, ..., n*Pn are sorted to select the data with the smallest value, which is then marked as k*Pk. The optimal number of operating water pumps is k, and the optimal total shaft power of the water pumps is k*Pk. By calculating the optimal result, the number of operating water pumps and the specific frequency of water pump operation can be determined.

[0100] The following are embodiments of the present invention used to further illustrate the calculation process:

[0101] Example of pump flow self-test (flow self-test module): The pump is currently running stably at a frequency of 50Hz and a shaft power of 12.8kW. Based on the pump flow-shaft power curve or the mathematical formula of the curve at the current frequency, the current water flow rate of the water system can be directly calculated to be 120m3 / h. Then, based on the pump flow-head curve or the mathematical formula of the curve at the current frequency, the current head of the system can be calculated to be 30.3m.

[0102] Alternatively, if the pressure difference across the pump is known to be 30.3m (the pressure difference across the pump is the head of the current system) and the frequency is 50Hz, the water flow rate of the current water system can be directly calculated as 120m³ / h by combining the pump flow rate-head curve or the mathematical formula of the curve at the current frequency.

[0103] Example of pump head self-test (head self-test module): Given that the required flow rate is 120 m3 / h and the pump operating frequency is 50 Hz, the required head of the system can be calculated directly using the pump flow rate-head curve or the mathematical formula of the curve at the current frequency. The head is 30.3 m (the pressure difference before and after the pump is the head of the current system).

[0104] Alternatively, given that the required water flow rate of the system is 120 m³ / h, the shaft power of the pump is 12.8 kW. Based on the pump flow rate-shaft power curve or the mathematical formula of the curve at different frequencies, the frequency of the currently operating pump is calculated to be 50 Hz. Then, combined with the pump flow rate-head curve or the mathematical formula of the curve at the current frequency, the required head of the system is calculated to be 30.3 m.

[0105] Example of dynamic optimization control for a pump set (optimization control module): The existing pump set consists of three pumps (2 in operation and 1 on standby). Each pump has a design flow rate of 100 m³ / h at 50Hz and a design head of 32 m. The pump shaft power is 11.4 kW. The maximum design flow rate of the entire pipeline network is 200 m³ / h. At this flow rate, the system head is 32 m. Assuming the pipe diameter is DN200, the flow velocity in the pipe is 1.651 m / s. If the current system requires a flow rate of 150 m³ / h, the flow velocity in the pipe is 1.238 m / s, and the required system head is 32 * (1.238 / 1.651)^2 = 18 m.

[0106] A. When the number of pumps in operation is 1, and the single pump meets the requirements of 18m head and 150m3 / h flow rate, the frequency is 44Hz. Based on the flow rate of 150m3 / h and the frequency of 44Hz, combined with the mathematical formula of the pump flow rate-head curve or curve at the current frequency, the shaft power of the pump at this time is calculated to be 11.14kW.

[0107] B. When there are 2 pumps in operation, and each pump meets the requirements of 18m head and 150 / 2 = 75m3 / h flow rate (in the case of parallel pumps, the head is the same and the flow rate is superimposed, simplifying the calculation, at this time all pumps are considered to be pumps with the same parameters), the frequency is 37.5Hz. Based on the flow rate of 150 / 2 = 75m3 / h and 37.5Hz, combined with the mathematical formula of the pump flow rate-head curve or curve at the current frequency, the shaft power of a single pump is calculated to be 4.82kW, and the total shaft power of the pump is 2*4.82 = 9.64kW.

[0108] When there are 3 pumps in operation, and each pump meets the requirements of 18m head and 150 / 3 = 50m³ / h flow rate (in the case of parallel pumps, the head is the same and the flow rate is superimposed, simplifying the calculation, and assuming that all pumps have the same parameters), the frequency is 36.6Hz. Based on the flow rate of 150 / 3 = 50m³ / h and 36.6Hz, combined with the mathematical formula of the pump flow rate-head curve or curve at the current frequency, the shaft power of a single pump is calculated to be 3.55kW, and the total shaft power of the pumps is 3*3.55 = 10.65kW.

[0109] Based on the above calculation results, the values ​​of 11.14kW, 9.64kW, and 10.65kW are compared. The smallest value is 9.64kW. Therefore, the optimal number of pump units to operate is 2, and the frequency of each unit is 37.5Hz.

[0110] Figure 2 A block diagram of a multi-pump joint optimization system according to the present invention is shown.

[0111] A second aspect of the present invention also provides a multi-pump-based joint optimization system 2, the system comprising: a memory 21 and a processor 22, wherein the memory includes a multi-pump-based joint optimization program, and the multi-pump-based joint optimization program, when executed by the processor, performs the following steps:

[0112] The self-test module obtains the operating frequency, shaft power, and head data of the target pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data.

[0113] The head self-test module calculates the required head of the system based on the total water flow, operating frequency, and shaft power to obtain the required head.

[0114] By optimizing the control module, when the system's operating head and water flow rate are both the system's required values, it is determined to be in a stable state. In the stable state, the parameters of the target pump group are obtained, including: the number of pumps in parallel is n, the required head of the system is H, the required flow rate is Q, the pump operating frequency is f, and the pump shaft power is P.

[0115] When the number of pumps in the target pump group is 1, the frequency of a single pump is f1 if it meets the head H and Q. Based on Q and f1, and combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency, the shaft power of the pump at this time is calculated to be P1.

[0116] When the number of pumps in the target pump group is 2, if a single pump meets the requirements of head H and Q / 2 (in the case of pumps in parallel, the head is the same and the flow rate is superimposed, simplifying the calculation, at this time all pumps are treated as pumps with the same parameters), the frequency is f2. Based on Q / 2 and f2, combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of a single pump is calculated to be P2, and the total shaft power of the pump is 2*P2.

[0117] When the number of pumps in the target pump group is n, if a single pump meets the requirements of head H and Q / n (in the case of pumps in parallel, the head is the same and the flow rate is superimposed, simplifying the calculation, at this time all pumps are treated as pumps with the same parameters), the frequency is fn. Based on Q / n and fn, combined with the pump flow rate-head curve or the mathematical formula of the curve at the current frequency, the shaft power of a single pump is calculated to be Pn, and the total shaft power of the pump is n*Pn.

[0118] Based on the above calculation results, the values ​​of P1, 2*P2, ..., n*Pn are sorted to select the data with the smallest value and marked as k*Pk. The optimal number of pumps in operation is k, and the optimal total shaft power of the pumps is k*Pk.

[0119] Based on the optimal number of operating water pumps and the optimal total shaft power of water pumps, an optimized configuration scheme is generated.

[0120] A prediction model based on LSTM is constructed. The model is trained using historical monitoring data of the target pump group. The prediction model is then used to make predictions based on the monitoring data of the target pump group in the current period, so as to obtain the predicted pump monitoring data for the next period.

[0121] Each pump is evaluated based on the health operation data of the target pump group, and a monitoring and early warning range is generated. The early warning pump information is generated by combining the predicted pump monitoring data. Based on the early warning pump information and the optimized control module, multiple early warning configuration schemes for the next cycle are generated.

[0122] It should be noted that in the optimal number of operating pumps (k) and the optimal total shaft power (k*Pk), when the water system is operating stably according to its needs, the current head and flow rate of the water system are the system's required values. The actual dynamic optimization control of the pump set must meet the current system's required flow rate and head, that is, to minimize the power consumption of the pump set (i.e., minimize the total shaft power) while meeting the current system's required flow rate and head. The method of this invention can dynamically calculate the scheme with the lowest power consumption. The target pump set is considered to be in a stable operating state after the equipment has been running for a period of time. The required head (H) can be obtained by calculating the required head using this invention, and the required flow rate (Q) is a preset value, determined according to actual needs. The optimized configuration scheme includes information such as the selected number of operating pumps and the set operating frequency.

[0123] According to an embodiment of the present invention, the step of acquiring the operating frequency, shaft power, and head data of the target water pump set during stable operation through the flow self-test module, and calculating the total water flow based on the operating frequency, shaft power, and head data, specifically involves:

[0124] When the target pump set is running stably, the total water flow of the current water system is calculated based on the operating frequency and shaft power of the target pump set, combined with the pump flow-shaft power curve or the mathematical formula of the curve at the current frequency.

[0125] Based on the total water flow, the real-time head of the current system is calculated according to the pump flow-head curve or the mathematical formula of the curve at the current frequency.

[0126] It should be noted that the flow self-checking module is used to calculate the total water flow data in real time and to calculate whether the total water flow meets the current required flow rate Q, where Q is a preset value. In the mathematical formula of the water pump flow-shaft power curve or curve, the mathematical formula of the curve is the mathematical expression of the water pump flow-shaft power curve, and the two calculation methods are consistent. In the mathematical formula of the curve, each relationship curve appearing in this embodiment has a corresponding mathematical expression.

[0127] According to an embodiment of the present invention, the step of acquiring the operating frequency, shaft power, and head data of the target water pump set during stable operation through the flow self-test module, and calculating the total water flow based on the operating frequency, shaft power, and head data, further includes:

[0128] When the target pump set is running stably, the head data is obtained based on the real-time pressure data of the target pump set (the pressure difference before and after the pump is the head of the current system). Combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency of the target pump set, the total water flow of the current water system is calculated.

[0129] It should be noted that in this embodiment, the calculation of the total water flow includes the two methods mentioned above, and the specific method to be used depends on the actual data acquired and the application.

[0130] According to an embodiment of the present invention, the step of calculating the required head of the system based on the total water flow, operating frequency, and shaft power using the head self-test module to obtain the required head is specifically as follows:

[0131] By using preset flow meters and preset pump sensor control reference points, and through PID control, when the actual flow rate of the target pump group reaches the system's required flow rate, the required head of the water system is calculated based on the pump flow-head curve or the mathematical formula of the curve at the current operating frequency.

[0132] It should be noted that the water system refers to the entire target water pump set.

[0133] According to an embodiment of the present invention, the step of calculating the required head of the system based on the total water flow, operating frequency, and shaft power using a head self-test module to obtain the required head further includes:

[0134] If the required water flow rate and the shaft power of the pump are known, the current frequency of the target pump set can be calculated based on the pump flow rate-shaft power curve or the mathematical formula of the curve at different operating frequencies. Then, the required head of the water system can be calculated by combining the pump flow rate-head curve or the mathematical formula of the curve at the current frequency.

[0135] It should be noted that there are two methods for calculating the required head in this embodiment, and the specific method to be used depends on the actual situation.

[0136] According to an embodiment of the present invention, the step of constructing an LSTM-based prediction model, training the model using historical monitoring data of the target pump group, and predicting the next period's pump monitoring data by comparing the prediction model with the current period's monitoring data of the target pump group, includes:

[0137] Construct a prediction model based on LSTM;

[0138] Obtain historical monitoring data of the target water pump set, including water pressure, power, and flow rate data;

[0139] Based on the time dimension, historical monitoring data is serialized to form a time series dataset;

[0140] The time series dataset is divided into a training set and a test set according to a preset ratio. The training set and the test set are then imported into the prediction model for training, and the trained prediction model is obtained.

[0141] It should be noted that the historical monitoring data refers to monitoring data from a recent period, which is representative and can effectively reflect the current operating status of the water pump set, and is used as training data.

[0142] According to an embodiment of the present invention, the step of predicting the pump monitoring data for the next period by using a prediction model and the monitoring data of the target pump group in the current period specifically involves:

[0143] Acquire the monitoring data of the target pump group in the current cycle and mark it as the current monitoring data;

[0144] Based on the time dimension, the current monitoring data is serialized and imported into the prediction model for data prediction, generating the predicted pump monitoring data for the next cycle.

[0145] It should be noted that the current monitoring data includes the pressure, power, flow rate, temperature, vibration, and current data of the water pump set. These monitoring data often exhibit periodic changes over time and have a certain degree of predictability. This invention uses an LSTM model for data prediction, effectively performing predictive analysis on the data.

[0146] According to an embodiment of the present invention, the evaluation of each pump based on the health operation data of the target pump group and the generation of a monitoring and early warning range, the generation of early warning pump information by combining the predicted pump monitoring data, and the generation of multiple early warning configuration schemes for the next cycle based on the early warning pump information and the optimized control module, specifically:

[0147] Acquire the health operation data of the target water pump set, including the water pump set's operating time, equipment life assessment, and maintenance record data;

[0148] Based on the health operation data, the monitoring operation parameter range of each pump in the pump group is evaluated, and the monitoring and early warning range information of the pump group is generated.

[0149] Taking one pump in the target pump group as the unit of analysis, the time-dimensional predictive monitoring numerical fluctuation analysis of the pump is carried out based on the predicted pump monitoring data to obtain the monitoring fluctuation curve and the duration of extreme values.

[0150] Based on the monitoring fluctuation curve, extreme value duration and monitoring warning range information, a warning analysis is performed, and water pumps exceeding the warning range are marked as warning water pumps, thus forming warning water pump information;

[0151] Let the number of warning pumps be N. The target pump group is combined into multiple groups of simulated pump groups. Each simulated pump group includes all non-warning pumps and 0 to N warning pumps. The optimization control module configures the multiple simulated pump groups in the optimal way to generate multiple warning configuration schemes.

[0152] It should be noted that the health operation data includes data corresponding to each water pump. The monitoring and early warning range includes the range of various monitoring data indicators and the early warning duration threshold for each water pump. Due to the different health statuses of each water pump, the corresponding monitoring and early warning ranges are different and determined by the health status. For example, water pumps with good health status have a larger early warning range, better operational tolerance, and small monitoring fluctuations in a short period of time will not affect the overall operation; their early warning duration thresholds are also larger. Conversely, water pumps with poor health status have a smaller early warning range, requiring them to be kept within a relatively strict data fluctuation range to prevent anomalies; their early warning duration thresholds are also smaller. The monitoring fluctuation curve is a numerical fluctuation curve formed by analyzing and predicting data, which can well reflect the predicted monitoring value fluctuation trend. The extreme value duration is the duration within the interval corresponding to the maximum and minimum values ​​of the monitoring data, used to analyze the duration when the early warning range is exceeded, and comparing the duration with the early warning duration threshold. Water pumps exceeding the early warning range are marked as early warning water pumps, including comparisons of monitoring data indicator ranges and early warning duration thresholds. The number of warning pumps is set to N. Multiple pumps are combined to form multiple simulated pump groups. Specifically, combinations of pumps that may encounter emergency situations are generated based on the warning pumps. For example, when the non-warning pump is one pump A and the warning pumps are two pumps B1 and B2, the multiple simulated pump groups include four combinations: (A), (A, B1), (A, B2), and (A, B1, B2). These generate four warning schemes. When a warning pump malfunctions, dynamic pump operation can be controlled through the corresponding warning scheme to ensure the real-time safe, efficient, and low-energy-consumption operation of the pump group.

[0153] Based on the above calculation results, the values ​​of P1, 2*P2, ..., n*Pn are sorted to select the data with the smallest value, which is then marked as k*Pk. The optimal number of operating water pumps is k, and the optimal total shaft power of the water pumps is k*Pk. By calculating the optimal result, the number of operating water pumps and the specific frequency of water pump operation can be determined.

[0154] The following are embodiments of the present invention used to further illustrate the calculation process:

[0155] Example of pump flow self-test (flow self-test module): The pump is currently running stably at a frequency of 50Hz and a shaft power of 12.8kW. Based on the pump flow-shaft power curve or the mathematical formula of the curve at the current frequency, the current water flow rate of the water system can be directly calculated to be 120m3 / h. Then, based on the pump flow-head curve or the mathematical formula of the curve at the current frequency, the current head of the system can be calculated to be 30.3m.

[0156] Alternatively, if the pressure difference across the pump is known to be 30.3m (the pressure difference across the pump is the head of the current system) and the frequency is 50Hz, the water flow rate of the current water system can be directly calculated as 120m³ / h by combining the pump flow rate-head curve or the mathematical formula of the curve at the current frequency.

[0157] Example of pump head self-test (head self-test module): Given that the required flow rate is 120 m3 / h and the pump operating frequency is 50 Hz, the required head of the system can be calculated directly using the pump flow rate-head curve or the mathematical formula of the curve at the current frequency. The head is 30.3 m (the pressure difference before and after the pump is the head of the current system).

[0158] Alternatively, given that the required water flow rate of the system is 120 m³ / h, the shaft power of the pump is 12.8 kW. Based on the pump flow rate-shaft power curve or the mathematical formula of the curve at different frequencies, the frequency of the currently operating pump is calculated to be 50 Hz. Then, combined with the pump flow rate-head curve or the mathematical formula of the curve at the current frequency, the required head of the system is calculated to be 30.3 m.

[0159] Example of dynamic optimization control for a pump set (optimization control module): The existing pump set consists of three pumps (2 in operation and 1 on standby). Each pump has a design flow rate of 100 m³ / h at 50Hz and a design head of 32 m. The shaft power of the pump is 11.4 kW. The maximum design flow rate of the entire pipeline network is 200 m³ / h. At this flow rate, the system head is 32 m. Assuming the pipe diameter is DN200, the flow velocity in the pipe is 1.651 m / s. If the current system requires a flow rate of 150 m³ / h, the flow velocity in the pipe is 1.238 m / s, and the required system head is 32 * (1238 / 1651)^2 = 18 m.

[0160] A. When the number of pumps in operation is 1, and the single pump meets the requirements of 18m head and 150m3 / h flow rate, the frequency is 44Hz. Based on the flow rate of 150m3 / h and the frequency of 44Hz, combined with the mathematical formula of the pump flow rate-head curve or curve at the current frequency, the shaft power of the pump at this time is calculated to be 11.14kW.

[0161] B. When there are 2 pumps in operation, and each pump meets the requirements of 18m head and 150 / 2 = 75m3 / h flow rate (in the case of parallel pumps, the head is the same and the flow rate is superimposed, simplifying the calculation, at this time all pumps are considered to be pumps with the same parameters), the frequency is 37.5Hz. Based on the flow rate of 150 / 2 = 75m3 / h and 37.5Hz, combined with the mathematical formula of the pump flow rate-head curve or curve at the current frequency, the shaft power of a single pump is calculated to be 4.82kW, and the total shaft power of the pump is 2*4.82 = 9.64kW.

[0162] When there are 3 pumps in operation, and each pump meets the requirements of 18m head and 150 / 3 = 50m³ / h flow rate (in the case of parallel pumps, the head is the same and the flow rate is superimposed, simplifying the calculation, and assuming that all pumps have the same parameters), the frequency is 36.6Hz. Based on the flow rate of 150 / 3 = 50m³ / h and 36.6Hz, combined with the mathematical formula of the pump flow rate-head curve or curve at the current frequency, the shaft power of a single pump is calculated to be 3.55kW, and the total shaft power of the pumps is 3*3.55 = 10.65kW.

[0163] Based on the above calculation results, the values ​​of 11.14kW, 9.64kW, and 10.65kW are compared. The smallest value is 9.64kW. Therefore, the optimal number of pump units to operate is 2, and the frequency of each unit is 37.5Hz.

[0164] A third aspect of the present invention also provides a computer-readable storage medium comprising a multi-pump-based joint optimization program, wherein when executed by a processor, the multi-pump-based joint optimization program implements the steps of the multi-pump-based joint optimization method as described in any of the preceding claims.

[0165] This invention discloses a joint optimization analysis method, system, and storage medium based on multiple water pumps. Through a flow self-checking module, it acquires the operating frequency, shaft power, and head data of the target water pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data. Through a head self-checking module, it calculates the required head of the system based on the total water flow, operating frequency, and shaft power, obtaining the required head. Through an optimization control module, when both the system's operating head and water flow are the required values, it is determined to be in a stable state. In the stable state, the optimization control module generates the optimal configuration scheme for the current water pump set.

[0166] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0167] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0168] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0169] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0170] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0171] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A joint optimization method based on multiple water pumps, characterized in that, include: The self-test module obtains the operating frequency, shaft power, and head data of the target pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data. The head self-test module calculates the required head of the system based on the total water flow, operating frequency, and shaft power to obtain the required head. By optimizing the control module, when the system's operating head and water flow rate are both the system's required values, it is determined to be in a stable state. In the stable state, the parameters of the target pump group are obtained, including: the number of pumps in parallel is n, the required head of the system is H, the required flow rate is Q, the pump operating frequency is f, and the pump shaft power is P. When the number of pumps in the target pump group is 1, the frequency of a single pump is f1 if it meets the head H and Q. Based on Q and fl, and combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency, the shaft power of the pump at this time is calculated to be P1. When the number of pumps in the target pump group is 2, the frequency of each pump is f2 when the head H and Q / 2 are met. Based on Q / 2 and f2, combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of the single pump is P2 and the total shaft power of the pump is 2*P2. When the number of pumps in the target pump set is n, the frequency of a single pump is fn when the head H and Q / n are met. Based on Q / n and fn, and combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of a single pump is Pn and the total shaft power of the pump is n*Pn. Based on the above calculation results, the values ​​of P1, 2*P2, ..., n*Pn are sorted to select the data with the smallest value and marked as k*Pk. The optimal number of pumps in operation is k, and the optimal total shaft power of the pumps is k*Pk. Based on the optimal number of operating water pumps and the optimal total shaft power of the water pumps, an optimized configuration scheme is generated. A prediction model based on LSTM is constructed. The model is trained using historical monitoring data of the target pump group. The prediction model is then used to make predictions based on the monitoring data of the target pump group in the current period, so as to obtain the predicted pump monitoring data for the next period. Each pump is evaluated based on the health operation data of the target pump group, and a monitoring and early warning range is generated. The early warning pump information is generated by combining the predicted pump monitoring data. Based on the early warning pump information and the optimized control module, multiple early warning configuration schemes for the next cycle are generated.

2. The method for joint optimization based on multiple water pumps according to claim 1, characterized in that, The process involves using a flow self-test module to acquire data on the operating frequency, shaft power, and head of the target pump set during stable operation. Based on this data, the total water flow rate is calculated. Specifically: When the target pump set is running stably, the total water flow of the current water system is calculated based on the operating frequency and shaft power of the target pump set, combined with the pump flow-shaft power curve or the mathematical formula of the curve at the current frequency. Based on the total water flow, the real-time head of the current system is calculated according to the pump flow-head curve or the mathematical formula of the curve at the current frequency.

3. The joint optimization method based on multiple water pumps according to claim 1, characterized in that, The method of acquiring the operating frequency, shaft power, and head data of the target pump set during stable operation through the flow self-test module, and calculating the total water flow based on the operating frequency, shaft power, and head data, further includes: When the target pump set is running stably, the head data is obtained based on the real-time pressure data of the target pump set. Combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency of the target pump set, the total water flow of the current water system is calculated.

4. The method for joint optimization based on multiple water pumps according to claim 1, characterized in that, The head self-checking module calculates the required head of the system based on the total water flow, operating frequency, and shaft power to obtain the required head. Specifically: By using preset flow meters and preset pump sensor control reference points, and through PID control, when the actual flow rate of the target pump group reaches the system's required flow rate, the required head of the water system is calculated based on the pump flow-head curve or the mathematical formula of the curve at the current operating frequency.

5. The joint optimization method based on multiple water pumps according to claim 1, characterized in that, The method of calculating the required system head based on total water flow, operating frequency, and shaft power using a head self-checking module to obtain the required head also includes: If the required water flow rate and the shaft power of the pump are known, the current frequency of the target pump set can be calculated based on the pump flow rate-shaft power curve or the mathematical formula of the curve at different operating frequencies. Then, the required head of the water system can be calculated by combining the pump flow rate-head curve or the mathematical formula of the curve at the current frequency.

6. The joint optimization method based on multiple water pumps according to claim 1, characterized in that, The construction of the LSTM-based prediction model involves training the model using historical monitoring data of the target pump group, and then using the prediction model to predict the pump monitoring data for the next period using the monitoring data of the target pump group in the current period. This includes: Construct a prediction model based on LSTM; Obtain historical monitoring data of the target water pump set, including water pressure, power, and flow rate data; Based on the time dimension, historical monitoring data is serialized to form a time series dataset; The time series dataset is divided into a training set and a test set according to a preset ratio. The training set and the test set are then imported into the prediction model for training, and the trained prediction model is obtained.

7. The joint optimization method based on multiple water pumps according to claim 6, characterized in that, The process of using a prediction model and the monitoring data of the target pump group in the current cycle to predict the pump monitoring data for the next cycle is as follows: Acquire the monitoring data of the target pump group in the current cycle and mark it as the current monitoring data; Based on the time dimension, the current monitoring data is serialized and imported into the prediction model for data prediction, generating the predicted pump monitoring data for the next cycle.

8. The joint optimization method based on multiple water pumps according to claim 7, characterized in that, The process involves evaluating each pump based on the health operation data of the target pump group, generating a monitoring and early warning range, combining this with predicted pump monitoring data to generate early warning pump information, and then using this early warning pump information and the optimized control module to generate multiple early warning configuration schemes for the next cycle. Specifically: Acquire the health operation data of the target water pump set, including the water pump set's operating time, equipment life assessment, and maintenance record data; Based on the health operation data, the monitoring operation parameter range of each pump in the pump group is evaluated, and the monitoring and early warning range information of the pump group is generated. Taking one pump in the target pump group as the unit of analysis, the time-dimensional predictive monitoring numerical fluctuation analysis of the pump is carried out based on the predicted pump monitoring data to obtain the monitoring fluctuation curve and the duration of extreme values. Based on the monitoring fluctuation curve, extreme value duration and monitoring warning range information, a warning analysis is performed, and water pumps exceeding the warning range are marked as warning water pumps, thus forming warning water pump information; Let the number of warning pumps be N. The target pump group is combined into multiple groups of simulated pump groups. Each simulated pump group includes all non-warning pumps and 0 to N warning pumps. The optimization control module configures the multiple simulated pump groups in the optimal way to generate multiple warning configuration schemes.

9. A joint optimization system based on multiple water pumps, characterized in that, The system includes: a memory and a processor. The memory includes a multi-pump-based joint optimization program. When the processor executes the multi-pump-based joint optimization program, it performs the following steps: The self-test module obtains the operating frequency, shaft power, and head data of the target pump set during stable operation, and calculates the total water flow based on the operating frequency, shaft power, and head data. The head self-test module calculates the required head of the system based on the total water flow, operating frequency, and shaft power to obtain the required head. By optimizing the control module, when the system's operating head and water flow rate are both the system's required values, it is determined to be in a stable state. In the stable state, the parameters of the target pump group are obtained, including: the number of pumps in parallel is n, the required head of the system is H, the required flow rate is Q, the pump operating frequency is f, and the pump shaft power is P. When the number of pumps in the target pump group is 1, the frequency of a single pump is f1 if it meets the head H and Q. Based on Q and f1, and combined with the pump flow-head curve or the mathematical formula of the curve at the current operating frequency, the shaft power of the pump at this time is calculated to be P1. When the number of pumps in the target pump group is 2, the frequency of each pump is f2 when the head H and Q / 2 are met. Based on Q / 2 and f2, combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of the single pump is P2 and the total shaft power of the pump is 2*P2. When the number of pumps in the target pump set is n, the frequency of a single pump is fn when the head H and Q / n are met. Based on Q / n and fn, and combined with the pump flow-head curve or the mathematical formula of the curve at the current frequency, the shaft power of a single pump is Pn and the total shaft power of the pump is n*Pn. Based on the above calculation results, the values ​​of P1, 2*P2, ..., n*Pn are sorted to select the data with the smallest value and marked as k*Pk. The optimal number of pumps in operation is k, and the optimal total shaft power of the pumps is k*Pk. Based on the optimal number of operating water pumps and the optimal total shaft power of the water pumps, an optimized configuration scheme is generated. A prediction model based on LSTM is constructed. The model is trained using historical monitoring data of the target pump group. The prediction model is then used to make predictions based on the monitoring data of the target pump group in the current period, so as to obtain the predicted pump monitoring data for the next period. Each pump is evaluated based on the health operation data of the target pump group, and a monitoring and early warning range is generated. The early warning pump information is generated by combining the predicted pump monitoring data. Based on the early warning pump information and the optimized control module, multiple early warning configuration schemes for the next cycle are generated.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a multi-pump-based joint optimization program, which, when executed by a processor, implements the steps of the multi-pump-based joint optimization method as described in any one of claims 1 to 8.

Images