Pump station group dispatching system and method based on digital twinning

Abstract

The application provides a pump station group scheduling system and method based on digital twinning, and relates to the technical field of pump station scheduling.The pump station group digital twinning model is constructed according to the bearing temperature curve, the flow curve and the pump cabin pressure curve when the rated power works in history, the solving condition I, the solving condition II and the solving condition III are set according to the task quantity and the task time, the scheduling quantity that each pump station should obtain is solved in MATLAB, the scheduling quantity is simulated in the pump station group digital twinning model, the simulated running state is obtained, the running state score is constructed, the health state score is constructed according to the data of the pump station, the ability of each pump station to complete the task is judged by analyzing the order of the running state score and the health state score, if there is a situation that cannot be completed, the scheduling quantity is redistributed, the failure caused by the overload operation of the pump station is avoided, and the smooth progress of the scheduling task is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pump station scheduling, in particular to a pump station group scheduling system and method based on digital twinning. BACKGROUND

[0002] As the core infrastructure in the fields of water conservancy projects, urban water supply, and agricultural irrigation, a pump station group is usually composed of multiple pump stations distributed in a wide range, and undertakes key tasks such as flow regulation, pressure control, and water resource allocation. With the development of intelligent technology, the scheduling requirements of the pump station group are becoming increasingly complex, and need to take into account task efficiency, equipment reliability, and energy consumption optimization. Especially in cross-regional water resource allocation and emergency drainage scenarios, the coordinated operation of the pump station group needs to dynamically adapt to different working conditions and achieve multi-objective optimization. Digital twinning technology provides a technical basis for the coordinated interaction of physical equipment and virtual models of the pump station group through real-time data mapping and simulation modeling, and is suitable for scenarios such as equipment state prediction, scheduling strategy verification, and dynamic optimization, becoming an important direction for improving the intelligent scheduling capability of the pump station group.

[0003] In the prior art, the patent with publication number CN117450050A discloses a drainage pump station intelligent scheduling system based on digital twinning technology. The system stores historical data of pump stations in the control area during scheduling work in a database; a model training module constructs a pump station digital twinning model; a data acquisition module acquires real-time data; and the real-time data is sent to an intelligent scheduling module; and the intelligent scheduling module obtains a scheduling scheme according to the real-time data and the pump station digital twinning model.

[0004] Although the disclosed technical document realizes the scheduling of pump stations, it does not take into account the different scheduling capabilities of different pump stations, nor does it consider the problem of reduced scheduling capability caused by actual working conditions of the pump stations, which can easily lead to an unsuccessful scheduling process, thereby affecting normal production activities.

[0005] The above information disclosed in the background section is only used to enhance the understanding of the background of the present disclosure, and therefore it can include information that does not constitute the prior art known to those of ordinary skill in the art. SUMMARY

[0006] The present application aims to provide a pump station group scheduling system and method based on digital twinning to solve the problems raised in the background.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0008] A pump station group scheduling method based on digital twinning, the specific steps of which include

[0009] Step 1: Obtain the bearing temperature curve, flow curve, and pump cabin pressure curve of each pump station when operating at rated power, train the bearing temperature model, pump cabin pressure model, and flow model of each pump station respectively, and aggregate the models of all pump stations to obtain the pump station group digital twin model;

[0010] Step 2: Set the scheduling quantity of each pump station according to the scheduled task quantity and task time, set solving condition I, solving condition II, and solving condition III, and solve in MATLAB to obtain the scheduling quantity of each pump station;

[0011] Step 3: Each pump station obtains the operation duration of the pump station according to the scheduling quantity and inputs the pump station group digital twin model to obtain the bearing temperature curve, pump cabin pressure curve, and flow curve of each pump station, obtains the flow stability according to the flow curve, obtains the pressure stability according to the pump cabin pressure curve, and obtains the operation state score of each pump station in combination with the temperature threshold of each pump station;

[0012] Step 4: Obtain the equipment duration, operation duration, repair frequency, and duration since the last minor repair of each pump station, construct a health state score, sort the pump stations by the operation state score and the health state score, determine whether each pump station meets the scheduling quantity according to the sorting, adjust the task quantity of the pump station that does not meet the scheduling quantity and redistribute the task quantity of other pump stations, and alarm if there is always a pump station that cannot meet the scheduling quantity.

[0013] Further, the bearing temperature curve, flow curve, and pump cabin pressure curve of each pump station in the history when operating at rated power are obtained respectively through the work log of each pump station, the horizontal axis of the curve is the operation duration, the starting point of the operation duration is 0, and the terminal point is the actual operation duration of the equipment, and the vertical axis is the numerical value of each type of data;

[0014] Take a single pump station as a unit, take the start-up duration as a training set, and input the bearing temperature curve, pump cabin pressure curve, and flow curve corresponding to different start-up durations into the convolutional neural network model for training to obtain the bearing temperature model, pump cabin pressure model, and flow model of each pump station, aggregate the models of all pump stations to obtain the pump station group digital twin model.

[0015] Further, the rated flow and rated power of each pump station are obtained, the scheduled task quantity and task time of the pump station group are obtained from the superior, the scheduled task quantity is distributed among the pump stations, the scheduling quantity of each pump station is set, and the solving is performed in MATLAB, and the solving conditions are set:

[0016] The formula according to solving condition I is as follows:

[0017]

[0018] Wherein, V is the scheduled task amount, V i The scheduling amount of the pump station numbered i, i is the pump station number search variable, i∈N, 1≤i≤M;

[0019] The formula for solving condition II is as follows:

[0020]

[0021] Wherein, MAX represents selecting the maximum value, Q i The rated flow of the pump station numbered i, T z is the task time;

[0022] The formula for solving condition III is as follows:

[0023]

[0024] Wherein, MIN is the minimum function, P i The rated power of the pump station numbered i;

[0025] Get the scheduling amount of each pump station after solving.

[0026] Further, each pump station obtains the running duration according to the obtained scheduling amount and the rated flow of each pump station, and the formula is as follows:

[0027]

[0028] Wherein, T i The running duration of the pump station numbered i, Q i The rated flow of the pump station numbered i, i is the pump station number search variable, i∈N, 1≤i≤M;

[0029] Input the start-up duration of each pump station into the corresponding pump station group digital twin model, obtain the bearing temperature curve, pump cabin pressure curve and flow curve of each pump station, and obtain the flow stability according to the flow curve of each pump station, and the formula is as follows:

[0030]

[0031] Wherein, σ Q-i The flow stability value when the running duration of the pump station numbered i is T i , T i is the running duration, Q i (t) is the flow of the pump station numbered i when the duration is t, Q i The average flow of the pump station numbered i, t is the running duration search variable, 0<t≤T i , i is the pump station number search variable, i∈N, 1≤i≤M;

[0032] The pressure stability is obtained according to the pump cabin pressure curve of each pump station, and the formula is as follows:

[0033]

[0034] Wherein, σ p-i represents the pressure stability value of the pump station numbered i when the operation time is T i , T i is the operation time, p i (t) represents the pressure of the pump station numbered i when the time is t, p i represents the average pressure of the pump station numbered i, i is the operation time retrieval variable, 0<t≤T i , i is the pump station number retrieval variable, i∈N, 1≤i≤M.

[0035] The temperature threshold of each pump station is obtained, the highest temperature is obtained according to the bearing temperature curve of each pump station, and the operation state score of each pump is constructed, and the formula is as follows:

[0036]

[0037] Wherein, SY i represents the operation state score of the pump station numbered i, σ p-i represents the pressure stability of the pump station numbered i, σ Q-i represents the flow stability of the pump station numbered i, T max- i represents the highest temperature of the pump station numbered i, T th- i represents the temperature threshold of the pump station numbered i, i is the pump station number retrieval variable, i∈N, 1≤i≤M.

[0038] Further, the equipment time, operation time, overhaul times and time from the last minor repair of each pump station are obtained through the maintenance log, the minor repair and minor repair are required maintenance levels of equipment maintenance, wherein the equipment time is the time from the commissioning of the pump to the current, the operation time is the use time after the pump is commissioned, and the health state score of each pump station is constructed, and the formula is as follows:

[0039]

[0040] Wherein, SJ i represents the health state score of the pump station numbered i, T equip-i represents the equipment time of the pump station numbered i, T run-i represents the operation time of the pump station numbered i, N mid-i represents the overhaul times of the pump station numbered i, T last-irepresents the time length of the pump station numbered i from the last minor repair, i is a pump station number search variable, i is in N, 1<=i<=M, k1 and k2 are coefficients, k1<k2.

[0041] Further, the health state score and the running state score of all pump stations are respectively sorted from high to low, the pump stations ranked in the last third in the health state score sorting are obtained, the pump stations ranked in the last third in the running state score sorting are obtained, the pump stations in the last third in both sortings are screened out, and the pump stations not meeting the scheduling quantity this time are marked.

[0042] Further, the scheduling quantity re-distribution logic is as follows:

[0043] Firstly, the pump station not meeting the scheduling quantity this time is adjusted, the number of the pump station is set as j, the adjusted scheduling quantity of the pump station is set as half of the scheduling quantity before adjustment, and the remaining task quantity is distributed among other pump stations, the adjusted scheduling quantity of other pump stations is re-solved in MATLAB, and the solving condition I is modified, and the formula is as follows:

[0044]

[0045] Wherein, V is the scheduled task quantity, V j represents the scheduling quantity of the pump station numbered j before adjustment, V i represents the scheduling quantity of the pump station numbered i, i is a pump station number search variable, i is in N, 1<=i<=M;

[0046] The solving condition II and the solving condition III are not modified, the scheduling quantity of each pump station after solving is re-obtained, and the scheduling quantity is sent to the corresponding pump station.

[0047] Further, each pump station executes steps 3 and 4 again according to the re-obtained scheduling quantity, until the score difference of all pump stations is greater than zero, and then the scheduling task is started to be executed, and when the score difference of all pump stations cannot be solved to be greater than zero, an alarm is given.

[0048] The application also includes a pump station group scheduling system based on digital twinning, which is used for executing the pump station group scheduling method based on digital twinning.

[0049] The model construction module is used for obtaining the bearing temperature curve, the flow curve and the pump cabin pressure curve of each pump station when the pump station is operated at rated power, training the bearing temperature model, the pump cabin pressure model and the flow model of each pump station respectively, and obtaining the pump station group digital twinning model by collecting the models of all pump stations.

[0050] The scheduling quantity solving module is configured to set the scheduling quantity of each pump station according to the scheduled task quantity and the task time, and set the solving condition I, the solving condition II and the solving condition III, and solve the scheduling quantity of each pump station in MATLAB;

[0051] The operation evaluation module is configured to obtain the operation duration of each pump station according to the scheduling quantity, input the pump station group digital twin model, obtain the bearing temperature curve, the pump cabin pressure curve and the flow curve of each pump station, obtain the flow stability according to the flow curve, obtain the pressure stability according to the pump cabin pressure curve, and obtain the operation state score of each pump station in combination with the temperature threshold of each pump station.

[0052] The state analysis module is configured to obtain the equipment duration, the operation duration, the overhaul times and the duration from the last minor repair of each pump station, construct the health state score, sort each pump station through the operation state score and the health state score, judge whether each pump station meets the scheduling quantity according to the sorting, adjust the task quantity of the pump station that does not meet the scheduling quantity and reassign the task quantity of other pump stations if the pump station that does not meet the scheduling quantity exists, and alarm if there is always a pump station that cannot meet the scheduling quantity.

[0053] Compared with the prior art, the beneficial effects of the present application are:

[0054] The present application simulates the scheduling quantity of each pump station by constructing the pump station group digital twin model, constructs the operation state score according to the data of the simulation operation, compares and analyzes the health state score and the operation state score of each pump station, assigns the scheduling quantity suitable for each pump station, avoids the failure caused by the overload operation of the pump station, and ensures the smooth progress of the scheduling task. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 The present application is a whole method flowchart;

[0056] Figure 2 The present application is a system structure schematic diagram. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below in combination with specific embodiments.

[0058] It should be noted that the technical terms or scientific terms used in the present application should be understood as the general meaning understood by those skilled in the art to which the present application belongs, unless otherwise defined. The terms "first", "second" and the like used in the present application do not represent any order, number or importance, but are only used to distinguish different components. The terms "include" or "contain" and the like mean that the elements or objects before the terms cover the elements or objects listed after the terms and their equivalents, and do not exclude other elements or objects. The terms "connected" or "connected" and the like are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to represent relative positional relationships, and when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0059] Embodiment:

[0060] Please refer to Figure 1 The present application provides a technical solution:

[0061] A pump station group scheduling method based on digital twinning, the specific steps comprising:

[0062] Step 1: Obtain the bearing temperature curve, flow curve and pump cabin pressure curve of each pump station when running at rated power, take the pump station as the unit, respectively train the bearing temperature model, pump cabin pressure model and flow model of each pump station, and collect the models of all pump stations to obtain the pump station group digital twinning model.

[0063] The step 1 includes the following contents:

[0064] Through the work log of each pump station, the bearing temperature curve, flow curve and pump cabin pressure curve of each pump station in history when running at rated power are obtained, the horizontal axis of the curve is the running time, the starting point of the running time is 0, and the terminal point is the actual running time of the equipment, and the vertical axis is the numerical value of each type of data.

[0065] Take a single pump station as the unit, take the starting time as the training set, and input the bearing temperature curve, pump cabin pressure curve and flow curve corresponding to different starting time into the convolutional neural network model for training to obtain the bearing temperature model, pump cabin pressure model and flow model of each pump station. Collect the models of all pump stations to obtain the pump station group digital twinning model.

[0066] As a preferred embodiment, a one-dimensional convolutional neural network is used in the convolutional neural network, the number of convolution kernels is 32, the size of the convolution kernel is 5x5, a maximum pooling layer is used, the pooling window is 2, the step is 2, and the number of neurons is 128.

[0067] By acquiring the historical operation data of the pump stations and training the digital twin model, a high-precision simulation foundation is laid for the entire scheduling system. By using a convolutional neural network combined with the starting duration of the pump stations and the actual operation curve (such as bearing temperature, flow rate, and pressure), the model can dynamically reflect the behavior characteristics of the pump stations under different working conditions. This step not only provides key inputs for the subsequent optimization of task quantity allocation, but also enables global collaborative analysis by aggregating the models of each pump station, allowing the multi-constraint condition solving in step 2 to be based on a unified virtual environment, ensuring the rationality and feasibility of the scheduling scheme. At the same time, the high fidelity of the model provides reliable data support for the real-time state prediction and score calculation in step 3.

[0068] Step 2: According to the scheduled task quantity and task time, set the scheduling quantity of each pump station, and set solving condition I, solving condition II, and solving condition III, and solve in MATLAB to obtain the scheduling quantity of each pump station.

[0069] The step 2 includes the following contents:

[0070] Obtain the rated flow rate and rated power of each pump station, obtain the scheduled task quantity and task time of the pump station group from the upper level, allocate the scheduled task quantity among the pump stations, set the scheduling quantity of each pump station and solve in MATLAB, and set the solving conditions:

[0071] The solving condition I is based on the following formula:

[0072]

[0073] Wherein, V is the scheduled task quantity, V i represents the scheduling quantity of the pump station numbered i, i is the pump station number retrieval variable, i ∈ N, 1 ≤ i ≤ M;

[0074] Wherein, by means of the sum of the scheduling quantity of each pump station being equal to the scheduled task quantity, it is ensured that the work of the pump station group can meet the scheduled task quantity.

[0075] The solving condition II is based on the following formula:

[0076]

[0077] Wherein, MAX represents the maximum value, Q i represents the rated flow rate of the pump station numbered i, T z is the task time;

[0078] Wherein, by judging the longest pump station working time, it is judged that the current scheduling task can be completed within the specified time.

[0079] The solving condition III is based on the following formula:

[0080]

[0081] Where MIN is the minimization function, P i This indicates the rated power of pump station numbered i;

[0082] By multiplying the working time of each pump station by its corresponding power, the power consumption of each pump station can be determined. The minimum power consumption can be obtained by minimizing the function, thereby achieving the goal of energy conservation.

[0083] In a preferred embodiment, the scheduling quantities of each pumping station are solved using the interior-point method, with a maximum iteration count of 1000 and an optimality tolerance of 10. -6 To ensure accuracy, if the problem remains unsolved, a global alert will be issued.

[0084] Obtain the scheduling amount for each pump station after the solution is obtained.

[0085] By setting triple constraints on task allocation (total task volume, time limit, and energy consumption minimization), a balance between scientific rigor and economic efficiency in the scheduling scheme is achieved. MATLAB is used to quickly handle complex constraint problems, ensuring that the scheduling load for each pump station meets both task requirements and operational efficiency and energy-saving goals. The solution results in this step provide direct parameters for the operation time calculation and digital twin model input in step 3, laying the foundation for subsequent state prediction and the construction of the scoring system. Furthermore, the dynamic allocation logic of the task volume implicitly considers the potential performance differences among pump stations, preparing for the determination and redistribution mechanism of scoring differences in step 4, thus forming a prerequisite for scheduling optimization.

[0086] Step 3: Each pump station obtains its own operating time based on the scheduling volume and inputs it into the digital twin model of the pump station group. The bearing temperature curve, pump chamber pressure curve and flow curve of each pump station are obtained. The flow stability is obtained based on the flow curve and the pressure stability is obtained based on the pump chamber pressure curve. Combined with the temperature threshold of each pump station, the operating status score of each pump station is obtained.

[0087] Step 3 includes the following:

[0088] The operating time for each pumping station is determined based on the obtained dispatch volume and the rated flow rate of each station, using the following formula:

[0089]

[0090] Among them, T i Q represents the operating time of pump station numbered i. i This represents the rated flow rate of pump station numbered i, where i is the pump station number retrieval variable, i∈N, 1≤i≤M;

[0091] The required operating time of a pump is determined by dividing the workload of each pumping station by the rated flow rate. The larger the workload, the longer the operating time; the larger the rated flow rate, the shorter the operating time.

[0092] The startup time of each pump station is input into the corresponding digital twin model of the pump station group to obtain the bearing temperature curve, pump chamber pressure curve, and flow curve of each pump station. The flow stability is obtained based on the flow curve of each pump station, using the following formula:

[0093]

[0094] Where, σ Q-i This indicates that the operating time of pump station numbered i is T. i The flow stability value at time, T i For operating time, Q i (t) represents the flow rate of pump station i at time t. This represents the average flow rate of pump station i, where t is the operating time retrieval variable, and 0 represents the average flow rate of pump station i. <t≤T i , where i is the pump station number retrieval variable, i∈N, 1≤i≤M;

[0095] Among them, the standard deviation of the flow curve is used to judge the stability of the flow rate. The flow rate stability represents the operating status of the pump. The smaller the standard deviation, the better the operating status of the pump. The larger the standard deviation, the more unstable the flow rate may be due to problems such as corrosion or sealing. These unstable factors indicate that the pump may have problems in the future operation process, so they need to be included in the operating status score.

[0096] Pressure stability is obtained based on the pressure curve of the pump compartment of each pumping station, using the following formula:

[0097]

[0098] Where, σ p-i This indicates that the operating time of pump station numbered i is T. i Pressure stability value at time, T i For the operating time, p i (t) represents the pressure at pump station i after a duration of t. This represents the average pressure of pump station numbered i, where t is the operating time retrieval variable, and 0 represents the average pressure of pump station i. <t≤T i , where i is the pump station number retrieval variable, i∈N, 1≤i≤M;

[0099] Among them, the pressure of the pump chamber is an important factor affecting the normal operation of the pump. The normal pump chamber pressure should be stable. Stability means that the pump has a strong ability to work continuously. If the standard deviation of the pump chamber pressure is large, it indicates that the pump may have problems with sealing and blockage, which may cause the scheduling process to be unsuccessful. Therefore, it needs to be included in the consideration of the operation status score.

[0100] The temperature threshold for each pump station is obtained from the supplier of the pumps. The highest temperature is determined based on the bearing temperature curve of each pump station, and an operating status score for each pump is constructed using the following formula:

[0101]

[0102] Among them, SY i σ represents the operating status score of pump station numbered i. p-i σ represents the pressure stability of pump station numbered i. Q-i T represents the flow stability of pump station numbered i. max-i This represents the highest temperature, T, at pump station numbered i. th-i Let i represent the temperature threshold of pump station i, where i is the pump station number retrieval variable, i∈N, 1≤i≤M.

[0103] The pump's operating status is determined by the difference between the highest temperature and the temperature threshold. During continuous pump operation, the bearing temperature will rise. However, under conditions of stable heat dissipation system and its own stable heat generation, the highest temperature generally will not exceed the temperature threshold. But when the pump is in an abnormal state, such as shaft eccentricity or ball bearing corrosion, the bearing temperature will continue to rise, causing it to exceed the temperature threshold. Temperature is also an important factor in judging the normal operation of the pump. Pressure stability, flow stability, and highest temperature are all included in the status score. By considering all three factors, it is possible to avoid misjudgment due to local factors. The higher the pressure stability value, the higher the operating status score; the higher the flow stability value, the higher the operating status score; and when the highest temperature exceeds the temperature threshold, the greater the excess, the more exponentially the operating status score will increase.

[0104] Based on a digital twin model, the bearing temperature, pump chamber pressure, and flow rate curves of the pumping station are predicted. By quantifying flow stability and pressure stability, the dynamic operating status is transformed into measurable scoring indicators. By comparing the temperature threshold with the maximum temperature, the risk of equipment overheating can be identified in advance, avoiding equipment damage caused by abnormal operating conditions. The operating status scoring formula integrates multi-dimensional data of flow rate, pressure, and temperature, providing a real-time operational dimension basis for the comparison of health status scores in step 4. The prediction results of this step are directly related to the short-term performance of the pumping station, providing key criteria for subsequent dynamic adjustments.

[0105] Step 4: Obtain the equipment duration, operation duration, medium repair times, and the duration since the last minor repair for each pumping station, construct a health status score, sort based on the operation status score and health status score of each pumping station, and determine whether each pumping station meets the dispatching volume according to the sorting. If there is a pumping station that does not meet the requirement, adjust the task volume of this pumping station and reallocate the task volume of other pumping stations. If there is always a pumping station that cannot meet the dispatching volume, issue an alarm.

[0106] The said Step 4 includes the following contents:

[0107] Step 401: Obtain the equipment duration, operation duration, medium repair times, and the duration since the last minor repair for each pumping station through the maintenance log. The medium repair and minor repair are the maintenance levels required for equipment maintenance. The equipment duration is the duration from the commissioning of the pump to the current time, and the operation duration is the usage duration after the pump is commissioned. Construct the health status score for each pumping station, and the basis formula is as follows:

[0108]

[0109] Among them, SJ i represents the health status score of the pumping station numbered i, T equip-i represents the equipment duration of the pumping station numbered i, T run-i represents the operation duration of the pumping station numbered i, N mid-i represents the medium repair times of the pumping station numbered i, T last-i represents the duration since the last minor repair of the pumping station numbered i. i is the pumping station number retrieval variable, i ∈ N, 1 ≤ i ≤ M, and k1 and k2 are coefficients, k1 < k2.

[0110] Among them, the equipment duration, operation duration, medium repair times, and the duration since the last minor repair all affect the health status of the pump. The longer the equipment duration, the lower the health status score; the longer the operation duration, the lower the health status score; the more medium repair times, it means the more failure times of this pump, indicating that the pump is less healthy and the health status score is lower; the longer the duration since the last minor repair, it means the pump has not been maintained for a long time and the health status score is also lower. k1 < k2 represents that the influencing factor of the operation duration is greater than the equipment duration, to avoid the influence of the health status score of some pumping stations due to long equipment time and low working intensity.

[0111] Extract the equipment duration, operation duration, medium repair times, and the duration since the last minor repair of the pumping stations from the maintenance logs, and construct a health status score in combination with the exponential decay function to quantify the long-term health status of the pumping stations. This score reflects the comprehensive impact of equipment aging, usage frequency, and maintenance history on reliability. The setting of k1 < k2 highlights that the operation duration has a much greater negative impact on the health status than the equipment duration. Step 401 provides static health dimension data for the calculation of the score difference in Step 402, which complements the dynamic operation status score in Step 3, ensuring that the scheduling decision considers both real-time performance and the long-term reliability of the equipment. At the same time, the formula design of the health score implies guidance for the maintenance strategy, indirectly optimizing the subsequent operation and maintenance management.

[0112] Step 402: Sort the health status scores and operation status scores of all pumping stations from high to low respectively. Obtain the pumping stations ranked in the last one-third of the health status score ranking, and obtain the pumping stations ranked in the last one-third of the operation status score ranking. Screen out the pumping stations that are in the last one-third in both rankings, and label them as the pumping stations that do not meet the current scheduling volume. Redistribute the scheduling volume of all pumping stations.

[0113] Sort the two scores and screen out the pumping stations that are in the last one-third in both rankings, which means that the selected pumping stations are the most unstable ones among all pumping stations during operation and need to be considered emphatically. In order to avoid unexpected situations in actual work, adjust their scheduling volume to improve their stability in the scheduling task.

[0114] Step 403: The logic of scheduling volume redistribution is as follows:

[0115] First, adjust the pumping stations that do not meet the current scheduling volume. Set its number as j, and set its adjusted scheduling volume to half of the scheduling volume before adjustment. At the same time, distribute the remaining task volume among other pumping stations. Resolve the adjusted scheduling volume of other pumping stations in MATLAB, and modify the solution condition I. The formula is as follows:

[0116]

[0117] Among them, V is the scheduled task volume, V j represents the scheduling volume of the pumping station numbered j before adjustment, V i represents the scheduling volume of the pumping station numbered i, i is the pumping station number retrieval variable, i ∈ N, 1 ≤ i ≤ M;

[0118] Optimize the pumping stations that cannot complete the current scheduling volume in the new round of task distribution, so that they only complete half of the original planned task volume, and distribute the remaining task volume among other pumping stations.

[0119] Without modifying solution conditions II and III, re-obtain the scheduling quantity for each pump station after the solution is obtained, and send the scheduling quantity to the corresponding pump station.

[0120] For substandard pump stations, their scheduling workload is halved, and the scheduling workloads of other pump stations are recalculated, while the total task constraints are updated. This step dynamically adjusts task allocation to prevent faulty pump stations from dragging down the overall task, while using MATLAB for rapid re-optimization to ensure that the remaining pump stations efficiently fill the task gaps under time and energy constraints. The reallocation logic demonstrates the system's resilience and fault tolerance, fixing local problems while maintaining the priority of the global objective. Furthermore, the adjusted scheduling workload provides new input for the re-prediction in step 3 and the iterative decision in step 4, forming a crucial link in the closed-loop feedback.

[0121] Step 404: Each pump station executes steps 3 and 4 again based on the re-acquired scheduling amount until the score difference of all pump stations is greater than zero. Then, the scheduling task is started. If it is impossible to find a solution that makes the score difference of all pump stations greater than zero, an alarm is triggered.

[0122] By iteratively executing steps 3 and 4, the scheduling quantity is continuously updated and the pump station status is reassessed until all pump stations G i A value ≥0 or an alarm is triggered. The iterative process enables incremental optimization of the scheduling scheme, ensuring that the system approaches the optimal solution amidst dynamic changes. If all conditions cannot be met ultimately, the alarm mechanism prompts manual intervention to prevent task interruption due to collective equipment failure. This step not only strengthens the system's self-healing capabilities but also feeds back problems that cannot be automatically resolved to the maintenance team, prompting equipment repair or replacement and indirectly optimizing the health status of the pump station group (feedback to the data update in step 401). The iterative logic and alarm mechanism together ensure the ultimate feasibility of scheduling while providing data support for long-term maintenance decisions.

[0123] Please refer to Figure 2 The present invention also includes a pump station group scheduling system based on digital twins, used to execute the above-described pump station group scheduling method based on digital twins, comprising:

[0124] The model building module is used to obtain the bearing temperature curve, flow rate curve, and pump chamber pressure curve of each pump station when it is running at rated power. The bearing temperature model, pump chamber pressure model, and flow rate model of each pump station are trained separately. The models of all pump stations are summarized to obtain a digital twin model of the pump station group.

[0125] The scheduling quantity solution module is used to set the scheduling quantity of each pump station based on the scheduling task quantity and task time, and to set solution conditions I, solution conditions II and solution conditions III, and then solve and obtain the scheduling quantity of each pump station in MATLAB.

[0126] The operation evaluation module is used to obtain the operating time of each pump station according to the scheduling volume and input it into the digital twin model of the pump station group. It obtains the bearing temperature curve, pump chamber pressure curve and flow curve of each pump station, obtains the flow stability based on the flow curve, obtains the pressure stability based on the pump chamber pressure curve, and obtains the operating status score of each pump station by combining the temperature threshold of each pump station.

[0127] The status analysis module is used to obtain the equipment duration, running time, number of intermediate repairs, and time since the last minor repair for each pump station, and construct a health status score. The pump stations are sorted by their running status score and health status score. Based on the sorting, it is determined whether each pump station meets the scheduling requirements. If there is a pump station that does not meet the requirements, the task of this pump station is adjusted and the task of other pump stations is redistributed. If there are always pump stations that cannot meet the scheduling requirements, an alarm is triggered.

[0128] 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.

[0129] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0130] The units described 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 can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0131] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A pump station group scheduling method based on digital twins, characterized in that, The specific steps include: Step 1: Obtain the bearing temperature curve, flow rate curve, and pump chamber pressure curve of each pump station when it is running at rated power. Train the bearing temperature model, pump chamber pressure model, and flow rate model of each pump station separately. Summarize the models of all pump stations to obtain the digital twin model of the pump station group. Step 2: Based on the scheduling workload and time, set the scheduling workload for each pump station, and set solution conditions I, II, and III, and solve them in MATLAB to obtain the scheduling workload for each pump station. Step 3: Each pump station obtains its own operating time based on the scheduling volume and inputs it into the digital twin model of the pump station group. The bearing temperature curve, pump chamber pressure curve and flow curve of each pump station are obtained. The flow stability is obtained based on the flow curve and the pressure stability is obtained based on the pump chamber pressure curve. Combined with the temperature threshold of each pump station, the operating status score of each pump station is obtained. Step 4: Obtain the equipment duration, running time, number of intermediate repairs, and time since the last minor repair for each pump station, construct a health status score, sort the pump stations by their running status score and health status score, and determine whether each pump station meets the scheduling requirements based on the sorting. If there are pump stations that do not meet the requirements, adjust the task load of this pump station and redistribute the task load to other pump stations. If there are pump stations that cannot meet the scheduling requirements, an alarm will be triggered. The health status scores and operation status scores of all pumping stations are sorted from high to low. Pumping stations ranked in the bottom three in the health status score ranking are selected, and pumping stations ranked in the bottom third in the operation status score ranking are selected. Pumping stations ranked in the bottom third in both rankings are selected and marked as pumping stations that do not meet the current scheduling volume. The scheduling volume of all pumping stations is then redistributed. The scheduling redistribution logic is as follows: First, adjust the pump stations that do not meet the current dispatch volume, and assign them the number [number]. The adjusted scheduling amount is set to half of the original scheduling amount. The remaining workload is then distributed among other pumping stations. The adjusted scheduling amounts for the other pumping stations are re-solved in MATLAB, with the solution condition I modified. The formula used is as follows: in, For the amount of tasks to be scheduled, Indicates the number is The pump station adjusted the previous scheduling volume. Indicates the number is The dispatch volume of the pumping station, Retrieve variables for pump station number. , ; Without modifying solution conditions II and III, re-obtain the scheduling amount for each pump station after the solution is obtained, and send the scheduling amount to the corresponding pump station; Each pump station executes steps 3 and 4 again based on the re-acquired scheduling data until the score difference of all pump stations is greater than zero. Then, the scheduling task is executed. If it is impossible to find a solution that makes the score difference of all pump stations greater than zero, an alarm is triggered.

2. The pump station group scheduling method based on digital twin according to claim 1, characterized in that: By using the work logs of each pumping station, the bearing temperature curve, flow rate curve, and pump chamber pressure curve of each pumping station during rated power operation in history are obtained. The horizontal axis of the curve represents the running time, with the starting point of the running time being 0 and the ending point being the actual running time of the equipment. The vertical axis represents the values ​​of each type of data. Using a single pumping station as a unit, the startup time as the training set, and the bearing temperature curve, pump chamber pressure curve, and flow rate curve corresponding to different startup times as labels, the convolutional neural network model is trained to obtain the bearing temperature model, pump chamber pressure model, and flow rate model for each pumping station. The models of all pumping stations are then aggregated to obtain a digital twin model of the pumping station group.

3. The pump station group scheduling method based on digital twin according to claim 1, characterized in that: Obtain the rated flow and rated power of each pumping station, retrieve the scheduling workload and time of the pumping station group from the superior unit, allocate the scheduling workload among the pumping stations, set the scheduling workload for each pumping station, and solve it in MATLAB. The logic for setting the solution conditions is as follows: The formula used to solve condition I is as follows: in, For the amount of tasks to be scheduled, Indicates the number is The dispatch volume of the pumping station, Retrieve variables for pump station number. , ; The formula used to solve condition II is as follows: in, This indicates selecting the maximum value. Indicates the number is The rated flow rate of the pumping station, For task time; The formula used to solve condition III is as follows: in, To minimize the function, Indicates the number is The rated power of the pumping station; Obtain the scheduling amount for each pump station after the solution is obtained.

4. The pump station group scheduling method based on digital twin according to claim 1, characterized in that: The operating time for each pumping station is determined based on the obtained dispatch volume and the rated flow rate of each station, using the following formula: in, Indicates the number is The operating time of the pumping station Indicates the number is The rated flow rate of the pumping station, Retrieve variables for pump station number. , ; The startup time of each pump station is input into the corresponding digital twin model of the pump station group to obtain the bearing temperature curve, pump chamber pressure curve, and flow curve of each pump station. The flow stability is obtained based on the flow curve of each pump station, using the following formula: in, Indicates the number is The pump station operates for a period of time of The data flow stability value at that time. For operating time, Indicates the number is The pumping station duration is Traffic flow at that time Indicates the number is The average flow rate of the pumping station, Variables for runtime are retrieved. , Retrieve variables for pump station number. , ; Pressure stability is obtained based on the pressure curve of the pump compartment of each pumping station, using the following formula: in, Indicates the number is The pump station operates for a period of time of Pressure stability values ​​at that time For operating time, Indicates the number is The pumping station duration is The pressure of time, Indicates the number is The average pressure of the pumping station, Variables for runtime are retrieved. , Retrieve variables for pump station number. , ; The temperature threshold for each pump station is obtained, and the highest temperature is obtained based on the bearing temperature curve of each pump station. An operating status score for each pump is then constructed, using the following formula: in, Indicates the number is The operational status score of the pumping station Indicates the number is The pump station operates for a period of time of Pressure stability values ​​at that time Indicates the number is The pump station operates for a period of time of The data flow stability value at that time. Indicates the number is The highest temperature of the pumping station, Indicates the number is The temperature threshold of the pumping station, Retrieve variables for pump station number. , .

5. A pump station group scheduling method based on digital twins according to claim 4, characterized in that: The equipment duration, operating time, number of intermediate repairs, and time since the last minor repair are obtained from the maintenance logs for each pumping station. The intermediate and minor repairs refer to the maintenance levels required for equipment maintenance. The equipment duration is the time from pump commissioning to the present, and the operating time is the operating time of the pump after commissioning. A health status score for each pumping station is constructed based on the following formula: in, Indicates the number is The health status score of the pumping station Indicates the number is The equipment time of the pumping station, Indicates the number is The operating time of the pumping station, Indicates the number is The number of times the pump station needs to undergo intermediate repairs. Indicates the number is The distance of the pumping station from the last minor repair time. Retrieve variables for pump station number. , , and For coefficients, .

6. A pump station group scheduling system based on digital twins, used to execute the pump station group scheduling method based on digital twins as described in any one of claims 1-5, characterized in that, include: The model building module is used to obtain the bearing temperature curve, flow rate curve, and pump chamber pressure curve of each pump station when it is running at rated power. The bearing temperature model, pump chamber pressure model, and flow rate model of each pump station are trained separately. The models of all pump stations are summarized to obtain a digital twin model of the pump station group. The scheduling quantity solution module is used to set the scheduling quantity of each pump station based on the scheduling task quantity and task time, and to set solution conditions I, solution conditions II and solution conditions III, and then solve and obtain the scheduling quantity of each pump station in MATLAB. The operation evaluation module is used to obtain the operating time of each pump station according to the scheduling volume and input it into the digital twin model of the pump station group. It obtains the bearing temperature curve, pump chamber pressure curve and flow curve of each pump station, obtains the flow stability based on the flow curve, obtains the pressure stability based on the pump chamber pressure curve, and obtains the operating status score of each pump station by combining the temperature threshold of each pump station. The status analysis module is used to obtain the equipment duration, running time, number of intermediate repairs, and time since the last minor repair for each pump station, and construct a health status score. The pump stations are sorted by their running status score and health status score. Based on the sorting, it is determined whether each pump station meets the scheduling requirements. If there is a pump station that does not meet the requirements, the task of this pump station is adjusted and the task of other pump stations is redistributed. If there are always pump stations that cannot meet the scheduling requirements, an alarm is triggered.

Images