Pumping station operation optimization method and system based on digital twin

Abstract

This application proposes a pump station operation optimization method and system based on digital twins, which relates to the field of data processing technology. The method includes: obtaining fault area data and non-fault area data of fault events that occur during the operation of the pump station; determining the upstream impact area of ​​the faulty water pump in the non-faulty water pump based on the similarity of changes and timestamps between the fault area data and the non-fault area data; determining the degree of influence of the upstream impact area on the faulty water pump based on the proportional relationship between stable paired data and fluctuating paired data matched between the fault area data and the non-fault area data in the upstream impact area; determining the fluctuation range of the faulty water pump based on the degree of influence of the faulty water pump in multiple fault events; using the fluctuation range as the dynamic constraint boundary of each parameter of the faulty water pump, constructing a digital twin model to determine the optimal parameter configuration and deploying it to the pump station. This application can achieve coordinated optimization of pump station operation efficiency and equipment stability.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a pump station operation optimization method and system based on digital twins. Background Art

[0002] In a pumping station with multiple pumps operating in parallel, each pump can be assigned its own specific flow path and flow pattern by designing an operation plan. With the development of intelligent control technology, it is possible to develop an operation control strategy for the pumping station to optimize its operation, thereby reducing energy consumption or lowering water supply costs.

[0003] Existing pump station operation optimization methods often treat the entire pump station system as a whole with identical physical characteristics and operating conditions. These methods, through homogenized optimization strategies, achieve theoretically optimal energy efficiency, such as the lowest theoretical total energy consumption. However, these methods ignore the regional characteristics of actual pump equipment operation. This can lead to overloads in vulnerable areas when implementing these strategies, impacting the overall stability of the pump station system. This is especially true for pump units that experience prolonged high loads, where internally constructed parameters may remain close to warning values. Without targeted constraints in these optimization methods, the risk of failure is significant, severely impacting the operational stability of the pump station system. Summary of the Invention

[0004] The present application aims to solve one of the technical problems in the related art at least to a certain extent.

[0005] To this end, the first purpose of this application is to propose a pump station operation optimization method based on digital twins to achieve the optimal pump station operation plan while taking into account the stability of the pump station operation.

[0006] The second purpose of this application is to propose a pump station operation optimization system based on digital twin.

[0007] To achieve the above objectives, the first embodiment of the present application proposes a pump station operation optimization method based on digital twins, including:

[0008] Acquire fault data of a fault event that occurs during operation of the pump station, wherein the fault data includes fault area data of the faulty water pump and non-fault area data of the non-faulty water pump;

[0009] Determine the upstream impact area of ​​the faulty water pump in the non-faulty water pump based on the change similarity and time stamp between the faulty area data and the non-faulty area data;

[0010] Based on the matching relationship between the fault area data and the non-fault area data in the upstream impact area, the stable paired data and the fluctuating paired data are determined, and the degree of influence of the upstream impact area on the faulty water pump is determined based on the proportional relationship between the stable paired data and the fluctuating paired data;

[0011] Determine the degree of influence of the faulty water pump based on the corresponding impact levels of each upstream impact area affecting the faulty water pump in multiple fault events, as well as the proportion of each upstream impact area within the pumping station. The impact level is used to determine the fluctuation range of the faulty water pump.

[0012] Using the fluctuation range as the dynamic constraint boundary of each parameter of the faulty water pump, a digital twin model is constructed to determine the optimal parameter configuration and deploy it to the pumping station.

[0013] As a preferred embodiment, determining the impact degree of the upstream impact area on the faulty water pump includes:

[0014] Select matching pairs with consistent change trends between the fault area data and the non-fault area data in the upstream impact area as stable paired data;

[0015] Select matching pairs with different change trends between the fault area data and the non-fault area data in the upstream impact area as fluctuation paired data;

[0016] Calculating the first change similarity between the stable paired data, and using the complementary probability of the first change similarity as the second change similarity;

[0017] The first change similarity is used as the weight of the proportion of stable paired data in the non-fault area data in the upstream impact area, and the second change similarity is used as the weight of the proportion of fluctuating paired data in the non-fault area data in the upstream impact area. A weighted sum is taken to obtain the impact degree.

[0018] As a preferred embodiment, determining the impact degree of the faulty water pump includes:

[0019] Obtain the median of multiple impact levels of any upstream impact area on any faulty water pump in each fault event as the representative impact level of any faulty water pump;

[0020] The mode of the representative impact degree of each upstream impact area on any faulty water pump is selected as the typical impact degree of any faulty water pump;

[0021] The proportion of each upstream impact area corresponding to any faulty water pump within the pumping station is used as the weight of the typical impact degree to obtain the impact degree of any faulty water pump.

[0022] As a preferred embodiment, determining the upstream impact area of ​​the faulty water pump in the non-faulty water pump includes:

[0023] Determine the affected area of ​​the faulty water pump in the non-faulty water pump based on the similarity between the data in the faulty area and the data in the non-faulty area;

[0024] When the time stamp of the non-fault area data in the affected area is earlier than the time stamp of the fault area data, the non-fault area is determined to be the upstream affected area.

[0025] As a preferred embodiment, the fault data is time series data, and determining the affected area of ​​the faulty water pump among the non-faulty water pumps includes:

[0026] generating a first change curve of the fault area data and a second change curve of the non-fault area data based on the degree of change of the fault data at adjacent moments;

[0027] A non-fault area where the similarity between the first change curve and the second change curve is greater than a similarity threshold is determined as an affected area.

[0028] As a preferred embodiment, determining the stable paired data and the fluctuating paired data includes:

[0029] Obtaining a third variation curve of the affected area;

[0030] The first change curve and the third change curve are matched using a dynamic time warping algorithm to determine stable pairing data in which one-to-one correspondence is found between the first change curve and the third change curve, as well as fluctuating pairing data in which one data in the first change curve corresponds to multiple data in the third change curve.

[0031] As a preferred embodiment, the process of obtaining the upstream impact area includes:

[0032] Obtain matching pairs consisting of non-fault area data and fault area data within the affected area;

[0033] In the matching pairs, the percentage of data in the non-fault area within the affected area whose timestamp is earlier than that of the data in the fault area;

[0034] When the proportion corresponding to the affected area exceeds a preset proportion threshold, the affected area is determined to be an upstream affected area.

[0035] As a preferred implementation, determining the optimal parameter configuration includes:

[0036] Based on dynamic constraint boundaries, the physical, behavioral, and data layers of the digital twin model are constructed to obtain simulation results of the faulty water pump in the pumping station.

[0037] When the simulation results pass the model verification, the parameters corresponding to the simulation results are used as the optimal parameter configuration.

[0038] As a preferred embodiment, the process of obtaining the fluctuation range includes:

[0039] The negative exponent with the degree of influence as the base of the natural constant is used to obtain the upper limit of the fluctuation of the faulty water pump to determine the fluctuation range.

[0040] To achieve the above objectives, the second embodiment of the present application proposes a pump station operation optimization system based on digital twins, including:

[0041] A data acquisition module is used to acquire fault data of fault events that occur during the operation of the pump station, wherein the fault data includes fault area data of the faulty water pump and non-fault area data of the non-faulty water pump;

[0042] An upstream impact area determination module is used to determine the upstream impact area of ​​the faulty water pump among the non-faulty water pumps based on the similarity of changes and timestamps between the faulty area data and the non-faulty area data;

[0043] An impact degree determination module is used to determine stable paired data and fluctuating paired data based on the matching relationship between the fault area data and the non-fault area data in the upstream impact area, and determine the impact degree of the upstream impact area on the faulty water pump based on the proportional relationship between the stable paired data and the fluctuating paired data;

[0044] An impact degree determination module is used to determine the impact degree of the faulty water pump based on the impact degrees corresponding to the upstream impact areas affecting the faulty water pump in multiple fault events, and the proportion of each upstream impact area within the pumping station. The impact degree is used to determine the fluctuation range of the faulty water pump;

[0045] The optimization module is used to build a digital twin model using the fluctuation range as the dynamic constraint boundary of each parameter of the faulty water pump to determine the optimal parameter configuration and deploy it to the pumping station.

[0046] The digital twin-based pump station operation optimization method and system provided in the present application analyzes fault data to determine the upstream impact area that may have an impact on the faulty water pump, and further determines the degree of impact of the upstream area on the faulty water pump by matching the faulty area data with the non-fault area data in the upstream impact area. The data matching relationship can reflect the impact of the upstream impact area on the data change of the faulty water pump and determine the degree of impact of the upstream impact area. By analyzing the impact levels of various fault events, the degree of impact of the faulty water pump under the coupled influence of multiple upstream impact areas is determined, reflecting the operating status of the faulty water pump, identifying the water pumps that are relatively weak or have higher risks during the operation of the pumping station, and setting an appropriate safety fluctuation range for them based on the degree of impact. In the process of establishing the digital twin model, different dynamic constraints are set for each water pump to make the model closer to the actual situation. By deploying the optimal parameter configuration simulated by the model to the actual pumping station, the stability risk of the pumping station operation is resolved. Under the theoretical optimal condition, the reliability of the actual operation can be guaranteed, and the coordinated optimization of the pumping station operation efficiency and equipment stability can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0048] Figure 1 A schematic flow chart of a pump station operation optimization method based on digital twins provided in an embodiment of the present application;

[0049] Figure 2 An example matching diagram provided in an embodiment of the present application;

[0050] Figure 3 A system block diagram of a pump station operation optimization system based on digital twins provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0052] The following describes the pump station operation optimization method and system based on digital twin according to the embodiments of the present application with reference to the accompanying drawings.

[0053] Figure 1 A flow chart of a pump station operation optimization method based on digital twins provided in an embodiment of the present application.

[0054] Existing pump station operation optimization methods often employ a unified optimization standard, failing to fully consider the differences in failure probability and tolerance across different areas within the system. This homogenized optimization strategy ignores the regional characteristics of actual equipment operation. While the resulting optimization solution theoretically achieves optimal energy efficiency, it may overload certain vulnerable areas in actual operation, thus affecting the overall stability of the system. This is particularly true for pump units that experience long-term high loads. The lack of targeted optimization constraints can easily lead to localized failures, ultimately making it difficult to ensure the long-term stability of the operation optimization method.

[0055] To address this issue, the present invention provides a pump station operation optimization method based on digital twins to achieve the optimal pump station operation plan while taking into account the stability of the pump station operation. Figure 1 As shown in Figure 1, the pump station operation optimization method based on digital twin includes the following steps:

[0056] Step 101 : Acquire fault data of a fault event that occurs during operation of a pump station, wherein the fault data includes fault area data of a faulty water pump and non-fault area data of a non-faulty water pump.

[0057] Optionally, multiple fault events may occur during the historical operation of the pump station. For each fault event, fault data related to the fault event is obtained.

[0058] Fault data can be collected from multiple sources, such as a sequence of parameters such as flow rate, pressure, and power at each moment, as fault data at each moment. To ensure data representativeness, fault data collection should cover different operating conditions (such as peak and off-peak periods), seasonal environmental conditions, and data from various typical fault modes. For example, fault data from multiple fault events can be collected on an annual basis.

[0059] It should be noted that during the operation of the pump station system, if a water pump unit fails, the water flow that originally passed through the failed water pump may be redistributed to other normally operating water pumps, which will affect the operating stability of other water pumps, or reduce the efficiency of other water pumps and increase energy consumption; or it may cause the relevant water pumps to stop running through the interlocking protection mechanism, affecting the normal operation of water supply or drainage. Therefore, the failure of a specific water pump unit is often not an isolated incident, but is directly affected by the operating status of other areas in the system. In other words, for each failure event, a comprehensive analysis should be conducted in combination with the impact of other water pumps on the failed water pump. Based on this, the fault data collected in the embodiment of the present application is data for all water pumps within the coverage area of ​​the pump station, including both the fault area data of the failed water pump and the non-fault area data of the non-fault water pump.

[0060] Optionally, to ensure data validity, data within 30 minutes before each fault event occurs may be obtained as fault data.

[0061] Step 102 : determining the upstream impact area of ​​the faulty water pump in the non-faulty water pump based on the change similarity and timestamps between the faulty area data and the non-faulty area data.

[0062] Optionally, for a faulty water pump, there are two types of associated areas within the pump station: one is the upstream impact area (i.e., the water pump units that have a direct or indirect hydraulic / electrical impact on the faulty area), and the other is the downstream impact area (i.e., the water pump units affected by the conduction effect of the faulty area). Both of these areas have a high correlation with the data change rate of the faulty water pump area. Faulty area data with similar changes and non-faulty area data may have an influencing relationship. In this embodiment of the present application, by analyzing the degree of change similarity and timestamps, it is possible to preliminarily determine the upstream impact area that affects the faulty water pump.

[0063] In an embodiment of the present application, based on the degree of similarity of changes between the fault area data and the non-fault area data, the impact area of ​​the faulty water pump in the non-faulty water pump is determined; when the timestamp of the non-fault area data in the impact area is earlier than the timestamp of the fault area data, the non-fault area is determined to be the upstream impact area.

[0064] Since fault data is time series data, the changes before and after the fault data can be reflected by the changes in the fault data on the timeline.

[0065] In an embodiment of the present application, based on the degree of change of fault data at adjacent moments, a first change curve of the fault area data and a second change curve of the non-fault area data are generated; the non-fault area where the similarity between the first change curve and the second change curve is greater than a similarity threshold is determined as the affected area.

[0066] As an example, for each pump in the pumping station, the cosine similarity between the data at the previous moment and the data at the next moment is calculated. The 1-cosine similarity is used as the degree of change at each moment to characterize the data variation. The first variation curve is formed by the degree of continuous time-series change in the data in the faulty area; the second variation curve is formed by the degree of continuous time-series change in the data in the non-faulty area.

[0067] Furthermore, the cosine similarity between the first change curve and each second change curve is calculated as the similarity between the fault area data and the non-fault area data. When the similarity is greater than a similarity threshold, the corresponding non-fault area is determined to be the affected area of ​​the fault area.

[0068] As an example, the similarity threshold may be set to 0.5. In other embodiments, it may be set according to actual conditions and is not limited here.

[0069] It is understandable that the data changes in the affected area are similar to those of the faulty water pump, and there is a high probability that the pump affects the faulty water pump or is affected by the faulty water pump, that is, the upstream affected area or the downstream affected area.

[0070] As a possible implementation, when the timestamp of the data in the non-fault area in the impacted area is earlier than the timestamp of the data in the faulty area, the non-faulty area is determined to be the upstream impacted area.

[0071] As another possible implementation, matching pairs consisting of non-fault area data and fault area data within the impacted area are obtained; in the matching pairs, the proportion of the non-fault area data within the impacted area whose timestamp is earlier than the timestamp of the fault area data is obtained; when the proportion corresponding to the impacted area exceeds a preset proportion threshold, the impacted area is determined to be an upstream impacted area.

[0072] Among them, matching pairs can be obtained through data processing methods that can achieve curve matching, such as dynamic time warping algorithm (DTW), nonlinear fitting matching method or machine learning mapping.

[0073] As an example, the preset proportion threshold may be set to 0.7. In other embodiments, it may be set to other values ​​according to actual conditions, which is not limited here.

[0074] By setting the ratio, data changes with most timestamps earlier than the faulty water pump can be taken into account, covering as many other water pumps that may have an impact on the faulty water pump as possible.

[0075] It is understandable that the changes in the non-fault area earlier than the faulty water pump are likely to be one of the factors affecting the faulty water pump. During the operation of the pumping station, this influencing factor needs to be taken into account and recorded as the upstream influencing area.

[0076] Step 103 : Based on the matching relationship between the fault area data and the non-fault area data in the upstream impact area, stable paired data and fluctuating paired data are determined, and based on the proportional relationship between the stable paired data and the fluctuating paired data, the degree of influence of the upstream impact area on the faulty water pump is determined.

[0077] Optionally, different upstream influencing areas may have different impacts on the faulty area. Since a faulty pump may be affected by multiple upstream influencing areas simultaneously, the closer the change in the faulty area's data is to that of a specific upstream influencing area, the greater the impact. Therefore, the impact of upstream influencing areas on the faulty pump is comprehensively considered based on stable paired data with consistent change trends and fluctuating paired data with different change trends.

[0078] In an embodiment of the present application, matching pairs with consistent change trends between the fault area data and the non-fault area data in the upstream impact area are selected as stable matching data; matching pairs with different change trends between the fault area data and the non-fault area data in the upstream impact area are selected as fluctuating matching data.

[0079] In different matching methods, the representations of consistent change trends and differences are different. Taking the method of obtaining matching pairs using the DTW algorithm as an example, the third change curve of the affected area is obtained, wherein the third change curve is used to characterize the degree of change of the fault data in the affected area at adjacent moments; the first change curve and the third change curve are matched using the dynamic time warping algorithm to determine the stable paired data in the first change curve that corresponds one-to-one with the third change curve, as well as the fluctuating paired data in which one data in the first change curve corresponds to multiple data in the third change curve.

[0080] like Figure 2 In the matching example diagram shown, the upper solid line is the third change curve, the lower solid line is the first change curve, and each dotted line connects a matching pair. In DTW, if the correspondence between the two curves is one-to-one when matching, that is, the dotted line is vertical, indicating that the change trends of the two data are consistent, and the one-to-one matching pair is determined to be stable paired data. In the case of a one-to-many correspondence between the data, that is, the dotted line is tilted, indicating that some data have different change trends, and the multiple matching pairs with one-to-many matching are determined to be fluctuating paired data.

[0081] In other embodiments, some accidental fluctuation data may be eliminated.

[0082] As an example, for DTW matching of any third change curve with the first change curve, all matching point pairs are obtained based on the sequential order of the time points on the third change curve. The time difference between the point in the upstream influence area and the point in the downstream influence area in each matching pair is calculated (upstream point time minus downstream point time) to obtain a time difference sequence. Using a sliding window method, multiple consecutive, identical, and integer time periods are obtained. Time periods with fewer than five elements are screened out, as they are likely to be accidental fluctuations and are less representative, and are therefore discarded. The resulting matching pairs are recorded as stable paired data.

[0083] For any DTW match between the third change curve and the first change curve, a greater proportion of continuous one-to-one matches indicates a greater similarity in trends and a greater degree of impact. For non-continuous one-to-one matches, a greater proportion of upstream one-to-one and one-to-many matches indicates that changes upstream can cause greater changes downstream, and thus a greater degree of impact.

[0084] Therefore, the first change similarity between the stable paired data is calculated, and the complementary probability of the first change similarity is used as the second change similarity; the first change similarity is used as the weight of the proportion of the stable paired data in the non-fault area data in the upstream impact area, and the second change similarity is used as the weight of the proportion of the fluctuating paired data in the non-fault area data in the upstream impact area, and a weighted sum is performed to obtain the degree of influence.

[0085] In the embodiment of the present application, the first change similarity can be obtained by calculating the cosine similarity. In other embodiments, it can also be calculated by other similarity calculation methods. The value range of the first change similarity is [0, 1], and the second change similarity = 1-first change similarity.

[0086] It can be understood that the greater the degree of impact, the greater the impact of the data change in the upstream impact area on the faulty water pump.

[0087] Step 104, determining the degree of influence of the faulty water pump based on the corresponding influence degrees of each upstream influence area affecting the faulty water pump in multiple fault events, and the proportion of each upstream influence area within the pumping station, wherein the degree of influence is used to determine the fluctuation range of the faulty water pump.

[0088] Optionally, any faulty water pump may have acted as a faulty water pump in multiple fault events during its historical operation. By obtaining the degree of impact of the faulty water pump in each fault event, the degree of impact of the faulty water pump on each upstream affected area can be comprehensively determined.

[0089] As an example, the median of multiple impact levels of any upstream influence area on any faulty water pump in each fault event is obtained as the representative impact level of any faulty water pump; the mode of the representative impact levels of each upstream influence area on any faulty water pump is selected as the typical impact level of any faulty water pump; the proportion of each upstream influence area corresponding to any faulty water pump within the pumping station range is used as the weight of the typical impact level to obtain the impact level of any faulty water pump.

[0090] For any faulty water pump, the more upstream influence areas that affect the faulty water pump, the greater the influence of multiple upstream influence areas on the faulty water pump, and the greater the degree of influence on the faulty water pump, that is, the more likely the faulty water pump is to be affected by other water pumps and fail.

[0091] Furthermore, the greater the degree of impact on the faulty water pump, the more likely changes in other water pumps are to cause significant changes in the data of the faulty water pump. In other words, the lower the tolerance of the faulty water pump, the lower the fluctuation range needs to be set for the corresponding water pump.

[0092] As an example, the degree of influence is used as a negative exponent of a natural constant to calculate the upper limit of the fluctuation of the faulty water pump, and then determine the fluctuation range, that is, [0, ] as the fluctuation range, where v represents the degree of influence.

[0093] For example, for a faulty water pump, the maximum fluctuation value calculated through the above steps is 0.02. Then [0, 0.02] is used as the fluctuation range of various parameters of the faulty water pump, such as speed fluctuation rate ≤ 2% / s, power fluctuation rate ≤ 2% / s, etc. That is, during the pump station operation optimization process, if different operating modes are adjusted according to actual conditions, it is necessary to ensure that the changes in various parameters of any water pump do not exceed the corresponding fluctuation range.

[0094] In step 105 , a digital twin model is constructed using the fluctuation range as the dynamic constraint boundary of each parameter of the faulty water pump to determine the optimal parameter configuration and deploy it to the pumping station.

[0095] Optionally, by determining the fluctuation range, a safety threshold range for the faulty water pump can be set to construct a digital twin model.

[0096] In an embodiment of the present application, the physical layer, behavioral layer, and data layer of the digital twin model are constructed based on the dynamic constraint boundaries to obtain the simulation results of the faulty water pump in the pumping station; when the simulation results pass the model verification, the parameters corresponding to the simulation results are used as the optimal parameter configuration.

[0097] The digital twin model construction employs a three-tiered architecture: the physical layer integrates the CFD fluid dynamics model and pump characteristic curves; the behavioral layer embeds a dynamic constraint rule engine and parameter cross-boundary propagation algorithm; and the data layer establishes a knowledge graph containing the influence feature vectors. Model validation is achieved through virtual-to-real mapping calibration and constraint sensitivity analysis. An online learning framework is designed to implement sliding window updates of the influence index and adaptive adjustment of the constraint range.

[0098] By constructing a digital twin model, the probabilistic degree of influence is converted into dynamic parameter constraints, so that the digital twin model has risk adaptive characteristics, and the output of the obtained digital twin model will be closer to the actual working conditions.

[0099] Furthermore, the constructed digital twin model achieves safe and efficient operation optimization by building a complete preview evaluation closed loop in a virtual environment.

[0100] First, various extreme operating conditions are simulated on the digital twin platform, and the response of key system parameters is monitored in real time to ensure that all operating constraints are strictly met. At the same time, multi-physics field coupling simulation is used to accurately evaluate the energy efficiency improvement potential under different optimization schemes. Finally, after fully verifying the feasibility of the scheme in a virtual environment, the verified optimal parameter configuration is deployed to the actual pump station.

[0101] The embodiment of the present application avoids the risks of direct on-site testing through a digital rehearsal process, and realizes closed-loop optimization from simulation verification to actual application. While ensuring the safe and stable operation of the system, it significantly improves the overall operating efficiency and reduces the risk of unplanned downtime.

[0102] In this embodiment, fault data of a fault event occurring during the operation of a pumping station is obtained, wherein the fault data includes the fault area data of the faulty pump and the non-fault area data of non-faulty pumps. Based on the similarity and timestamp of changes between the fault area data and the non-fault area data, the upstream impact area of ​​the faulty pump in the non-faulty pumps is determined. Based on the matching relationship between the fault area data and the non-fault area data in the upstream impact area, stable paired data and fluctuating paired data are determined. The degree of influence of the upstream impact area on the faulty pump is determined based on the proportional relationship between the stable paired data and the fluctuating paired data. The degree of influence of the faulty pump is determined based on the corresponding influence degree of each upstream impact area affecting the faulty pump in multiple fault events and the proportion of each upstream impact area within the pumping station. The degree of influence is used to determine the fluctuation range of the faulty pump. Using the fluctuation range as the dynamic constraint boundary of each parameter of the faulty pump, a digital twin model is constructed to determine the optimal parameter configuration and deploy it to the pumping station. This method can ensure the reliability of actual operation under the theoretical optimality and achieve the coordinated optimization of pumping station operation efficiency and equipment stability.

[0103] In order to implement the above embodiments, the present application also proposes a pump station operation optimization system based on digital twins.

[0104] Figure 3 A system block diagram of a pump station operation optimization system based on digital twins provided in an embodiment of the present application.

[0105] like Figure 3 As shown, the pump station operation optimization system based on digital twin includes: a data acquisition module 100, an upstream impact area determination module 200, an impact degree determination module 300, an impact degree determination module 400 and an optimization module 500.

[0106] The data acquisition module 100 is used to acquire fault data of a fault event that occurs during the operation of the pump station, wherein the fault data includes fault area data of the faulty water pump and non-fault area data of the non-faulty water pump;

[0107] An upstream impact region determination module 200 is configured to determine an upstream impact region of the faulty water pump among non-faulty water pumps based on a change similarity and a timestamp between the faulty area data and the non-faulty area data;

[0108] An impact degree determination module 300 is configured to determine stable paired data and fluctuating paired data based on a matching relationship between the faulty area data and the non-faulty area data in the upstream impact area, and determine the impact degree of the upstream impact area on the faulty water pump based on a proportional relationship between the stable paired data and the fluctuating paired data;

[0109] An impact degree determination module 400 is configured to determine the impact degree of the faulty water pump based on the impact degrees corresponding to the upstream impact areas affecting the faulty water pump during multiple fault events, and the proportion of each upstream impact area within the pumping station. The impact degree is used to determine the fluctuation range of the faulty water pump.

[0110] The optimization module 500 is used to construct a digital twin model using the fluctuation range as the dynamic constraint boundary of each parameter of the faulty water pump to determine the optimal parameter configuration and deploy it to the pump station.

[0111] It should be noted that the above explanation of the embodiment of the pump station operation optimization method based on digital twin is also applicable to the pump station operation optimization system based on digital twin of this embodiment, and will not be repeated here.

[0112] In the descriptions of the foregoing embodiments, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are mutually inconsistent.

[0113] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0114] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0115] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0116] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A pump station operation optimization method based on digital twin, characterized in that: The following steps are involved: Acquire fault data of a fault event that occurs during operation of the pump station, wherein the fault data includes fault area data of the faulty water pump and non-fault area data of the non-faulty water pump; determining an upstream impact area of ​​the faulty water pump in the non-faulty water pump based on a degree of similarity of changes and a timestamp between the faulty area data and the non-faulty area data; Determining stable paired data and fluctuating paired data based on a matching relationship between the faulty area data and the non-faulty area data in the upstream impact area, and determining the degree of influence of the upstream impact area on the faulty water pump according to a proportional relationship between the stable paired data and the fluctuating paired data; Determining the degree of influence of the faulty water pump according to the corresponding influence degrees of each upstream influence area affecting the faulty water pump in multiple fault events and the proportion of each upstream influence area within the pumping station, wherein the degree of influence is used to determine the fluctuation range of the faulty water pump; Using the fluctuation range as the dynamic constraint boundary of each parameter of the faulty water pump, a digital twin model is constructed to determine the optimal parameter configuration and deploy it to the pumping station.

2. The pump station operation optimization method based on digital twin according to claim 1 is characterized in that: Determining the degree of influence of the upstream impact area on the faulty water pump includes: Selecting matching pairs with consistent change trends between the fault area data and the non-fault area data in the upstream impact area as the stable paired data; Selecting matching pairs with different change trends between the fault area data and the non-fault area data in the upstream impact area as the fluctuation paired data; Calculating a first change similarity between the stable paired data, and using the complementary probability of the first change similarity as a second change similarity; The first change similarity is used as the weight of the proportion of the stable paired data in the non-fault area data in the upstream impact area, and the second change similarity is used as the weight of the proportion of the fluctuating paired data in the non-fault area data in the upstream impact area. A weighted sum is performed to obtain the impact degree.

3. The pump station operation optimization method based on digital twin according to claim 1 is characterized in that: Determining the degree of impact of the faulty water pump includes: Obtaining the median of multiple impact levels of any upstream impact area on any faulty water pump in each fault event as the representative impact level of the faulty water pump; Selecting the mode of the representative impact degree of each upstream impact area on any one of the faulty water pumps as the typical impact degree of any one of the faulty water pumps; The proportion of each upstream impact area corresponding to any faulty water pump within the pump station is used as the weight of the typical impact degree to obtain the impact degree of any faulty water pump.

4. The pump station operation optimization method based on digital twin according to claim 1 is characterized in that: Determining the upstream impact area of ​​the faulty water pump in the non-faulty water pump includes: determining an impact area of ​​the faulty water pump in the non-faulty water pump based on a degree of similarity between the faulty area data and the non-faulty area data; When the timestamp of the non-fault area data in the affected area is earlier than the timestamp of the fault area data, the non-fault area is determined to be the upstream affected area.

5. The pump station operation optimization method based on digital twin according to claim 4 is characterized in that: The fault data is time series data, and determining the affected area of ​​the faulty water pump in the non-faulty water pump includes: generating a first change curve of the fault area data and a second change curve of the non-fault area data based on the degree of change of the fault data at adjacent moments; A non-fault area where the similarity between the first change curve and the second change curve is greater than a similarity threshold is determined as the affected area.

6. The pump station operation optimization method based on digital twin according to claim 5 is characterized in that: The determining of stable paired data and fluctuating paired data includes: Acquire a third change curve of the affected area, wherein the third change curve is used to represent a degree of change of fault data of the affected area at adjacent moments; The first change curve and the third change curve are matched using a dynamic time warping algorithm to determine stable pairing data in the first change curve that corresponds one-to-one with data in the third change curve, as well as fluctuating pairing data in which one data in the first change curve corresponds to multiple data in the third change curve.

7. The pump station operation optimization method based on digital twin according to claim 4 is characterized in that: The process of obtaining the upstream impact area includes: Acquire a matching pair consisting of non-fault area data and the fault area data in the affected area; Among the matching pairs, obtaining a ratio of the timestamp of the non-fault area data in the affected area that is earlier than the timestamp of the fault area data; When the proportion corresponding to the affected area exceeds a preset proportion threshold, the affected area is determined to be the upstream affected area.

8. The pump station operation optimization method based on digital twin according to claim 1 is characterized in that: Determining the optimal parameter configuration includes: Based on the dynamic constraint boundary, construct the physical layer, behavioral layer, and data layer of the digital twin model to obtain simulation results of the faulty water pump in the pumping station; When the simulation result passes the model verification, the parameters corresponding to the simulation result are used as the optimal parameter configuration.

9. The pump station operation optimization method based on digital twin according to claim 1 is characterized in that: The process of obtaining the fluctuation range includes: The fluctuation upper limit of the faulty water pump is obtained by taking the degree of influence as a negative exponent with a natural constant as a base, so as to determine the fluctuation range.

10. A pump station operation optimization system based on digital twin, characterized in that: include: A data acquisition module is used to acquire fault data of a fault event that occurs during the operation of the pump station, wherein the fault data includes fault area data of the faulty water pump and non-fault area data of the non-faulty water pump; an upstream impact area determination module, configured to determine an upstream impact area of ​​the faulty water pump in the non-faulty water pump based on a degree of similarity of changes and a timestamp between the faulty area data and the non-faulty area data; an impact degree determination module, configured to determine stable paired data and fluctuating paired data based on a matching relationship between the faulty area data and the non-faulty area data in the upstream impact area, and determine the impact degree of the upstream impact area on the faulty water pump according to a proportional relationship between the stable paired data and the fluctuating paired data; an impact degree determination module, configured to determine the impact degree of the faulty water pump based on the impact degrees corresponding to the upstream impact areas affecting the faulty water pump in multiple fault events, and the proportion of the upstream impact areas within the pumping station, wherein the impact degree is used to determine the fluctuation range of the faulty water pump; The optimization module is used to construct a digital twin model using the fluctuation range as the dynamic constraint boundary of each parameter of the faulty water pump to determine the optimal parameter configuration and deploy it to the pump station.

Images