Pump station operation optimization method and system based on digital twinning

Abstract

The present application belongs to the technical field of data processing, and discloses a pump station operation optimization method and system based on digital twinning, which comprises the following steps: obtaining equipment parameters of each water pump, constructing a pump station digital twinning model based on the equipment parameters, and calculating cumulative total water supply, system total energy consumption and working flow of each water pump in each time period; then, establishing an adaptation degree mapping model based on historical data of each water pump, and outputting an operation adaptation degree value according to the speed and flow; obtaining a water supply task, and constructing a space-time speed matrix; taking the space-time speed matrix as an optimization variable, constructing an optimization target based on the digital twinning model and the adaptation degree mapping model, and solving the optimization operation strategy combined with the constraint condition. Through the adaptation degree mapping model, the individual differences of each water pump are perceived, the optimization result takes into account the energy consumption and the equipment operation state, each water pump is matched with a suitable operation interval, and the service life of the equipment is prolonged.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically, to a method and system for optimizing pump station operation based on digital twins. Background Technology

[0002] Pumping stations are crucial infrastructure in urban water supply, agricultural irrigation, and industrial circulating water systems, undertaking the critical task of water resource transportation and allocation. With the maturity and widespread adoption of variable frequency drive (VFD) technology, VFD pumps have become the mainstream configuration in modern pumping stations. Compared to traditional fixed-frequency pumps, VFD pumps can flexibly adjust their speed according to actual needs, effectively reducing energy consumption while meeting water supply requirements. Pumping station energy consumption is a major energy-consuming aspect of the water supply industry. Therefore, how to fully leverage the adjustment advantages of VFD pumps to achieve efficient and energy-saving operation of pumping stations has become a key concern in the industry.

[0003] Currently, a relatively mature technical system has been established for the operation optimization of variable frequency pumping stations. Mainstream methods typically establish hydraulic and energy consumption models for the pumping station, aiming to minimize total system energy consumption, and employ intelligent optimization algorithms such as genetic algorithms and particle swarm optimization to solve for the optimal pump speed combination. However, this optimization method usually solves for a constant set of optimal speeds for a given water supply demand, with each pump maintaining a fixed speed throughout the entire operating cycle. While this approach is computationally simple, it lacks dynamic planning capabilities for the operating process, making it impossible to flexibly adjust operating strategies according to actual conditions and fully exploit the energy-saving potential of variable frequency pumps through flexible speed regulation. Secondly, existing methods do not consider the individual differences between pumps during optimization. In actual operation, even pumps of the same model will exhibit performance differences after a period of use due to factors such as wear and maintenance conditions, resulting in variations in efficiency, vibration, and temperature under the same operating conditions. Certain operating conditions unfavorable to specific pumps may exacerbate their damage, and existing methods lack the ability to perceive and adapt to these individual differences, making it difficult to match the most suitable operating range for each pump. Summary of the Invention

[0004] To overcome the aforementioned problems of the prior art, this invention proposes a pump station operation optimization method and system based on digital twins to solve the above problems.

[0005] This invention provides the following technical solution:

[0006] The pump station operation optimization method based on digital twins includes:

[0007] The equipment parameters of each water pump are obtained, and a digital twin model of the pumping station is constructed based on the equipment parameters. The digital twin model of the pumping station is used to calculate the cumulative total water supply, the total energy consumption of the system, and the working flow rate of each water pump in each time period according to the rotation speed of each water pump in each time period.

[0008] A fit mapping model for each water pump is established based on historical data of each water pump. The fit mapping model is used to output the operating fit value for a given period based on the rotational speed and operating flow rate of that period.

[0009] Obtain the water supply task, which includes the target total water supply volume and the planned total duration; construct a spatiotemporal rotation speed matrix, which is used to describe the rotation speed of each water pump in each time period;

[0010] Using the spatiotemporal rotational speed matrix as the optimization variable, an optimization objective is constructed based on the pump station digital twin model and the fit mapping model. Constraints are set in combination with water supply tasks and equipment parameters, and the optimal values ​​of each element in the spatiotemporal rotational speed matrix are obtained by solving the problem, thereby generating the corresponding optimized operation strategy.

[0011] Preferably, the digital twin model of the pumping station includes a basic model of equipment parameters, a single-period hydraulic response model, a single-period energy consumption calculation model, and a multi-period cumulative model;

[0012] The equipment parameter basic model is used to store the equipment parameters of each water pump;

[0013] The single-time hydraulic response model is constructed based on the head-flow curves of each pump, the pipeline resistance coefficient, and the design head. It is used to output the working flow of each pump and the total system flow during a single time period, according to the rotational speed of each pump in that time period.

[0014] The single-period energy consumption calculation model is constructed based on the efficiency-flow curves of each water pump, and is used to output the system power for that period based on the speed and operating flow of each water pump in a single period.

[0015] The multi-period cumulative model is used to multiply the total system flow rate of each period by the corresponding period duration and then sum them to obtain the cumulative total water supply, and to multiply the system power of each period by the corresponding period duration and then sum them to obtain the total system energy consumption.

[0016] Preferably, the single-time hydraulic response model includes:

[0017] Operating status determination steps: Determine the operating status of each water pump based on its rotation speed during the time period, and identify the water pumps with a rotation speed greater than zero as the operating water pumps during the time period;

[0018] Single pump hydraulic transformation steps: Transform the reference head-flow curve at rated speed according to the actual speed of each operating pump to generate the real-time head-flow characteristic curve of each operating pump at the current speed.

[0019] System parallel solution steps: Superimpose the real-time head-flow characteristic curves of each operating pump, calculate the intersection of the superimposed system parallel characteristic curve and the pipeline resistance curve constructed based on the pipeline resistance coefficient and design head, and determine the system operating head and total system flow rate for this period.

[0020] Flow reversal allocation step: Based on the determined system operating head, reverse the lookup in the real-time head-flow characteristic curve of each operating pump to obtain the operating flow rate of each operating pump during that time period.

[0021] Preferably, the single-period energy consumption calculation model includes:

[0022] Efficiency characteristic transformation steps: Based on the principle that the efficiency is equal at similar operating points, the baseline efficiency-flow curve at rated speed is transformed according to the actual speed to construct the real-time efficiency-flow characteristic curve of each operating pump at the current speed.

[0023] Efficiency query steps: Map the working flow rate of each operating pump output by the single-time hydraulic response model to the real-time efficiency flow characteristic curve to obtain the actual operating efficiency of each operating pump under the current operating conditions.

[0024] Power calculation steps: Calculate the shaft power of each operating pump based on its working flow rate, system operating head, and actual operating efficiency, and sum the shaft power of each operating pump to obtain the system power for that period.

[0025] Preferably, the steps for establishing the fitness mapping model include:

[0026] Extract the rotation speed, working flow rate, and adaptation status index sequence for each historical time period;

[0027] Statistical features are extracted and comprehensively evaluated for the adaptation status index sequence of each historical period, and the operational adaptation value of that historical period is calculated.

[0028] Construct a training sample set with time-period rotation speed and time-period working flow as input features and operation adaptability value as output label;

[0029] Regression modeling is performed on the training sample set to establish a mapping relationship between time period rotation speed and time period working flow and operation adaptability value, thus obtaining the adaptability mapping model.

[0030] Preferably, the adaptation state index sequence includes efficiency values, vibration values, and temperature values ​​at multiple sampling times;

[0031] The statistical feature extraction and comprehensive evaluation of the adaptation status index sequence for each historical period, and the calculation of the operational adaptation value for that historical period, include:

[0032] Calculate the mean and variance of each fitness state index sequence;

[0033] The deviation rate of an indicator is calculated based on the series mean of each indicator. The deviation rate represents the degree of deviation of the indicator from the normal value.

[0034] The stability of an indicator is calculated based on the series variance of each indicator, and the stability characterizes the degree of fluctuation of the indicator within a time period.

[0035] After normalizing the deviation rate and stability of each indicator, a weighted fusion is performed to obtain the operational adaptability value.

[0036] Preferably, the step of constructing the optimization objective based on the pump station digital twin model and the fit mapping model includes:

[0037] Input the spatiotemporal rotation speed matrix into the digital twin model of the pump station to obtain the total energy consumption of the system and the working flow rate of each pump in each time period;

[0038] Input the speed and flow rate of each water pump at each time period into the adaptation mapping model to obtain the operating adaptation value of each water pump at each time period.

[0039] The system overall fit is obtained by time-weighted summation of the operational fit values ​​of each pump at each time period. The system misfit is obtained by taking the negative value of the system overall fit.

[0040] The total energy consumption and system misfit of the system are normalized to obtain normalized energy consumption value and normalized misfit value;

[0041] The optimization objective is to minimize the weighted sum of the normalized energy consumption value and the normalized misfit value.

[0042] Preferably, the constraints include:

[0043] The deviation between the cumulative total water supply and the target total water supply is within the preset range; the sum of the time periods does not exceed the planned total duration; the length of each time period is not less than the minimum time period granularity; the speed of each water pump is zero or within its allowable speed range in each time period; the working flow rate of each water pump in each time period is within its allowable flow rate range; and the length of each time period is not less than the minimum time period granularity.

[0044] This invention also provides a pump station operation optimization system based on digital twins, used to implement a pump station operation optimization method based on digital twins, including:

[0045] The digital twin module is used to obtain the equipment parameters of each water pump and build a digital twin model of the pumping station based on the equipment parameters. The digital twin model of the pumping station is used to calculate the cumulative total water supply, the total energy consumption of the system, and the working flow rate of each water pump in each time period according to the rotation speed of each water pump in each time period.

[0046] The adaptability module is used to establish an adaptability mapping model for each water pump based on the historical data of each water pump. The adaptability mapping model is used to output the operating adaptability value for a given period based on the rotational speed and operating flow rate of that period.

[0047] The optimization preparation module is used to obtain the water supply task, which includes the target total water supply and the planned total duration; and to construct a spatiotemporal rotation speed matrix, which is used to describe the rotation speed of each water pump in each time period.

[0048] The optimization solution module is used to construct optimization objectives based on the spatiotemporal rotation speed matrix as the optimization variable, the pump station digital twin model and the fitness mapping model, and set constraints in combination with water supply tasks and equipment parameters to solve for the optimal values ​​of each element in the spatiotemporal rotation speed matrix and generate the corresponding optimized operation strategy.

[0049] This invention provides a method and system for optimizing pump station operation based on digital twins, which has the following beneficial effects:

[0050] By constructing a spatiotemporal speed matrix as an optimization variable, the speed of each pump in each time period is incorporated into a unified optimization framework, realizing dynamic planning for the entire operating cycle. Unlike traditional methods for solving constant speed, this method can flexibly plan the speed allocation strategy for each time period according to the overall requirements of the water supply task, fully tapping the energy-saving potential of variable frequency pumps through flexible speed adjustment.

[0051] By constructing a fit mapping model, the system learns the efficiency, vibration, and temperature performance of each pump under different operating conditions based on historical operating data. This establishes a mapping relationship between speed and flow rate and operational fit, enabling the perception and quantification of individual pump differences. During optimization, fit is incorporated into the optimization objective, ensuring that the optimization result not only pursues minimum energy consumption but also considers the operating status of each pump, avoiding prolonged operation under unfavorable conditions. This approach matches each pump with the most suitable operating range, mitigating equipment wear and performance degradation, extending equipment lifespan, and reducing maintenance costs. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating the pump station operation optimization method based on digital twins of the present invention.

[0053] Figure 2 This is a schematic diagram of the modules of the pump station operation optimization system based on digital twins of the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] Example 1

[0056] Please see Figure 1 In this embodiment, the pump station operation optimization method based on digital twins includes:

[0057] S1. Obtain the equipment parameters of each water pump, and construct a digital twin model of the pump station based on the equipment parameters. The digital twin model of the pump station is used to calculate the cumulative total water supply, the total energy consumption of the system, and the working flow rate of each water pump in each time period according to the rotation speed of each water pump in each time period.

[0058] The digital twin model of the pumping station includes a basic model of equipment parameters, a single-period hydraulic response model, a single-period energy consumption calculation model, and a multi-period cumulative model;

[0059] The equipment parameter basic model is used to store the equipment parameters of each water pump;

[0060] The single-time hydraulic response model is constructed based on the head-flow curves of each pump, the pipeline resistance coefficient, and the design head. It is used to output the working flow of each pump and the total system flow during a single time period, according to the rotational speed of each pump in that time period.

[0061] The single-period energy consumption calculation model is constructed based on the efficiency-flow curves of each water pump, and is used to output the system power for that period based on the speed and operating flow of each water pump in a single period.

[0062] The multi-period cumulative model is used to multiply the total system flow rate of each period by the corresponding period duration and then sum them to obtain the cumulative total water supply, and to multiply the system power of each period by the corresponding period duration and then sum them to obtain the total system energy consumption.

[0063] The single-time hydraulic response model includes:

[0064] Operating status determination steps: Determine the operating status of each water pump based on its rotation speed during the time period, and identify the water pumps with a rotation speed greater than zero as the operating water pumps during the time period;

[0065] Single pump hydraulic transformation steps: Transform the reference head-flow curve at rated speed according to the actual speed of each operating pump to generate the real-time head-flow characteristic curve of each operating pump at the current speed.

[0066] System parallel solution steps: Superimpose the real-time head-flow characteristic curves of each operating pump, calculate the intersection of the superimposed system parallel characteristic curve and the pipeline resistance curve constructed based on the pipeline resistance coefficient and design head, and determine the system operating head and total system flow rate for this period.

[0067] Flow reversal allocation step: Based on the determined system operating head, reverse the lookup in the real-time head-flow characteristic curve of each operating pump to obtain the operating flow rate of each operating pump during that time period.

[0068] The single-period energy consumption calculation model includes:

[0069] Efficiency characteristic transformation steps: Based on the principle that the efficiency is equal at similar operating points, the baseline efficiency-flow curve at rated speed is transformed according to the actual speed to construct the real-time efficiency-flow characteristic curve of each operating pump at the current speed.

[0070] Efficiency query steps: Map the working flow rate of each operating pump output by the single-time hydraulic response model to the real-time efficiency flow characteristic curve to obtain the actual operating efficiency of each operating pump under the current operating conditions.

[0071] Power calculation steps: Calculate the shaft power of each operating pump based on its working flow rate, system operating head, and actual operating efficiency, and sum the shaft power of each operating pump to obtain the system power for that period.

[0072] In this embodiment, it should be noted that equipment parameters refer to technical parameters related to the operating characteristics of the water pump and the system configuration. These typically include the head-flow rate curve, efficiency-flow rate curve, rated speed, allowable speed range, allowable flow rate range, as well as the pipeline resistance coefficient, design head, and minimum time-limit particle size for each water pump. The head-flow rate curve describes the head that the water pump can provide at different flow rates; the efficiency-flow rate curve describes the operating efficiency of the water pump at different flow rates. These parameters can be obtained from product manuals provided by the water pump manufacturer or from on-site measurements.

[0073] It should be noted that in the single-pump hydraulic transformation step, the reference head-flow rate curve at rated speed is transformed according to the actual speed, based on the pump similarity law. According to the pump similarity law, when the speed changes, the flow rate is directly proportional to the speed, and the head is directly proportional to the square of the speed. Through this transformation, the real-time head-flow rate characteristic curve of each operating pump at the current speed can be obtained.

[0074] It should be noted that the flow superposition in the parallel system solution step refers to adding the flow rates of each operating pump at the same head. Specifically, for multiple pumps operating in parallel, at the same head value, the total system flow rate is equal to the algebraic sum of the flow rates of each pump at that head. By repeating this process for different head values, the parallel system characteristic curve can be obtained. The pipe network resistance curve, constructed based on the pipe network resistance coefficient and the design head, describes the head required by the pipe network system at different flow rates. The design head corresponds to the static head portion, and the pipe network resistance coefficient determines the increase in head as the flow rate increases. The system operating point is the intersection of the system parallel characteristic curve and the pipe network resistance curve. The flow rate corresponding to this intersection point is the total system flow rate, and the corresponding head is the system operating head.

[0075] It should be noted that the efficiency characteristic transformation step, based on the principle of equal efficiency at similar operating points, means that the pump's operating efficiency remains constant under similar operating conditions. Therefore, the efficiency-flow curve at the actual speed can be obtained by scaling the efficiency-flow curve at the rated speed along the flow axis proportionally to the speed.

[0076] It should be noted that the shaft power of each operating pump in the power calculation step is calculated based on the pump's operating flow rate, system operating head, and actual operating efficiency, using the general method for pump power calculation.

[0077] S2. Based on the historical data of each water pump, establish a compatibility mapping model for each water pump. The compatibility mapping model is used to output the operating compatibility value for a given period based on the rotational speed and operating flow rate of that period.

[0078] The steps for establishing the fitness mapping model include:

[0079] Extract the rotation speed, working flow rate, and adaptation status index sequence for each historical time period;

[0080] Statistical features are extracted and comprehensively evaluated for the adaptation status index sequence of each historical period, and the operational adaptation value of that historical period is calculated.

[0081] Construct a training sample set with time-period rotation speed and time-period working flow as input features and operation adaptability value as output label;

[0082] Regression modeling is performed on the training sample set to establish a mapping relationship between time period rotation speed and time period working flow and operation adaptability value, thus obtaining the adaptability mapping model.

[0083] The adaptation state index sequence includes efficiency values, vibration values, and temperature values ​​at multiple sampling times;

[0084] The statistical feature extraction and comprehensive evaluation of the adaptation status index sequence for each historical period, and the calculation of the operational adaptation value for that historical period, include:

[0085] Calculate the mean and variance of each fitness state index sequence;

[0086] The deviation rate of an indicator is calculated based on the series mean of each indicator. The deviation rate represents the degree of deviation of the indicator from the normal value.

[0087] The stability of an indicator is calculated based on the series variance of each indicator, and the stability characterizes the degree of fluctuation of the indicator within a time period.

[0088] After normalizing the deviation rate and stability of each indicator, a weighted fusion is performed to obtain the operational adaptability value.

[0089] In this embodiment, it should be noted that the core idea of ​​the fit mapping model is to learn the operating performance of the water pump under different combinations of speed and flow rate through historical operating data, thereby establishing a mapping relationship between speed, flow rate and operating fit. The operating fit value is used to measure the quality of the water pump's operating state under specific operating conditions. The higher the value, the better the water pump's operating state under that operating condition, the less equipment wear and the lower the life loss.

[0090] It should be noted that historical data refers to monitoring data recorded during the past operation of the water pump, which can usually be obtained from the pump station's SCADA system, online monitoring system, or data acquisition device. Historical data is divided into time periods, within which the water pump operates at a relatively stable speed. For each historical time period, three types of information need to be extracted: time period speed, time period operating flow rate, and adaptation status index sequence.

[0091] It should be noted that the adaptation status index sequence includes three categories of indicators: efficiency, vibration, and temperature. Efficiency reflects the pump's energy conversion performance and can be calculated from measured flow rate, head, and power; higher efficiency indicates more efficient energy utilization. Vibration reflects the pump's mechanical operating state and is typically measured using vibration sensors mounted on the bearing housing or pump casing; lower vibration indicates smoother mechanical operation. Temperature reflects the pump's thermal state, usually referring to bearing temperature, which can be measured using temperature sensors; lower temperature indicates lower frictional losses. Multiple index values ​​are collected within each time period to form the index sequence.

[0092] It should be noted that the purpose of statistical feature extraction on the adaptation status indicator sequence is to transform multiple sampled values ​​within a time period into feature values ​​that can characterize the overall operating status of that time period. The sequence mean reflects the average level of the indicator over the entire time period, while the sequence variance reflects the degree of fluctuation of the indicator within the time period. In a period of good operating status, all indicators should remain within the normal range and relatively stable.

[0093] It should be noted that the deviation rate is calculated by comparing the series mean of each indicator with a preset normal value to obtain the relative degree of deviation. The normal value can be determined based on the equipment's factory parameters, industry standards, or historical best operating records. For efficiency values, the deviation rate represents the degree of decrease in actual efficiency relative to rated efficiency; for vibration and temperature values, the deviation rate represents the degree of increase in actual value relative to the normal operating baseline. A larger deviation rate indicates that the indicator deviates further from the normal state, and the worse the operating condition.

[0094] It should be noted that stability is calculated based on the series variance of each indicator. A larger variance indicates more drastic fluctuations in the indicator over a period of time, and a more unstable operating state. Stability can be obtained by inversely mapping variance; that is, a smaller variance indicates higher stability. Stable operating conditions are crucial for equipment health, as frequent fluctuations accelerate equipment fatigue and wear.

[0095] It should be noted that normalizing the deviation rate and stability of each indicator is to eliminate differences in units and numerical ranges between different indicators, enabling them to be compared and integrated under a unified scale. Normalization can employ methods such as maximum-minimum normalization, and the normalized deviation rate and stability values ​​typically range from 0 to 1. Weighted fusion involves summing the normalized features according to preset weights to obtain a comprehensive operational fitness value. The weights can be determined based on the importance of each indicator's impact on equipment health, or through expert experience or data analysis methods.

[0096] It should be noted that the purpose of regression modeling is to establish a mapping function from input features (time-period rotational speed and time-period operating flow rate) to output labels (operational fitness values). Regression modeling methods that can be used include, but are not limited to, multinomial regression, support vector regression, random forest regression, or neural network regression. After the model is trained, it can predict the pump's operational fitness value under any given combination of rotational speed and flow rate. This provides a rapid assessment of the pump's operating status for subsequent optimization solutions, avoiding the need for actual equipment operation to obtain status feedback during the optimization process.

[0097] S3. Obtain the water supply task, which includes the target total water supply and the planned total duration; construct a spatiotemporal rotation speed matrix, which is used to describe the rotation speed of each water pump in each time period;

[0098] In this embodiment, it should be noted that the water supply task refers to the water supply target that the pumping station needs to complete within a certain period of time. The target total water supply volume refers to the total amount of water that needs to be delivered to the pipeline network within the planned time, usually in cubic meters; the planned total duration refers to the time span for completing the water supply task, usually in hours. The water supply task can be determined based on water demand forecasts, scheduling plans, or instructions from higher authorities.

[0099] It should be noted that the spatiotemporal speed matrix is ​​a two-dimensional matrix structure, with rows corresponding to each water pump and columns corresponding to each time period. Each element in the matrix represents the speed value of the corresponding water pump during that time period. When an element has a value of zero, it indicates that the water pump is in a stopped state during that time period. The spatiotemporal speed matrix can comprehensively describe the operation scheme of all water pumps throughout the entire planning period. The spatiotemporal speed matrix is ​​the core variable for subsequent optimization solutions; the optimization objective is to determine the optimal values ​​of each element in this matrix.

[0100] S4. Using the spatiotemporal rotation speed matrix as the optimization variable, construct the optimization objective based on the pump station digital twin model and the fit mapping model. Combine the water supply task and equipment parameters to set constraints, solve for the optimal value of each element in the spatiotemporal rotation speed matrix, and generate the corresponding optimized operation strategy.

[0101] The steps for constructing the optimization objective based on the pump station digital twin model and the fit mapping model include:

[0102] Input the spatiotemporal rotation speed matrix into the digital twin model of the pump station to obtain the total energy consumption of the system and the working flow rate of each pump in each time period;

[0103] Input the speed and flow rate of each water pump at each time period into the adaptation mapping model to obtain the operating adaptation value of each water pump at each time period.

[0104] The system overall fit is obtained by time-weighted summation of the operational fit values ​​of each pump at each time period. The system misfit is obtained by taking the negative value of the system overall fit.

[0105] The total energy consumption and system misfit of the system are normalized to obtain normalized energy consumption value and normalized misfit value;

[0106] The optimization objective is to minimize the weighted sum of the normalized energy consumption value and the normalized misfit value.

[0107] The constraints include:

[0108] The deviation between the cumulative total water supply and the target total water supply is within the preset range; the sum of the time periods does not exceed the planned total duration; the length of each time period is not less than the minimum time period granularity; the speed of each water pump is zero or within its allowable speed range in each time period; the working flow rate of each water pump in each time period is within its allowable flow rate range; and the length of each time period is not less than the minimum time period granularity.

[0109] In this embodiment, it should be noted that by combining the aforementioned constructed pump station digital twin model and adaptability mapping model, a comprehensive optimization framework capable of simultaneously evaluating energy consumption and operational adaptability is formed. The optimization aims to find an optimal spatiotemporal speed matrix, under the premise of meeting water supply tasks and equipment safety constraints, so that the pump station operates in an energy-efficient and efficient manner.

[0110] It should be noted that the optimization target construction process first involves calculating the total system energy consumption and the operating flow rate of each pump in each time period under a given spatiotemporal speed matrix using a digital twin model of the pump station. This step transforms the speed decision into quantifiable energy consumption indicators and flow allocation results. Then, the speed and flow rate are input into the fit mapping model to obtain the operating fit value of each pump in each time period. This step evaluates the operational status of each pump under corresponding operating conditions.

[0111] It should be noted that the purpose of time-weighted summation of the operational adaptability values ​​of each pump at different time periods is to aggregate the dispersed adaptability values ​​into a comprehensive system-level index. Time weighting means weighting according to the duration of each time period; the longer the time period, the greater the impact of its adaptability value on the comprehensive result. This weighting method is consistent with reality because the longer a pump operates under a certain condition, the more significant the impact of that condition on the overall health of the equipment.

[0112] It should be noted that the purpose of taking the negative value of the overall system fit to obtain the system misfit is to unify the optimization objective into a minimization form. Since a higher fit value indicates better operation, while a lower energy consumption value indicates greater energy efficiency, the optimization directions of the two are opposite. By taking the negative value to convert the fit to misfit, a lower misfit indicates better operation, which aligns with the optimization direction of energy consumption, facilitating the construction of a unified minimization objective function.

[0113] It should be noted that the purpose of normalizing the total system energy consumption and system misfit is to eliminate the differences between the two in terms of dimensions and numerical range. Energy consumption is usually measured in kilowatt-hours and may have a large value; misfit is a dimensionless value, and its numerical range is related to the output of the fitness model. After normalization, both are mapped to a comparable scale range, usually between 0 and 1, making the subsequent weighted summation meaningful.

[0114] It should be noted that minimizing the weighted sum of normalized energy consumption and normalized misfit values ​​as the optimization objective reflects the trade-off between energy conservation and healthy equipment operation. The weighting coefficients can be set according to actual operational needs; increasing the weight of energy consumption increases the focus on energy conservation, while increasing the weight of misfit increases the focus on equipment health. By adjusting the weights, different operation management strategies can be flexibly adapted.

[0115] It should be noted that constraints can be divided into two categories: task constraints and equipment constraints. Task constraints are set to ensure that the water supply task can be completed, including that the deviation between the cumulative total water supply and the target total water supply is within a preset range, and the sum of the time periods does not exceed the planned total duration. Equipment constraints are set to ensure that each water pump operates within its safe operating range, including that the time period length is not less than the minimum time period granularity, the speed of each water pump is zero or within its allowable speed range in each time period, and the operating flow rate of each water pump in each time period is within its allowable flow rate range.

[0116] It should be noted that the cumulative total water supply is calculated by inputting the spatiotemporal rotational speed matrix into the pump station's digital twin model, and its relative deviation from the target total water supply should generally be controlled within ±5%. The total time period refers to the total duration of the time periods in the spatiotemporal rotational speed matrix, which cannot exceed the planned total duration. The constraint that the time period length should not be less than the minimum time period granularity limits the equipment switching frequency and avoids excessively frequent speed adjustments; the minimum time period granularity is typically set to 10 to 20 minutes.

[0117] It should be noted that optimization algorithms suitable for nonlinear programming problems, such as genetic algorithms or particle swarm optimization, can be used to solve the problem. After the solution is obtained, the optimal spatiotemporal speed matrix is ​​transformed into an executable operating strategy. This strategy specifies the speed setpoint that each pump should use in each time period, and can be directly sent to the pump station control system for execution or sent to relevant personnel for reference.

[0118] Example 2

[0119] Please see Figure 2 This invention provides a pump station operation optimization system based on digital twins, used to implement a pump station operation optimization method based on digital twins, including:

[0120] The digital twin module is used to obtain the equipment parameters of each water pump and build a digital twin model of the pumping station based on the equipment parameters. The digital twin model of the pumping station is used to calculate the cumulative total water supply, the total energy consumption of the system, and the working flow rate of each water pump in each time period according to the rotation speed of each water pump in each time period.

[0121] The adaptability module is used to establish an adaptability mapping model for each water pump based on the historical data of each water pump. The adaptability mapping model is used to output the operating adaptability value for a given period based on the rotational speed and operating flow rate of that period.

[0122] The optimization preparation module is used to obtain the water supply task, which includes the target total water supply and the planned total duration; and to construct a spatiotemporal rotation speed matrix, which is used to describe the rotation speed of each water pump in each time period.

[0123] The optimization solution module is used to construct optimization objectives based on the spatiotemporal rotation speed matrix as the optimization variable, the pump station digital twin model and the fitness mapping model, and set constraints in combination with water supply tasks and equipment parameters to solve for the optimal values ​​of each element in the spatiotemporal rotation speed matrix and generate the corresponding optimized operation strategy.

[0124] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0125] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

[0126] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for pump station operation optimization based on digital twinning, characterized in that, include: The equipment parameters of each water pump are obtained, and a digital twin model of the pumping station is constructed based on the equipment parameters. The digital twin model of the pumping station is used to calculate the cumulative total water supply, the total energy consumption of the system, and the working flow rate of each water pump in each time period according to the rotation speed of each water pump in each time period. A fit mapping model for each water pump is established based on historical data of each water pump. The fit mapping model is used to output the operating fit value for a given period based on the rotational speed and operating flow rate of that period. Obtain a water supply task, which includes a target total water supply volume and a planned total duration; Construct a spatiotemporal rotational speed matrix, which is used to describe the rotational speed of each water pump in each time period; Using the spatiotemporal rotation speed matrix as the optimization variable, the optimization objective is constructed based on the pump station digital twin model and the fit mapping model. Constraints are set in combination with water supply tasks and equipment parameters. The optimal values ​​of each element in the spatiotemporal rotation speed matrix are obtained by solving the problem, and the corresponding optimized operation strategy is generated. The steps for establishing the fitness mapping model include: Extract the rotation speed, working flow rate, and adaptation status index sequence for each historical time period; Statistical features are extracted and comprehensively evaluated for the adaptation status index sequence of each historical period, and the operational adaptation value of that historical period is calculated. Construct a training sample set with time-period rotation speed and time-period working flow as input features and operation adaptability value as output label; Regression modeling is performed on the training sample set to establish the mapping relationship between time period rotation speed and time period working flow and operation adaptability value, thus obtaining the adaptability mapping model; The adaptation state index sequence includes efficiency values, vibration values, and temperature values ​​at multiple sampling times; The statistical feature extraction and comprehensive evaluation of the adaptation status index sequence for each historical period, and the calculation of the operational adaptation value for that historical period, include: Calculate the mean and variance of each fitness state index sequence; The deviation rate of an indicator is calculated based on the series mean of each indicator. The deviation rate represents the degree of deviation of the indicator from the normal value. The stability of an indicator is calculated based on the series variance of each indicator, and the stability characterizes the degree of fluctuation of the indicator within a time period. After normalizing the deviation rate and stability of each indicator, a weighted fusion is performed to obtain the operational adaptability value. The steps for constructing the optimization objective based on the pump station digital twin model and the fit mapping model include: Input the spatiotemporal rotation speed matrix into the digital twin model of the pump station to obtain the total energy consumption of the system and the working flow rate of each pump in each time period; Input the speed and flow rate of each water pump at each time period into the adaptation mapping model to obtain the operating adaptation value of each water pump at each time period. The system overall fit is obtained by time-weighted summation of the operational fit values ​​of each pump at each time period. The system misfit is obtained by taking the negative value of the system overall fit. The total energy consumption and system misfit of the system are normalized to obtain normalized energy consumption value and normalized misfit value; The optimization objective is to minimize the weighted sum of the normalized energy consumption value and the normalized fit value. The constraints include: The deviation between the cumulative total water supply and the target total water supply is within the preset range; the sum of the time periods does not exceed the planned total duration; the length of each time period is not less than the minimum time period granularity; the speed of each water pump is zero or within its allowable speed range in each time period; the working flow rate of each water pump in each time period is within its allowable flow rate range; and the length of each time period is not less than the minimum time period granularity.

2. The digital-twin-based pump station operation optimization method according to claim 1, characterized in that, The digital twin model of the pumping station includes a basic model of equipment parameters, a single-period hydraulic response model, a single-period energy consumption calculation model, and a multi-period cumulative model; The equipment parameter basic model is used to store the equipment parameters of each water pump; The single-time hydraulic response model is constructed based on the head-flow curves of each pump, the pipeline resistance coefficient, and the design head. It is used to output the working flow of each pump and the total system flow during a single time period, according to the rotational speed of each pump in that time period. The single-period energy consumption calculation model is constructed based on the efficiency-flow curves of each water pump, and is used to output the system power for that period based on the speed and operating flow of each water pump in a single period. The multi-period cumulative model is used to multiply the total system flow rate of each period by the corresponding period duration and then sum them to obtain the cumulative total water supply, and to multiply the system power of each period by the corresponding period duration and then sum them to obtain the total system energy consumption.

3. The digital-twin-based pump station operation optimization method according to claim 2, characterized in that, The single-time hydraulic response model includes: Operating status determination steps: Determine the operating status of each water pump based on its rotation speed during the time period, and identify the water pumps with a rotation speed greater than zero as the operating water pumps during the time period; Single pump hydraulic transformation steps: Transform the reference head-flow curve at rated speed according to the actual speed of each operating pump to generate the real-time head-flow characteristic curve of each operating pump at the current speed. System parallel solution steps: Superimpose the real-time head-flow characteristic curves of each operating pump, calculate the intersection of the superimposed system parallel characteristic curve and the pipeline resistance curve constructed based on the pipeline resistance coefficient and design head, and determine the system operating head and total system flow rate for this period. Flow reversal allocation step: Based on the determined system operating head, reverse the lookup in the real-time head-flow characteristic curve of each operating pump to obtain the operating flow rate of each operating pump during that time period.

4. The pump station operation optimization method based on digital twin according to claim 3, characterized in that, The single-period energy consumption calculation model includes: Efficiency characteristic transformation steps: Based on the principle that the efficiency is equal at similar operating points, the baseline efficiency-flow curve at rated speed is transformed according to the actual speed to construct the real-time efficiency-flow characteristic curve of each operating pump at the current speed. Efficiency query steps: Map the working flow rate of each operating pump output by the single-time hydraulic response model to the real-time efficiency flow characteristic curve to obtain the actual operating efficiency of each operating pump under the current operating conditions. Power calculation steps: Calculate the shaft power of each operating pump based on its working flow rate, system operating head, and actual operating efficiency, and sum the shaft power of each operating pump to obtain the system power for that period.

5. A pump station operation optimization system based on digital twins, used to implement the pump station operation optimization method based on digital twins as described in any one of claims 1-4, characterized in that, include: The digital twin module is used to obtain the equipment parameters of each water pump and build a digital twin model of the pumping station based on the equipment parameters. The digital twin model of the pumping station is used to calculate the cumulative total water supply, the total energy consumption of the system, and the working flow rate of each water pump in each time period according to the rotation speed of each water pump in each time period. The adaptability module is used to establish an adaptability mapping model for each water pump based on the historical data of each water pump. The adaptability mapping model is used to output the operating adaptability value for a given period based on the rotational speed and operating flow rate of that period. The optimization preparation module is used to obtain water supply tasks, which include the target total water supply volume and the planned total duration; Construct a spatiotemporal rotational speed matrix, which is used to describe the rotational speed of each water pump in each time period; The optimization solution module is used to construct optimization objectives based on the spatiotemporal rotation speed matrix as the optimization variable, the pump station digital twin model and the fitness mapping model, and set constraints in combination with water supply tasks and equipment parameters to solve for the optimal values ​​of each element in the spatiotemporal rotation speed matrix and generate the corresponding optimized operation strategy.

Images