Intelligent photovoltaic water pumping system based on digital twin technology

Abstract

The application belongs to the technical field of photovoltaic water lifting, and specifically discloses an intelligent photovoltaic water lifting system based on digital twin technology, which comprises a real-time data synchronization module, a digital twin modeling module, a simulation analysis module, a decision support module, a photovoltaic power generation module, an energy storage module and a multi-stage pump station; the digital twin modeling module is used for constructing a three-dimensional parameterized model and a dynamic performance simulation model, the three-dimensional parameterized model is used for displaying the real-time state of the photovoltaic water lifting system in a visual manner, and the dynamic performance simulation model is used for simulating the dynamic performance of the photovoltaic water lifting system and obtaining a dynamic performance simulation result; the simulation analysis module is used for predicting the photovoltaic power generation capacity; and the decision support module is used for solving a target function based on the dynamic performance simulation result and the photovoltaic power generation capacity prediction result, generating a water lifting strategy, and implementing water lifting based on the water lifting strategy. Through the application, the efficiency and stability of water lifting can be considered in the case of high-lift water delivery.

Description

Technical Field

[0001] This application belongs to the field of photovoltaic water lifting technology, and more specifically, relates to an intelligent photovoltaic water lifting system based on digital twin technology. Background Technology

[0002] Traditional photovoltaic (PV) water lifting systems have significant limitations in high-lift water conveyance scenarios. A single pump station or a simple series structure is insufficient to meet lift requirements exceeding 1000 meters, leading to low water conveyance efficiency, excessive energy consumption, and the risk of pipeline rupture. Furthermore, the system lacks dynamic matching capabilities between PV power generation and load, relying on historical data to fail to analyze weather and system status in real time, resulting in energy waste or intermittent outages. Therefore, achieving both efficiency and stability in high-lift water conveyance is a pressing technical challenge in this field. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of this application is to achieve both high-lift water conveyance and water lifting efficiency and stability.

[0004] To achieve the above objectives, in a first aspect, this application provides an intelligent photovoltaic water lifting system based on digital twin technology, comprising: Real-time data synchronization module, digital twin modeling module, simulation analysis module, decision support module, photovoltaic power generation module, energy storage module, and multi-level pumping station; The real-time data synchronization module is used to acquire real-time monitoring data based on sensors and synchronize it to the digital twin modeling module and simulation analysis module. The digital twin modeling module is used to construct a three-dimensional parametric model and a dynamic performance simulation model corresponding to the photovoltaic water lifting system. The three-dimensional parametric model is used to visualize the real-time status of the photovoltaic water lifting system based on real-time monitoring data, and the dynamic performance simulation model is used to simulate the dynamic performance of the photovoltaic water lifting system based on real-time monitoring data and obtain the dynamic performance simulation results. The simulation analysis module is used to predict photovoltaic power generation based on meteorological forecast data and real-time monitoring data, and generate photovoltaic power generation prediction results. The decision support module is used to solve the objective function based on the dynamic performance simulation results and the photovoltaic power generation prediction results, generate the water lifting strategy, and control the photovoltaic power generation module, energy storage module and multi-stage pumping station based on the water lifting strategy. The objective function is configured with decision variables, which are used to control the photovoltaic power generation module, energy storage module and multi-stage pumping station. The decision variable values ​​in the optimal solution of the objective function are used to form the water lifting strategy.

[0005] In one possible implementation, the real-time monitoring data includes: ambient temperature. Ambient humidity Ambient wind speed Light intensity Battery current of energy storage module Battery temperature of energy storage module Pipeline pressure and pipeline flow ; Dynamic performance simulation results include: photovoltaic inverter efficiency Battery health capacity Flow coefficients of pumping stations at all levels and pipeline pressure coefficient ;in, Indicates the first The flow coefficient of the pumping station; The dynamic performance simulation model is specifically a physical information neural network, which is used to simulate ambient temperature. Ambient humidity Ambient wind speed Light intensity Battery current of energy storage module Battery temperature of energy storage module Pipeline pressure and pipeline flow Predict the current photovoltaic inverter efficiency of the photovoltaic water lifting system. Battery health capacity Flow coefficients of pumping stations at all levels and pipeline pressure coefficient The physical information neural network is obtained by training a neural network based on a physical constraint loss function.

[0006] In one possible implementation, the physical constraint loss function is determined by the following formula: ; in, Indicates the total loss. This represents the data fitting loss. This represents the physical constraint loss in photovoltaic efficiency. This indicates the pressure-flow constraint loss. and These are the physical constraint weighting coefficients.

[0007] In one possible implementation, the data fitting loss is determined by the following formula: ; Indicates the number of samples. Represents the square of the L2 norm. Indicates the first The label corresponding to each sample The physical information neural network is for the first The predicted results for each sample; The physical constraint loss of photovoltaic efficiency is determined by the following formula: ; ; The physical information neural network is for the first The photovoltaic inverter efficiency predicted for each sample. Indicates the rated efficiency of the photovoltaic panel. For temperature coefficient, Indicates the first The photovoltaic panel temperature corresponding to each sample The preset nominal temperature, Indicates the first The ambient temperature corresponding to each sample Indicates the first The light intensity corresponding to each sample These are preset coefficients; The pressure-flow constraint loss is determined by the following formula: ; Indicates the first The pipeline pressure corresponding to each sample Indicates the first The pipe flow rate corresponding to each sample The physical information neural network is for the first The predicted pipeline pressure coefficient for each sample. This indicates the cross-sectional area of ​​the pipe.

[0008] In one possible implementation, the algorithm used to predict photovoltaic power generation is the random forest regression algorithm.

[0009] In one possible implementation, the objective function is determined by the following formula: ; , , and These are the weighting coefficients. For energy utilization rate evaluation function, This is the equipment loss evaluation function. This is a function for evaluating water supply stability. The dynamic performance simulation results include: photovoltaic inverter efficiency, battery health capacity, and flow coefficients of pump stations at all levels; The photovoltaic power generation forecast results include the photovoltaic power generation forecasts for multiple time periods after the current moment; The decision variables in the objective function include: the working status of the photovoltaic power generation module, the energy storage charging power, the energy storage discharging power, the start-stop status of each level of pump station, and the operating frequency of each level of pump station; The energy efficiency evaluation function is determined by the following formula: ; The time period number indicates the nth time period after the current time. Each period, Indicates the number of time periods. Indicates the first Electricity purchased from the grid during a given time period Indicates the photovoltaic inverter efficiency; Indicates the first Forecasted photovoltaic power generation for each time period; This indicates the working status of the photovoltaic power generation module. Indicates the system's rated power; The equipment loss evaluation function is determined by the following formula: ; This refers to the equipment loss weighting coefficient corresponding to the energy storage module. Indicates the first Energy storage charging power during each time period, Indicates the first Energy storage discharge power during each time period, Indicates the battery's healthy capacity. For the first The equipment loss weighting coefficient corresponding to the pumping station. Indicates the number of stages in a multi-stage pumping station. Indicates the first The loss rate of the pumping station Indicates the first The single start-stop loss coefficient of a primary pumping station. Indicates the first The first time period Start-up and shutdown status of the pumping station; The water supply stability evaluation function is determined by the following formula: ; Indicates the first Water demand for each time period Indicates the first The flow coefficient of the pumping station Indicates the first The operating frequency of the pumping station.

[0010] In one possible implementation, the constraints corresponding to the objective function include: pump station control constraints, hydraulic safety constraints, and energy storage system constraints. The dynamic performance simulation results also include: pipeline pressure coefficient; The control constraints of the pumping station are determined by the following formula: ; Indicates the first Minimum operating frequency of the pumping station. Indicates the first The maximum operating frequency of the pumping station; Hydraulic safety constraints are determined by the following formula: ; ; This indicates the maximum flow rate in the pipeline. Indicates the pipeline pressure coefficient. Indicates the maximum pressure the pipeline can withstand; The constraints of the energy storage system are determined by the following formula: ; ; Indicates the first The state of charge of the energy storage module at each time period. Indicates energy storage charging efficiency. Indicates the energy storage discharge efficiency. Indicates the duration of each time period. This represents the lower limit of the state of charge range of the energy storage module. This indicates the upper limit of the state of charge range of the energy storage module.

[0011] In one possible implementation, the pump station control constraints also include: ; in, Indicates the first Start-up and shutdown status of the first-level pumping station during each time period Indicates the first Light intensity at different times This represents the light intensity threshold.

[0012] In one possible implementation, the photovoltaic water lifting system is equipped with sensors including flow meters and pressure sensors, which are used to monitor the water delivery status of the water lifting pipeline.

[0013] Secondly, this application provides an intelligent photovoltaic water lifting method based on digital twin technology, applied to the intelligent photovoltaic water lifting system based on digital twin technology described in the first aspect or any possible implementation of the first aspect, comprising: Obtain real-time monitoring data; Using a three-dimensional parametric model and based on real-time monitoring data, the real-time status of the photovoltaic water lifting system is displayed in a visual manner. Using a dynamic performance simulation model and based on real-time monitoring data, the dynamic performance of the photovoltaic water lifting system is simulated, and the dynamic performance simulation results are obtained. Based on meteorological forecast data and real-time monitoring data, photovoltaic power generation is predicted, and photovoltaic power generation prediction results are generated. Based on the dynamic performance simulation results and photovoltaic power generation prediction results, the objective function is solved, a water lifting strategy is generated, and the photovoltaic power generation module, energy storage module and multi-stage pumping station are controlled based on the water lifting strategy. The objective function is configured with decision variables, which are used to control the photovoltaic power generation module, energy storage module and multi-stage pumping station. The decision variable values ​​in the optimal solution of the objective function are used to form the water lifting strategy.

[0014] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art: The system utilizes a real-time data synchronization module to collect real-time monitoring data; a digital twin modeling module to construct a 3D parametric model and a dynamic performance simulation model, visualizing the system status and simulating dynamic performance based on real-time monitoring data; a simulation analysis module to predict photovoltaic power generation based on meteorological forecast data and real-time monitoring data; and a decision support module to integrate dynamic performance simulation results and power generation predictions to solve the objective function (whose decision variables control photovoltaic power generation, energy storage, and multi-stage pumping stations), generating the optimal water lifting strategy. Multi-stage pumping stations enable high-lift water conveyance. Simultaneously, this strategy dynamically adjusts the output of the photovoltaic power generation modules, the charging and discharging of the energy storage modules, and the coordinated operation of multi-stage pumping stations (such as optimizing pump speed and inter-stage scheduling). Thus, during high-lift water conveyance, predictive optimization reduces energy waste, improves pump efficiency, and ensures water lifting efficiency. Furthermore, energy storage buffers photovoltaic fluctuations, and the redundant design of multi-stage pumping stations ensures the stability and continuity of the water lifting process. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of the intelligent photovoltaic water lifting system based on digital twin technology provided in the embodiments of this application; Figure 2 This is a functional diagram of the real-time data synchronization module provided in an embodiment of this application; Figure 3 This is a functional schematic diagram of the digital twin modeling module provided in an embodiment of this application; Figure 4This is a functional diagram of the simulation analysis module provided in the embodiments of this application; Figure 5 This is a functional diagram of the decision support module provided in the embodiments of this application. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0017] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0018] In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more, for example, multiple processing units means two or more processing units, multiple elements means two or more elements, etc.

[0019] The embodiments of this application are described below with reference to the accompanying drawings.

[0020] Figure 1 This is a schematic diagram of the structure of the intelligent photovoltaic water lifting system based on digital twin technology provided in the embodiments of this application, as shown below. Figure 1 As shown, the system includes: a real-time data synchronization module, a digital twin modeling module, a simulation analysis module, and a decision support module; The system also includes photovoltaic power generation modules, energy storage modules, and multi-stage pumping stations; The real-time data synchronization module is used to acquire real-time monitoring data based on sensors and synchronize it to the digital twin modeling module and simulation analysis module. The digital twin modeling module is used to construct a three-dimensional parametric model and a dynamic performance simulation model corresponding to the photovoltaic water lifting system. The three-dimensional parametric model is used to visualize the real-time status of the photovoltaic water lifting system based on real-time monitoring data, and the dynamic performance simulation model is used to simulate the dynamic performance of the photovoltaic water lifting system based on real-time monitoring data and obtain the dynamic performance simulation results. The simulation analysis module is used to predict photovoltaic power generation based on meteorological forecast data and real-time monitoring data, and generate photovoltaic power generation prediction results. The decision support module is used to solve the objective function based on the dynamic performance simulation results and the photovoltaic power generation prediction results, generate the water lifting strategy, and control the photovoltaic power generation module, energy storage module and multi-stage pumping station based on the water lifting strategy. The objective function is configured with decision variables, which are used to control the photovoltaic power generation module, energy storage module and multi-stage pumping station. The decision variable values ​​in the optimal solution of the objective function are used to form the water lifting strategy.

[0021] Specifically, Figure 2 This is a functional diagram of the real-time data synchronization module provided in an embodiment of this application, as shown below. Figure 2 As shown, the system is equipped with sensors for monitoring the environment and sensors for monitoring the system status. Sensors for monitoring the environment include temperature sensors, humidity sensors, wind speed sensors, and light sensors. Sensors for monitoring the system status include flow meters, pressure sensors, and ammeters. The flow meters and pressure sensors are used to monitor the water delivery status of the water supply pipeline, replacing the function of the level sensor. Sensor data reliability is ensured through redundant design and regular calibration; a communication protocol is used to transmit data to the digital twin modeling module for data processing (including data preprocessing); ultimately, cloud storage, remote monitoring, and historical data storage are achieved.

[0022] Figure 3 This is a functional diagram of the digital twin modeling module provided in an embodiment of this application, such as... Figure 3 As shown, the digital twin modeling module constructs a three-dimensional parametric model of the photovoltaic water lifting system. It simulates the dynamic performance of the system through a simulation platform. The model parameters include solar panel angle, light intensity, ambient temperature, battery capacity, and inverter efficiency. This module integrates a cloud platform to support remote simulation model updates and system status visualization. It also achieves dynamic optimization through bidirectional interaction between the data twin and the physical system.

[0023] Figure 4 This is a functional diagram of the simulation analysis module provided in the embodiments of this application, such as... Figure 4 As shown, the simulation analysis module constructs a photovoltaic power generation prediction model based on the random forest regression algorithm, and dynamically optimizes the prediction model parameters by combining historical meteorological data and real-time meteorological data to improve the model's generalization ability; the root mean square error (RMSE) and mean absolute error (MAE) are used to quantify the model performance, and the prediction accuracy is improved through iterative training.

[0024] Figure 5 This is a functional diagram of the decision support module provided in an embodiment of this application, such as... Figure 5As shown, the decision support module formulates a time-sharing water pumping plan and dynamically optimizes the pump operation strategy based on dynamic performance simulation results and photovoltaic power generation prediction results; it issues instructions to the multi-stage pump station PLC control system through the industrial standard communication protocol to monitor battery power, water pumping flow and pipeline pressure in real time; it sets up an abnormal state alarm mechanism to provide real-time early warning and linkage processing for insufficient battery power, pump failure or abnormal pipeline pressure; when there is sufficient sunlight, the first-stage floating pump station is started first to draw water, and when there is insufficient sunlight, the second-stage pump station is maintained through the energy storage battery.

[0025] The photovoltaic power generation module converts solar energy into electrical energy through photovoltaic panels; the energy storage module uses energy storage battery packs to provide power supply for the secondary pumping station when sunlight is insufficient, and monitors the battery power in real time; the pumping station adopts a multi-level architecture, including a primary floating pumping station (equipped with floating multi-stage centrifugal pumps, deployed in the water source reservoir area to adapt to water level fluctuations) and secondary / tertiary ground pumping stations (for example, water is gradually pressurized and transported to a high-level water tank with a total head of 1100 meters through water lifting pipelines). The system operates in coordination through a PLC control system, which dynamically adjusts the pump start-up and shutdown sequence and power distribution according to the photovoltaic power generation. When sunlight is sufficient, the primary floating pumping station is started first, and when sunlight is insufficient, the secondary pumping station is maintained by the energy storage battery.

[0026] Understandably, the system utilizes a real-time data synchronization module to collect real-time monitoring data; a digital twin modeling module to construct a three-dimensional parametric model and a dynamic performance simulation model, visualizing the system state and simulating dynamic performance based on real-time monitoring data; a simulation analysis module to predict photovoltaic power generation based on meteorological forecast data and real-time monitoring data; and a decision support module to integrate dynamic performance simulation results and power generation predictions to solve the objective function (whose decision variables control photovoltaic power generation, energy storage, and multi-stage pumping stations), generating the optimal water lifting strategy. Multi-stage pumping stations enable high-lift water conveyance. Simultaneously, this strategy dynamically adjusts the output of the photovoltaic power generation module, the charging and discharging of the energy storage module, and the coordinated operation of multi-stage pumping stations (such as optimizing pump speed and inter-stage scheduling). Thus, during high-lift water conveyance, predictive optimization reduces energy waste, improves pump efficiency, and ensures water lifting efficiency. Furthermore, energy storage buffers photovoltaic fluctuations, and the redundant design of multi-stage pumping stations ensures the stability and continuity of the water lifting process.

[0027] In one possible implementation, the real-time monitoring data includes: ambient temperature. Ambient humidity Ambient wind speed Light intensity Battery current of energy storage module Battery temperature of energy storage module Pipeline pressure and pipeline flow ; Dynamic performance simulation results include: photovoltaic inverter efficiency Battery health capacity Flow coefficients of pumping stations at all levels and pipeline pressure coefficient ;in, Indicates the first The flow coefficient of the pumping station; The dynamic performance simulation model is specifically a physical information neural network, which is used to simulate light intensity. Ambient temperature Battery current of energy storage module Pipeline pressure and pipeline flow Predict the current photovoltaic inverter efficiency of the photovoltaic water lifting system. Battery health capacity Flow coefficients of pumping stations at all levels and pipeline pressure coefficient The physical information neural network is obtained by training a neural network based on a physical constraint loss function.

[0028] Here is the pipeline pressure coefficient This coefficient is used to characterize the combined influence of the inherent physical properties of the pipeline system (such as pipe wall roughness, local resistance, geometric changes, etc.) on fluid flow resistance or pressure loss at a specific moment.

[0029] Understandably, Physical Information Neural Networks (PINNs) directly embed the governing equations (physical constraints on photovoltaic efficiency, pressure-flow relationship constraints) into the loss function of the neural network, forcing the model output to conform to physical laws (such as the decay of photovoltaic efficiency with temperature, the relationship between pipeline flow and the square of pressure, etc.), thereby avoiding the physically unreasonable predictions that may occur in purely data-driven models. Simultaneously, the network dynamically calibrates physical parameters using real-time sensor data (illuminance, temperature, vibration, etc.), enabling the model to adapt to environmental changes. This dual mechanism of physical constraints and data correction not only compensates for the errors caused by the simplification assumptions of traditional physical models but also solves the generalization problem of pure data models under small samples or extreme conditions, ultimately achieving high-precision, interpretable dynamic performance prediction.

[0030] In one possible implementation, the physical constraint loss function is determined by the following formula: ; in, Indicates the total loss. This represents the data fitting loss. This represents the physical constraint loss in photovoltaic efficiency. This indicates the pressure-flow constraint loss. and These are the physical constraint weighting coefficients.

[0031] In one possible implementation, the data fitting loss is determined by the following formula: ; Indicates the number of samples. Represents the square of the L2 norm. Indicates the first One sample (ambient temperature) Ambient humidity Ambient wind speed Light intensity Battery current of energy storage module Battery temperature of energy storage module Pipeline pressure and pipeline flow The corresponding tag, The physical information neural network is for the first The predicted results for each sample (photovoltaic inverter efficiency) Battery health capacity Flow coefficients of pumping stations at all levels and pipeline pressure coefficient ); The physical constraint loss of photovoltaic efficiency is determined by the following formula: ; ; The physical information neural network is for the first The photovoltaic inverter efficiency predicted for each sample. Indicates the rated efficiency of the photovoltaic panel. For temperature coefficient, Indicates the first The photovoltaic panel temperature corresponding to each sample The preset nominal temperature, Indicates the first The ambient temperature corresponding to each sample Indicates the first The light intensity corresponding to each sample These are preset coefficients; The pressure-flow constraint loss is determined by the following formula: ; Indicates the first The pipeline pressure corresponding to each sample Indicates the first The pipe flow rate corresponding to each sample The physical information neural network is for the first The predicted pipeline pressure coefficient (pipeline resistance coefficient) for each sample. This indicates the cross-sectional area of ​​the pipe.

[0032] It should be noted that the aforementioned physical constraint loss on photovoltaic efficiency refers to the photovoltaic inverter efficiency predicted by the neural network through physical equations. This conforms to the temperature decay law of photovoltaic panel efficiency. When the photovoltaic panel temperature... Exceeding the nominal temperature At that time, through The degree of efficiency decay in coefficient quantization forces neural network predictions to... Must meet the rated efficiency of photovoltaic panels The physical law of linear decay with temperature increase allows for the embedding of engineering empirical formulas into the learning process, ensuring the physical rationality of the prediction results.

[0033] Regarding the pressure-flow relationship constraint loss mentioned above, this constraint term forces the model to learn the basic principles of pipe fluid mechanics through a mathematical formula: pipe pressure loss is proportional to the square of the flow velocity. Specifically, this constraint calculates the pipe pressure. , and based on pipeline flow and predicted pipeline pressure coefficient Calculated theoretical pressure loss The mean square error between the two. Wherein, the flow rate divided by the pipe cross-sectional area... The expected pressure drop is represented by the square of the obtained flow velocity multiplied by the drag coefficient. By minimizing this loss function, the model is constrained to ensure that the pressure value is consistent with the hydraulic relationship derived from the predicted pipe pressure coefficient, thereby ensuring that the prediction results conform to physical laws.

[0034] In one possible implementation, when constructing a 3D parametric model of a photovoltaic water lifting system, the geometric models of each component (photovoltaic panel, support structure, energy storage battery, water pump, pipes, etc.) are first accurately created in 3D modeling software, and static parameters, such as dimensions, materials, and rated power, are assigned to them. Next, these static parameters are set as adjustable variables. Subsequently, a real-time parameter interface is configured to input real-time monitoring data into the model, mapping this real-time data to corresponding locations or attributes in the model. For example, changes in light intensity can dynamically adjust the surface color of the photovoltaic panel or the power generation indicator, and flow meter data can update the visualization of water flow status in the pipes in real time. Ultimately, through this combination of static parameters and dynamically updated real-time parameters, the model not only accurately reflects the physical structure but also presents real-time information from various sensors in a 3D scene with intuitive visual effects (such as color changes, numerical labels, dynamic arrows, etc.), achieving real-time visual monitoring of the system's operating status.

[0035] In one possible implementation, the algorithm used to predict photovoltaic power generation is the random forest regression algorithm.

[0036] Specifically, meteorological forecast data provides information on potential weather conditions over a future period, such as sunlight intensity, temperature, and cloud cover, while real-time monitoring data reflects the actual operating status of the photovoltaic power generation modules, including power generation and equipment status. The random forest regression algorithm constructs multiple decision trees, each making predictions based on both meteorological forecast data and real-time monitoring data. By combining the prediction results from all decision trees, the algorithm effectively processes this complex and multi-dimensional data, capturing nonlinear relationships and interactions within the data. Ultimately, the algorithm generates a prediction of photovoltaic power generation.

[0037] In one possible implementation, the objective function described above is determined by the following formula: ; , , and These are the weighting coefficients. For energy utilization rate evaluation function, This is the equipment loss evaluation function. This is a function for evaluating water supply stability. The dynamic performance simulation results include: photovoltaic inverter efficiency, battery health capacity, and flow coefficients of pump stations at all levels; The photovoltaic power generation forecast results include the photovoltaic power generation forecasts for multiple time periods after the current moment; The decision variables in the objective function include: the working status of the photovoltaic power generation module, the energy storage charging power, the energy storage discharging power, the start-stop status of each level of pump station, and the operating frequency of each level of pump station; The energy efficiency evaluation function is determined by the following formula: ; This is the number of the time interval (time step), indicating the number of the intervals after the current time. Each period, Indicates the number of time periods. Indicates the first Electricity purchased from the grid during a given time period Indicates the photovoltaic inverter efficiency; Indicates the first Forecasted photovoltaic power generation for each time period; This indicates the working status of the photovoltaic power generation module (value is 0 or 1, 0 means off, 1 means running). This indicates the system's rated power (fixed value). The equipment loss evaluation function is determined by the following formula: ; This refers to the equipment loss weighting coefficient corresponding to the energy storage module. Indicates the first Energy storage charging power during each time period, Indicates the first Energy storage discharge power during each time period, This indicates the battery's healthy capacity (the actual usable capacity of the battery). For the first The equipment loss weighting coefficient corresponding to the pumping station. Indicates the number of stages in a multi-stage pumping station. Indicates the first The loss rate of the pumping station Indicates the first The single start-stop loss coefficient (fixed value) of the pumping station. Indicates the first The first time period The start / stop status of the pumping station (value is 0 or 1, 0 indicates stop, 1 indicates start); The water supply stability evaluation function is determined by the following formula: ; Indicates the first Water demand for each time period Indicates the first The flow coefficient of the pumping station Indicates the first The operating frequency of the pumping station.

[0038] For example, It is determined by the following formula: ; Indicates the first The first time period The usage time of the pumping station. This indicates the total lifespan of the pumps in each level of the pumping station. Indicates the first The load intensity index of the pumps in the pumping station; ; Indicates the first The rated power of the pumps in the pumping station. Indicates the first The actual operating power of the pumps in the pumping station. Indicates the first The rated current of the pumps in the pumping station. Indicates the first The actual operating current of the pumps in the pumping station. Indicates the first The rated speed of the water pumps in the pumping station. Indicates the first The actual operating speed of the pumps in the pumping station. , , These are weighting coefficients, typically set based on equipment type and importance (e.g., higher power weight). The actual operating power, actual operating current, and actual operating speed of the water pump can be obtained by statistically analyzing the pump's historical operating data.

[0039] It is understandable that the energy utilization rate sub-objective in the objective function... By minimizing the difference between grid-purchased electricity and photovoltaic power generation, the system is forced to prioritize the use of solar energy, thereby reducing energy costs; equipment loss sub-objective By quantifying battery charge / discharge losses and pump station failure rates, equipment aging caused by overcharging / discharging or high-frequency operation is suppressed; water supply stability sub-objective This directly minimizes the deviation between water demand and actual water supply, ensuring that the flow rate matches the demand. Therefore, this objective function achieves efficient water lifting under the premise of long-term reliable operation through joint optimization of efficiency, loss, and stability.

[0040] In one possible implementation, the constraints corresponding to the objective function include: pump station control constraints, hydraulic safety constraints, and energy storage system constraints. The dynamic performance simulation results also include: pipeline pressure coefficient; The control constraints of the pumping station are determined by the following formula: ; Indicates the first Minimum operating frequency of the pumping station. Indicates the first The maximum operating frequency of the pumping station; Hydraulic safety constraints are determined by the following formula: ; ; This indicates the maximum flow rate in the pipeline. Indicates the pipeline pressure coefficient. Indicates the maximum pressure the pipeline can withstand; The constraints of the energy storage system are determined by the following formula: ; ; Indicates the first The state of charge of the energy storage module at each time period (which reflects the ratio of the battery's currently available energy to its total capacity). Indicates energy storage charging efficiency. Indicates the energy storage discharge efficiency. Indicates the duration of each time period. This represents the lower limit of the state of charge range of the energy storage module. This indicates the upper limit of the state of charge range of the energy storage module.

[0041] In one possible implementation, the above-mentioned pump station control constraints also include: ; in, Indicates the first Start-up and shutdown status of the first-level pumping station during each time period Indicates the first Light intensity for each time period (which can be obtained from meteorological forecast data). This represents the light intensity threshold.

[0042] In one possible implementation, the photovoltaic water lifting system is equipped with sensors including flow meters and pressure sensors, which are used to monitor the water delivery status of the water lifting pipeline.

[0043] Understandably, the installation and maintenance of liquid level sensors are typically complex, requiring precise positioning and professional debugging. They are also susceptible to interference from environmental factors such as temperature changes, vibration, and impurities, all of which can lead to inaccurate measurement data, increasing operating costs and the risk of malfunction. Using flow meters and pressure sensors to monitor the water delivery status of water supply pipelines offers a more reliable and efficient alternative. Flow meters accurately measure water flow, while pressure sensors reflect real-time pressure changes within the pipeline. The combination of these two provides a comprehensive understanding of the dynamic situation during water delivery. Installation and maintenance are relatively simple, and they are less affected by environmental interference, providing a more stable alternative to liquid level sensors and ensuring the stable operation of the water delivery system.

[0044] This application also provides an intelligent photovoltaic water lifting method based on digital twin technology, applicable to any of the aforementioned intelligent photovoltaic water lifting systems based on digital twin technology, the method comprising: Obtain real-time monitoring data; Using a three-dimensional parametric model and based on real-time monitoring data, the real-time status of the photovoltaic water lifting system is displayed in a visual manner. Using a dynamic performance simulation model and based on real-time monitoring data, the dynamic performance of the photovoltaic water lifting system is simulated, and the dynamic performance simulation results are obtained. Based on meteorological forecast data and real-time monitoring data, photovoltaic power generation is predicted, and photovoltaic power generation prediction results are generated. Based on the dynamic performance simulation results and photovoltaic power generation prediction results, the objective function is solved, a water lifting strategy is generated, and the photovoltaic power generation module, energy storage module and multi-stage pumping station are controlled based on the water lifting strategy. The objective function is configured with decision variables, which are used to control the photovoltaic power generation module, energy storage module and multi-stage pumping station. The decision variable values ​​in the optimal solution of the objective function are used to form the water lifting strategy.

[0045] The system's dynamic performance is simulated by constructing a three-dimensional parametric model through a digital twin modeling module; the real-time data synchronization module collects real-time data from environmental sensors (temperature, humidity, wind speed, light, etc.) and system status sensors (flow meters, pressure sensors, etc.), and transmits the pre-processed data to the digital twin modeling module; the simulation analysis module predicts photovoltaic power generation based on a random forest regression algorithm combined with meteorological and real-time monitoring data; the decision support module formulates a time-sharing water pumping strategy based on the prediction results, and controls the coordinated operation of multiple pumping stations (the first-level floating pumping station adapts to water level fluctuations, and the second / third-level ground pumping stations progressively increase pressure and deliver water) through an industrial standard communication protocol. When there is sufficient sunlight, the first-level pumping station is activated first, and when there is insufficient sunlight, the second-level pumping station is maintained through an energy storage battery.

[0046] Therefore, the embodiments provided in this application can solve the problems of insufficient high-lift water conveyance capacity, lack of dynamic matching of photovoltaic power generation, strong sensor dependence and insufficient application of digital twin technology, realize high-lift water conveyance with a total head of 1100 meters, improve water conveyance efficiency, energy utilization rate, system reliability and remote operation and maintenance capabilities, and are suitable for complex terrain scenarios.

[0047] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An intelligent photovoltaic water pumping system based on digital twin technology, characterized in that, include: Real-time data synchronization module, digital twin modeling module, simulation analysis module, decision support module, photovoltaic power generation module, energy storage module, and multi-level pumping station; The real-time data synchronization module is used to acquire real-time monitoring data based on sensors and synchronize it to the digital twin modeling module and simulation analysis module. The digital twin modeling module is used to construct a three-dimensional parametric model and a dynamic performance simulation model corresponding to the photovoltaic water lifting system. The three-dimensional parametric model is used to visualize the real-time status of the photovoltaic water lifting system based on real-time monitoring data, and the dynamic performance simulation model is used to simulate the dynamic performance of the photovoltaic water lifting system based on real-time monitoring data and obtain the dynamic performance simulation results. The simulation analysis module is used to predict photovoltaic power generation based on meteorological forecast data and real-time monitoring data, and generate photovoltaic power generation prediction results. The decision support module is used to solve the objective function based on the dynamic performance simulation results and the photovoltaic power generation prediction results, generate the water lifting strategy, and control the photovoltaic power generation module, energy storage module and multi-stage pumping station based on the water lifting strategy. The objective function is configured with decision variables, which are used to control the photovoltaic power generation module, energy storage module and multi-stage pumping station. The decision variable values ​​in the optimal solution of the objective function are used to form the water lifting strategy. The objective function is determined by the following formula: ； , , and These are the weighting coefficients. For energy utilization rate evaluation function, This is a function for evaluating equipment loss. This is a function for evaluating water supply stability. The dynamic performance simulation results include: photovoltaic inverter efficiency, battery health capacity, and flow coefficients of pump stations at all levels; The photovoltaic power generation forecast results include the photovoltaic power generation forecasts for multiple time periods after the current moment; The decision variables in the objective function include: the working status of the photovoltaic power generation module, the energy storage charging power, the energy storage discharging power, the start-stop status of each level of pump station, and the operating frequency of each level of pump station; The energy efficiency evaluation function is determined by the following formula: ； The time period number indicates the nth time period after the current time. Each period, Indicates the number of time periods. Indicates the first Electricity purchased from the grid during a given time period Indicates the photovoltaic inverter efficiency; Indicates the first Forecasted photovoltaic power generation for each time period; This indicates the working status of the photovoltaic power generation module. Indicates the system's rated power; The equipment loss evaluation function is determined by the following formula: ； This refers to the equipment loss weighting coefficient corresponding to the energy storage module. Indicates the first Energy storage charging power during each time period, Indicates the first Energy storage discharge power during each time period, Indicates the battery's healthy capacity. For the first The equipment loss weighting coefficient corresponding to the pumping station. Indicates the number of stages in a multi-stage pumping station. Indicates the first The loss rate of the pumping station Indicates the first The single start-stop loss coefficient of a primary pumping station. Indicates the first The first time period Start-up and shutdown status of the pumping station; The water supply stability evaluation function is determined by the following formula: ； Indicates the first Water demand for each time period Indicates the first The flow coefficient of the pumping station Indicates the first The operating frequency of the pumping station.

2. The intelligent photovoltaic water lifting system based on digital twin technology according to claim 1, characterized in that, Real-time monitoring data includes: ambient temperature Ambient humidity Ambient wind speed Light intensity Battery current of energy storage module Battery temperature of energy storage module Pipeline pressure and pipeline flow ; Dynamic performance simulation results include: photovoltaic inverter efficiency Battery health capacity Flow coefficients of pumping stations at all levels and pipeline pressure coefficient ;in, Indicates the first The flow coefficient of the pumping station; The dynamic performance simulation model is specifically a physical information neural network, which is used to simulate ambient temperature. Ambient humidity Ambient wind speed Light intensity Battery current of energy storage module Battery temperature of energy storage module Pipeline pressure and pipeline flow Predict the current photovoltaic inverter efficiency of the photovoltaic water lifting system. Battery health capacity Flow coefficients of pumping stations at all levels and pipeline pressure coefficient The physical information neural network is obtained by training a neural network based on a physical constraint loss function.

3. The intelligent photovoltaic water lifting system based on digital twin technology according to claim 2, characterized in that, The physical constraint loss function is determined by the following formula: ； in, Indicates the total loss. This represents the data fitting loss. This represents the physical constraint loss in photovoltaic efficiency. This indicates the pressure-flow constraint loss. and These are the physical constraint weighting coefficients.

4. The intelligent photovoltaic water pumping system based on digital twin technology according to claim 3, characterized in that, The data fitting loss is determined by the following formula: ； Indicates the number of samples. Represents the square of the L2 norm. Indicates the first The label corresponding to each sample The physical information neural network is for the first The predicted results for each sample; The physical constraint loss of photovoltaic efficiency is determined by the following formula: ； ； The physical information neural network is for the first The photovoltaic inverter efficiency predicted for each sample. Indicates the rated efficiency of the photovoltaic panel. For temperature coefficient, Indicates the first The photovoltaic panel temperature corresponding to each sample The preset nominal temperature, Indicates the first The ambient temperature corresponding to each sample Indicates the first The light intensity corresponding to each sample These are preset coefficients; The pressure-flow constraint loss is determined by the following formula: ； Indicates the first The pipeline pressure corresponding to each sample Indicates the first The pipe flow rate corresponding to each sample The physical information neural network is for the first The predicted pipeline pressure coefficient for each sample. This indicates the cross-sectional area of ​​the pipe.

5. The intelligent photovoltaic water lifting system based on digital twin technology according to claim 1, characterized in that, The algorithm used to predict photovoltaic power generation is the random forest regression algorithm.

6. The intelligent photovoltaic water lifting system based on digital twin technology according to claim 1, characterized in that, The constraints corresponding to the objective function include: pump station control constraints, hydraulic safety constraints, and energy storage system constraints; The dynamic performance simulation results also include: pipeline pressure coefficient; The control constraints of the pumping station are determined by the following formula: ； Indicates the first Minimum operating frequency of the pumping station. Indicates the first The maximum operating frequency of the pumping station; Hydraulic safety constraints are determined by the following formula: ； ； This indicates the maximum flow rate in the pipeline. Indicates the pipeline pressure coefficient. Indicates the maximum pressure the pipeline can withstand; The constraints of the energy storage system are determined by the following formula: ； ； Indicates the first The state of charge of the energy storage module at each time period. Indicates energy storage charging efficiency. Indicates the energy storage discharge efficiency. Indicates the duration of each time period. This represents the lower limit of the state of charge range of the energy storage module. This indicates the upper limit of the state of charge range of the energy storage module.

7. The intelligent photovoltaic water lifting system based on digital twin technology according to claim 6, characterized in that, Pump station control constraints also include: ； in, Indicates the first Start-up and shutdown status of the first-level pumping station during each time period Indicates the first Light intensity at different times This represents the light intensity threshold.

8. The intelligent photovoltaic water lifting system based on digital twin technology according to claim 1, characterized in that, The sensors configured in the photovoltaic water lifting system include flow meters and pressure sensors, which are used to monitor the water delivery status of the water lifting pipeline.

9. A smart photovoltaic water lifting method based on digital twin technology, characterized in that, The system is applied to the intelligent photovoltaic water lifting system based on digital twin technology as described in any one of claims 1-8, comprising: Obtain real-time monitoring data; Using a three-dimensional parametric model and based on real-time monitoring data, the real-time status of the photovoltaic water lifting system is displayed in a visual manner. Using a dynamic performance simulation model and based on real-time monitoring data, the dynamic performance of the photovoltaic water lifting system is simulated, and the dynamic performance simulation results are obtained. Based on meteorological forecast data and real-time monitoring data, photovoltaic power generation is predicted, and photovoltaic power generation prediction results are generated. Based on the dynamic performance simulation results and photovoltaic power generation prediction results, the objective function is solved, a water lifting strategy is generated, and the photovoltaic power generation module, energy storage module and multi-stage pumping station are controlled based on the water lifting strategy. The objective function is configured with decision variables, which are used to control the photovoltaic power generation module, energy storage module and multi-stage pumping station. The decision variable values ​​in the optimal solution of the objective function are used to form the water lifting strategy.

Images