A method and system for intelligent control of a photovoltaic power station drainage pump networking

Abstract

The present application relates to the technical field of intelligent control, and more particularly to a kind of photovoltaic power station's drainage pump networking intelligent control method and system.Acquire multi-source data, based on weighted fusion algorithm, multi-source data is fused to obtain effective water level value;Based on effective water level value, build waterlogging risk prediction model, obtain water level prediction value;Based on multi-source data, effective water level value, water level prediction value, select current operation mode;Based on the constraint condition of pre-set, build the optimization model of target;Based on selecting current operation mode, the optimal target is obtained through the optimization model of target Drainage pump power setting value.The present application aims to provide a kind of photovoltaic power station drainage pump networking intelligent control method and system capable of realizing predictive drainage, economic scheduling and intelligent control, to improve the safety, economy and intelligent level of drainage system.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to a network-connected intelligent control method and system for drainage pumps in a photovoltaic power station. Background Technology

[0002] Photovoltaic power plants are typically built in open outdoor areas, equipped with drainage ditches, collection ponds, and drainage pumps to remove rainwater, prevent flooding, and ensure the safety of the power plant equipment. As a critical auxiliary device, the level of intelligence in the control of drainage pumps directly impacts the safety and economic efficiency of the power plant's operation.

[0003] However, current control of drainage pumps is still limited to manual control or simple logic control. It is impossible to accurately control the water level in the collection tank or to intelligently control the operation of drainage pumps according to different weather conditions, which poses a safety hazard to photovoltaic power stations.

[0004] In addition, the start-up, shutdown, and operation of the drainage pump are completely unrelated to the power generation status of the photovoltaic power station and the grid electricity price. It may be idle when the photovoltaic power generation is in progress, but consume electricity to drain water during peak electricity price periods, which is an uneconomical situation. Frequent start-up and shutdown and direct power frequency operation are also detrimental to the lifespan of the equipment.

[0005] To address the aforementioned issues, there is an urgent need for a network-based intelligent control method for drainage pumps in photovoltaic power plants. This method would enable multi-objective collaborative optimization, ensuring the safety of the photovoltaic power plant while also considering its economic efficiency and intelligent control. Summary of the Invention

[0006] In view of the above problems, this disclosure provides a networked intelligent control method and system for drainage pumps of photovoltaic power plants to overcome or at least partially solve the above problems. The purpose is to provide a networked intelligent control method and system for drainage pumps of photovoltaic power plants that can realize predictive drainage, economical scheduling, and intelligent control, so as to improve the safety, economy and intelligence level of the drainage system.

[0007] The objective of this invention can be achieved through the following technical solutions: The first aspect of the present invention provides a method for intelligent network control of drainage pumps in a photovoltaic power plant, comprising: Acquire multi-source data and fuse the multi-source data based on a weighted fusion algorithm to obtain effective water level values; Based on the effective water level value, a water accumulation risk prediction model is constructed to obtain the predicted water level value; Based on multi-source data, effective water level values, and predicted water levels, select the current operating mode; Based on preset constraints, construct an optimization model for the objective; Based on the selected current operating mode, the optimal drainage pump power setting value under the target is obtained through the target optimization model.

[0008] Furthermore, The multi-source data includes water level data, meteorological data, photovoltaic power generation data, power grid data, and drainage pump status data; The water level data includes the measured water level value H_actual(t) at time t; The meteorological data includes the measured rainfall intensity at time t, R_current(t), and the predicted rainfall over the next Δt period at time t, R_forecast(t,Δt). The photovoltaic power generation data includes the photovoltaic array power generation P_pv(t) at time t; The power grid data includes the power grid price C_grid(t) at time t; The drainage pump status data includes the drainage pump operating power P_pump(t) at time t; the drainage flow rate Q_pump(t) at time t; and the drainage pump operating temperature T_pump(t) at time t.

[0009] in, The measured water level value H_actual(t) at time t is collected by a water level sensor placed in the drainage pool, with a sampling frequency of 1 time / minute; The measured rainfall intensity R_current(t) at time t is obtained from the actual measurement data of the local weather station; The rainfall forecast for the future time period Δt at time t is obtained through the meteorological data API interface; The grid electricity price C_grid(t) at time t is obtained from the real-time grid electricity price data of the power trading platform API. The photovoltaic array power generation P_pv(t) at time t is obtained through the monitoring system of the photovoltaic power station; The operating power P_pump(t) of the drainage pump at time t is obtained by installing a smart meter in the distribution cabinet of the drainage pump; The drainage flow rate Q_pump(t) at time t is obtained by installing a flow meter on the drainage pipe.

[0010] The operating temperature of the drainage pump at time t, T_pump(t), is obtained by installing a temperature sensor on the motor windings or bearings of the drainage pump.

[0011] Furthermore, The effective water level value H_effective(t) is expressed as: H_effective(t) =ɑ1*H_actual(t)+ɑ2*H_trend(t); in, H_effective(t) represents the effective water level value; H_trend(t) represents the predicted value based on historical water level change trends; H_trend(t) is obtained by using a linear regression model for prediction: H_trend(t)=H_actual(t-1)+ΔH_avg ΔH_avg=[H_actual(t-1)-H_actual(tR)] / (R-1); in, R represents the length of the observation window, which is usually 3-5 time points; ΔH_avg represents the average rate of change of water level over the past R time points; a1 and a2 are weighting coefficients, and a1 + a2 = 1.

[0012] Furthermore, the process of constructing a flood risk prediction model based on the effective water level value to obtain the predicted water level value includes: The flood risk prediction model is expressed as follows: ; in, H_predict(t+Δt) represents the predicted water level at time t+Δt; A represents the catchment area of ​​the photovoltaic field; R_forecast(t,Δt) represents the predicted total rainfall over the time period Δt; β represents the surface permeability coefficient; Q_drain(t) represents the drainage capacity at time t; V_pool represents the effective volume of the drainage pool; Δt represents the prediction time step.

[0013] Furthermore, The current operating modes include: safety priority mode, economic optimization mode, predictive scheduling mode, and value creation mode; If H_effective(t) or H_predict(t+Δt) meets the triggering condition of the safety priority mode, then the safety priority mode is selected; If H_effective(t) and C_grid(t) satisfy the triggering condition of the economic optimization mode, then the economic optimization mode is selected; If R_forecast(t,Δt) satisfies the triggering condition for predictive scheduling mode, then predictive scheduling mode is selected; If H_effective(t) and P_pv(t) satisfy the triggering conditions for the value creation mode, then the value creation mode is selected.

[0014] Furthermore, The triggering conditions for the safety - priority mode are: \(H_{effective}(t)\geq H_{danger}\) or \(H_{predict}(t + \Delta t)\geq H_{danger}\); where \(H_{danger}\) represents the danger water level. The triggering conditions for the economic - optimization mode are: \(H_{warning}\leq H_{effective}(t)\lt H_{danger}\), and \(C_{grid}(t)\) is in the peak period; where \(H_{warning}\) represents the warning water level. The triggering conditions for the predictive - scheduling mode are: \(R_{forecast}(t,\Delta t)\geq R_{threshold}\). Among them, \(R_{threshold}\) is the rainfall threshold. The triggering conditions for the value - creation mode are: \(H_{effective}(t)\lt H_{low}\) and \(P_{pv}(t)\gt P_{min}\). Among them, \(H_{low}\) is the low - water - level threshold; \(P_{min}\) represents the minimum power of the drainage pump.

[0015] Furthermore, based on the preset constraint conditions, an optimization model for the target is constructed, including: The preset constraint conditions include: Power - balance constraint condition: \(P_{pump}(t)=P_{pv}(t)+P_{grid}(t)\); where \(P_{grid}(t)\) represents the power purchased from the power grid at the \(t\) - th time period. Water - level safety constraint condition: \(H_{effective}(t)\lt H_{danger}\). Equipment - limit constraint condition: \(P_{min}\leq P_{pump}(t)\leq P_{max}\).

[0016] Furthermore, based on selecting the current operating mode, the set value of the drainage - pump power under the optimal target is obtained through the optimization model of the target, including: The objective function of the optimization model of the target is expressed as: \(Min(J)=w1\times C_{total}+w2\times T_{accumulated}+w3\times\Delta P_{sum}\). Among them, \(C_{total}\) represents the total operating cost; \(T_{accumulated}\) represents the cumulative operating time of the equipment; \(\Delta P_{sum}\) represents the maximum power fluctuation; \(w1\), \(w2\), \(w3\) are weight coefficients, and \(w1 + w2+w3 = 1\).

[0017] Furthermore, ; Among them, C_grid(t) represents the grid electricity price in time period t, in yuan / kWh; C_grid(t) is obtained from the predicted electricity price curve obtained from the electricity trading market API; for example, peak C_grid(t) is 0.8-1.2 yuan / kWh, normal C_grid(t) is 0.5-0.8 yuan / kWh, and valley C_grid(t) is 0.2-0.5 yuan / kWh.

[0018] P_grid(t) represents the power (kW) purchased from the grid in time period t, which is a decision variable and is determined by the optimization algorithm; Δt: control cycle duration (hours), with a value of 0.25 hours (15 minutes).

[0019] C_reactive represents the reactive power compensation benefit coefficient, with the unit being yuan / kVarh, and is determined according to the local power grid company's reactive power compensation policy; for example, C_reactive is 0.01-0.05 yuan / kVarh. Q_reactive(t) represents the reactive power provided by the drainage pump in time period t, and the unit is kVar.

[0020] Furthermore, ; Where k represents the loss coefficient, which indicates the impact of the unit energy consumption of the drainage pump on the equipment lifespan, with the unit being 1 / kW. It is obtained through equipment lifespan testing or fitting of the lifespan curve provided by the manufacturer. For example, if it is assumed that the drainage pump needs maintenance or replacement when the cumulative operating energy consumption reaches a certain value, then k can be expressed as the reciprocal relationship between energy consumption and lifespan. P_pump(t) represents the active power of the drainage pump in time period t, in kW.

[0021] Δt represents the length of the time period, in hours.

[0022] Furthermore, ; Where P_pump(t) represents the power of the drainage pump in time period t; P_pump(t-1) represents the power of the drainage pump in time period t-1.

[0023] A second aspect of the present invention provides a networked intelligent control system for drainage pumps in a photovoltaic power plant, comprising: Data fusion module: Acquires multi-source data and fuses the multi-source data based on a weighted fusion algorithm to obtain the effective water level value; Model building module: Based on the effective water level value, a water accumulation risk prediction model is built to obtain the predicted water level value; based on the preset constraints, an optimization model of the objective is built. Operating mode selection module: Selects the current operating mode based on multi-source data, effective water level value, and predicted water level value; Target optimization module: Based on the selected current operating mode, the optimal drainage pump power setting value is obtained through the target optimization model.

[0024] A third aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the processors to perform the intelligent network control method for drainage pumps of a photovoltaic power station as described in the first aspect.

[0025] The technical solution proposed in this application can bring the following beneficial effects: 1. This invention obtains effective water level values ​​by weighted fusion of multi-source data. Compared with directly measured water level values, it can reflect the expected water level earlier. Furthermore, through the water accumulation risk prediction model, it can predict the risk of urban flooding several hours in advance based on meteorological forecast data and initiate preventive drainage to reserve safe capacity for core areas.

[0026] 2. This invention employs a multi-objective optimization control strategy. When drainage demand occurs, the system prioritizes scheduling the current surplus photovoltaic power for power supply, achieving zero-cost local utilization of green electricity and reducing grid-purchased electricity. During non-drainage periods, the system controls the water pump frequency converter to operate in a specific state, making it a controllable reactive load to dynamically compensate for the reactive power of the power station.

[0027] 3. This invention avoids the sudden start-up of all water pumps at full load under extreme weather conditions through predictive scheduling, and instead switches to smooth, advance, stepped operation, reducing mechanical and electrical shocks.

[0028] The above description is merely an overview of the technical solution disclosed herein. In order to better understand the technical means of this disclosure and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this disclosure more apparent and understandable, specific embodiments of this disclosure are described below. Attached Figure Description

[0029] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this disclosure. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a flowchart illustrating the steps of a photovoltaic power station drainage pump network intelligent control method provided in the embodiments of this specification; Figure 2 This is a schematic diagram of a structural damage recognition system based on multi-scale feature fusion provided in the embodiments of this specification. Detailed Implementation

[0030] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. The technical solutions provided by various embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0031] This application provides a method and system for networked intelligent control of drainage pumps in a photovoltaic power plant.

[0032] The first aspect of the technical solution of the present invention is as follows: Figure 1 The diagram shown is a flowchart illustrating the steps of a photovoltaic power station drainage pump network intelligent control method provided in an embodiment of this specification. A method for network-connected intelligent control of drainage pumps in a photovoltaic power plant is provided, including: S101. Obtain multi-source data and fuse the multi-source data based on a weighted fusion algorithm to obtain an effective water level value; S102. Based on the effective water level value, construct a water accumulation risk prediction model to obtain the predicted water level value; S103. Based on multi-source data, effective water level values, and predicted water level values, select the current operating mode; S104. Based on the preset constraints, construct an optimization model for the objective; S105. Based on the selection of the current operating mode, the optimal drainage pump power setting value under the target is obtained through the target optimization model.

[0033] Furthermore, The multi-source data includes water level data, meteorological data, photovoltaic power generation data, power grid data, and drainage pump status data; The water level data includes the measured water level value H_actual(t) at time t; The meteorological data includes the measured rainfall intensity at time t, R_current(t), and the predicted rainfall over the next Δt period at time t, R_forecast(t,Δt). The photovoltaic power generation data includes the photovoltaic array power generation P_pv(t) at time t; The power grid data includes the power grid price C_grid(t) at time t; The drainage pump status data includes the drainage pump operating power P_pump(t) at time t; the drainage flow rate Q_pump(t) at time t; and the drainage pump operating temperature T_pump(t) at time t.

[0034] Among them, for example, The measured water level value H_actual(t) at time t is collected by a water level sensor placed in the drainage pool, with a sampling frequency of 1 time / minute; The measured rainfall intensity R_current(t) at time t is obtained from the actual measurement data of the local weather station; The rainfall forecast for the future time period Δt at time t is obtained through the meteorological data API interface; The grid electricity price C_grid(t) at time t is obtained from the real-time grid electricity price data of the power trading platform API. The photovoltaic array power generation P_pv(t) at time t is obtained through the monitoring system of the photovoltaic power station; The operating power P_pump(t) of the drainage pump at time t is obtained by installing a smart meter in the distribution cabinet of the drainage pump; The drainage flow rate Q_pump(t) at time t is obtained by installing a flow meter on the drainage pipe.

[0035] The operating temperature of the drainage pump at time t, T_pump(t), is obtained by installing a temperature sensor on the motor windings or bearings of the drainage pump.

[0036] Furthermore, The effective water level value H_effective(t) is expressed as: H_effective(t) =ɑ1*H_actual(t)+ɑ2*H_trend(t); in, H_effective(t) represents the effective water level value; H_trend(t) represents the predicted value based on historical water level change trends; H_trend(t) is obtained by prediction using a linear regression model: H_trend(t)=H_actual(t-1)+ΔH_avg; ΔH_avg=[H_actual(t-1)-H_actual(tR)] / (R-1); in, R represents the length of the observation window, which is usually 3-5 time points; ΔH_avg represents the average rate of change of water level over the past R time points; a1 and a2 are weighting coefficients, and a1+a2=1. a1 and a2 are set by experts based on historical data from the past 10 years. For example, a1=0.8 and a2=0.2.

[0037] Further, constructing a waterlogging risk prediction model based on the effective water level value to obtain a water level prediction value includes: The waterlogging risk prediction model is expressed as: ; Among them, H_predict(t + Δt) represents the predicted water level at time t + Δt; A represents the catchment area of the photovoltaic field, unit: square meters (m 2 ), obtained through site survey; R_forecast(t, Δt) represents the total rainfall within the predicted time period Δt; β represents the surface infiltration coefficient, with a value range of 0 - 1, determined through soil type tests; Q_drain(t) represents the drainage capacity at time t, unit: cubic meters per hour (m 3 / h); V_pool represents the effective volume of the drainage pool, unit: cubic meters (m 3 ); Δt represents the predicted time step, unit: hours (h).

[0038] Judging the risk level based on H_predict(t + Δt), including low risk, medium risk, and high risk; Among them, The determination condition for low risk is: H_predict(t + Δt) < H_warning, where H_warning is the warning water level; The determination condition for medium risk is: H_warning ≤ H_predict(t + Δt) < H_danger, where H_danger is the danger water level; The determination condition for high risk is: H_predict(t + Δt) ≥ H_danger.

[0039] Further, The current operation mode includes: safety - first mode, economic optimization mode, predictive scheduling mode, and value - creation mode; If H_effective(t) or H_predict(t + Δt) meets the trigger condition of the safety - first mode, then select the safety - first mode; If H_effective(t) and C_grid(t) meet the trigger condition of the economic optimization mode, then select the economic optimization mode; If R_forecast(t, Δt) meets the trigger condition of the predictive scheduling mode, then select the predictive scheduling mode; If \(H_{effective}(t)\) and \(P_{pv}(t)\) meet the triggering conditions of the value creation mode, then the value creation mode is selected.

[0040] Furthermore, The triggering conditions for the safety - first mode are: \(H_{effective}(t)\geq H_{danger}\) or \(H_{predict}(t + \Delta t)\geq H_{danger}\); where \(H_{danger}\) represents the danger water level;示例性的Δt取值为1,单位为h; \(H_{danger}\) represents the highest safety - limit water level of the photovoltaic power station's drainage system; when \(H_{effective}(t)\) or \(H_{predict}(t + 1)\) reaches or exceeds this value, the power station will face serious safety risks.

[0041] The triggering conditions for the economic - optimization mode are: \(H_{warning}\leq H_{effective}(t)\lt H_{danger}\), and \(C_{grid}(t)\) is in the peak period; where \(H_{warning}\) represents the warning water level.

[0042] The triggering conditions for the predictive - scheduling mode are: \(R_{forecast}(t,\Delta t)\geq R_{threshold}\); where \(R_{threshold}\) is the rainfall threshold.

[0043] The triggering conditions for the value creation mode are: \(H_{effective}(t)\lt H_{low}\) and \(P_{pv}(t)\gt P_{min}\); where \(P_{min}\) represents the minimum power of the drainage pump; \(H_{low}\) is the low - water - level threshold; when the water level is lower than this value, the drainage demand is basically eliminated, and the system can consider switching to the energy - saving mode or the value creation mode.

[0044] Exemplarily, the specific operation method is as follows: Mode 1: Safety - first mode, including the following conditions, Triggering conditions: \(H_{effective}(t)\geq H_{danger}\) or \(H_{predict}(t + 1)\geq H_{danger}\) Control strategy: All drainage pumps operate at full power to ensure the safety of the power station first; When this strategy is enabled, immediately send full - power start commands to all available drainage pumps, the system shields all economic - optimization algorithms, and records this event as "emergency drainage"; Exit condition: This mode can be exited if and only if H_effective(t) < H_safe and H_predict(t + 1) < H_warning; H_safe is the safe exit water level, which is lower than H_warning. This design is to prevent the mode from oscillating due to frequent fluctuations of the water level near the threshold.

[0045] Mode 2: Economic optimization mode: This mode is the main operating mode of the system when there is no emergency risk. Its core is to minimize the operating cost on the premise of meeting the drainage demand; it includes the following conditions, Trigger condition: H_warning ≤ H_effective(t) < H_danger and the electricity price is in the peak period. Control strategy:优先使用光伏电力排水,结合电网优化运行; First, calculate the drainage demand power P_need: According to the current water level H_actual(t), the target water level H_target (usually set as the intermediate value between H_low and H_warning), and the volume of the drainage pool, estimate the drainage volume and time required to lower the water level to H_target, and then inversely calculate the required average drainage power; Then, optimize the power source: (1)、光伏电力优先:若P_pv(t)≥P_need,则排水泵所需电力全部由当前光伏富余功率提供;控制指令为启动水泵,功率设定为P_need; (2)、混合供电:若P_pv(t)<P_need,则执行混合调度:首先,分配全部可用光伏富余功率P_pv(t)给排水泵;对于功率缺口ΔP=P_need-P_pv(t),判断电网电价C_grid(t),若处于谷电或平电时段,则从电网取电补充;若处于峰电时段且缺口不大,可进入“等待或降功率运行”子状态,适当延长排水时间,等待光伏功率上升或电价降低; Finally, send the calculated power setting value P_pump_set(t) of each drainage pump to the frequency converter or start-stop controller of each drainage pump; Exit condition: Exit immediately when the conditions for a higher-level mode (safety first, predictive scheduling) are triggered; or when the water level drops below H_low, enter the "value creation mode" or standby.

[0046] Mode 3: Predictive scheduling mode: Actively intervene before rainfall to smooth the drainage load and avoid impacting the power system and equipment of the power station; it includes the following conditions, Trigger condition: R_forecast(t,3)≥R_threshold, where R_threshold is the rainfall threshold control strategy: Start the drainage pump in advance to lower the water level and reserve capacity for the upcoming rainfall. The goal is to pre-lower the water level to a lower target value H_pre_target (such as H_low) before the rainfall arrives, so as to create storage space for the upcoming rainwater. First, calculate the amount of water that needs to be drained in advance: V_pre=[H_effective(t)-H_pre_target]×V_pool; Then, evenly distribute the drainage task over the future Δt period, and calculate the average required drainage power P_pre_avg; Since there is sufficient time, P_pre_avg is usually much lower than the rated power of the pump. Then, take P_pre_avg as the new drainage demand, and call the power source optimization logic of the "economic optimization mode" to find the drainage scheduling plan with the lowest cost in the future Δt period. This usually means using the predicted photovoltaic power generation peak in this period for drainage as much as possible. The system generates a pre-scheduling plan for the next few hours and starts to operate the pump at low power according to the plan. Exit condition: When the predicted rainfall event starts or ends and the actual water level risk is lifted, smoothly transition to the "economic optimization mode" or the "safety first mode" according to the current water level and power situation.

[0047] Mode 4: Value creation mode: This mode creates additional value for the power station by using the idle drainage pump capacity on the premise of ensuring absolute safety; including the following conditions, Trigger condition: H_effective(t)<H_low and P_pv(t)>P_min; Control strategy: The drainage pump participates in reactive power compensation as a controllable load to improve the power factor. The goal is to use the drainage pump as a controllable reactive load to improve the power factor at the grid connection point of the power station. First, monitor the real-time power factor PF(t) at the grid connection point of the power station; If PF(t) is lower than the target value (such as 0.95), the system calculates the reactive power Q_need(t) that needs to be compensated; Control the frequency converter of the drainage pump to make it work in a specific low active power and high reactive power (inductive or capacitive) state; For example, let the pump motor run idle or at low speed, and control it through the frequency converter to absorb or emit the specified reactive power Q_pump(t) from the power grid, where |Q_pump(t)|≤Q_pump_max (the maximum reactive power capacity of the pump); In this mode, the pump does not act as a drainage device but as a static var compensator. Exit conditions: Exit immediately when any trigger condition is no longer met; in particular, when the water level rises above H_low, this mode must be exited immediately in preparation for drainage.

[0048] The decision is made according to a fixed priority: safety priority mode > predictive scheduling mode > economic optimization mode > value creation mode. Switch to this mode immediately as long as the "safety priority mode" conditions are met.

[0049] If the safety conditions are not met, but the "predictive scheduling mode" conditions are met, then switch to predictive scheduling mode.

[0050] If neither of the above conditions is met, and the water level is within the normal range, then the economic optimization mode will be used.

[0051] Value creation mode can only be entered when the water level is extremely low and there is a surplus of photovoltaic power.

[0052] Furthermore, the construction of the optimization model for the objective based on preset constraints includes: The preset constraints include: Power balance constraint: P_pump(t) = P_pv(t) + P_grid(t); where P_grid(t) represents the power purchased from the grid in time period t, which is a decision variable and is determined by the optimization algorithm; Water level safety constraint: H_effective(t) <H_danger; Equipment constraints: P_min≤P_pump(t)≤P_max; P_min is the minimum power of the drainage pump; P_max is the maximum power of the drainage pump.

[0053] Furthermore, based on the selected current operating mode, the optimal drainage pump power setting value under the target is obtained through the target optimization model, including: The objective function of the optimization model for the target is expressed as: Min(J)=w1*C_total+w2*T_accumulated+w3*ΔP_sum; Where C_total represents the total operating cost; T_accumulated represents the cumulative operating time of the equipment; ΔP_sum represents the maximum power fluctuation; C_total, T_accumulated, and T_accumulated are normalized using the Min-Max normalization method; w1, w2, and w3 are weight coefficients determined by the analytic hierarchy process, and w1+w2+w3=1.

[0054] Furthermore, ; in, C_grid(t) represents the grid electricity price in time period t, in yuan / kWh; C_grid(t) is obtained from the predicted electricity price curve obtained from the electricity trading market API; for example, peak C_grid(t) is 0.8-1.2 yuan / kWh, normal C_grid(t) is 0.5-0.8 yuan / kWh, and valley C_grid(t) is 0.2-0.5 yuan / kWh.

[0055] P_grid(t) represents the power (kW) purchased from the grid in time period t, which is a decision variable and is determined by the optimization algorithm; Δt: control cycle duration (hours), with a value of 0.25 hours (15 minutes).

[0056] C_reactive represents the reactive power compensation benefit coefficient, with the unit being yuan / kVarh, and is determined according to the local power grid company's reactive power compensation policy; for example, C_reactive is 0.01-0.05 yuan / kVarh. Q_reactive(t) represents the reactive power provided by the drainage pump in time period t, and the unit is kVar.

[0057] Furthermore, ; Where k represents the loss coefficient, which indicates the impact of the unit energy consumption of the drainage pump on the equipment lifespan, with the unit being 1 / kW. It is obtained through equipment lifespan testing or fitting of the lifespan curve provided by the manufacturer. For example, if it is assumed that the drainage pump needs maintenance or replacement when the cumulative operating energy consumption reaches a certain value, then k can be expressed as the reciprocal relationship between energy consumption and lifespan. P_pump(t) represents the active power of the drainage pump in time period t, in kW.

[0058] Δt represents the length of the time period, in hours.

[0059] Furthermore, ; Where P_pump(t) represents the power of the drainage pump in time period t; P_pump(t-1) represents the power of the drainage pump in time period t-1.

[0060] The optimal power setpoint for the drainage pump under the target objective is obtained through the optimization model. The solution steps are as follows: 1. Initialization: Based on the current operating mode, obtain the current state and set the initial solution; 2. Gradient Calculation: Calculate the gradient of the objective function. J; 3. Constraint processing: Convert constraints into standard form; 4. Iterative solution: Iterate using the selected optimization algorithm; 5. Convergence Criterion: Satisfies J is less than the threshold or the maximum number of iterations has been reached; 6. Output: Output the optimal drainage pump power setpoint. .

[0061] The solution process for the optimization model of the above objectives is existing technology and will not be described in detail in this embodiment.

[0062] The second aspect of the technical solution of the present invention is as follows: Figure 2 The diagram shown is a structural schematic of a structural damage recognition system based on multi-scale feature fusion provided in an embodiment of this specification. A networked intelligent control system 200 for drainage pumps in a photovoltaic power plant is provided, comprising: Data fusion module 201: Acquires multi-source data and fuses the multi-source data based on a weighted fusion algorithm to obtain an effective water level value; Model building module 202: Based on the effective water level value, construct a water accumulation risk prediction model to obtain the predicted water level value; based on the preset constraints, construct an optimization model for the target. Operation mode selection module 203: Selects the current operation mode based on multi-source data, effective water level value, and predicted water level value; Target optimization module 204: Based on the selection of the current operating mode, the optimal drainage pump power setting value under the target is obtained through the target optimization model.

[0063] A third aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the processors to perform the intelligent network control method for drainage pumps of a photovoltaic power station as described in the first aspect.

[0064] This embodiment includes the following levels for implementing the operation of the data fusion module, model building module, operation mode selection module, and target optimization module; Specifically, it includes the perception layer, execution layer, communication layer, and cloud platform layer.

[0065] The sensing layer includes a water level sensor, which is placed at key locations in the drainage pool to monitor the water level in real time, in meters (m); a meteorological sensor, which monitors rainfall, temperature, humidity, etc.; a meteorological data acquisition module, which is used to access meteorological forecast data and obtain future rainfall prediction data; and a power monitoring module, which includes monitoring photovoltaic power generation P_pv(t) and grid electricity price C_grid(t).

[0066] The execution layer includes: a drainage pump with variable frequency control and a power range of [P_min, P_max]; and an intelligent control cabinet containing a PLC / controller that receives control commands and adjusts the drainage pump power.

[0067] The communication layer includes: local communication: the sensor communicates with the control cabinet via wired (e.g., 4-20mA) or wireless (e.g., LoRa); remote communication: the control cabinet communicates with the cloud platform via 4G / 5G or Ethernet.

[0068] The cloud platform layer includes: storing historical data, running weighted fusion algorithms, water accumulation risk prediction models, target optimization models, and determining the appropriate operating mode for the current state.

[0069] As can be seen from the above examples, the effective water level value obtained by weighted fusion of multi-source data can reflect the expected water level earlier than the directly measured water level value. Furthermore, the water accumulation risk prediction model can predict the risk of urban flooding several hours in advance based on meteorological forecast data and initiate preventive drainage to reserve safe capacity for core areas.

[0070] As can be seen from the above embodiments, through a multi-objective optimization control strategy, when drainage demand occurs, the system prioritizes scheduling the current surplus photovoltaic power for power supply, realizing zero-cost local utilization of green electricity and reducing grid-purchased electricity; during non-drainage periods, the system controls the water pump frequency converter to work in a specific state, making it a controllable reactive load to dynamically compensate the power station for reactive power.

[0071] As can be seen from the above examples, predictive scheduling avoids the sudden start-up of all water pumps at full capacity under extreme weather conditions, and instead enables smooth, advance, stepped operation, reducing mechanical and electrical shocks.

[0072] This embodiment can divide the method into functional modules based on the above method example. For example, each function can be assigned to a separate module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0073] When dividing the functions into modules corresponding to each function, it may include: a data fusion module, a model building module, a running mode selection module, and a target optimization module. It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here.

[0074] This embodiment also provides a computer-readable storage medium (including but not limited to disk storage, CD-ROM, optical storage, etc.) storing computer program code. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to realize the intelligent network control method for drainage pumps of a photovoltaic power station provided in the above embodiment.

[0075] This embodiment also provides a computer program product. When the computer program product is run on a computer, it causes the computer to perform the aforementioned steps to realize the intelligent network control method for drainage pumps of a photovoltaic power station provided in the above embodiment. The beneficial effects of the above embodiments can be found in the corresponding methods described above, and will not be repeated here.

[0076] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0077] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms. In the description of this disclosure, it should be understood that if terms such as "upper," "lower," "front," "rear," "left," and "right" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, they are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the indicated position or element must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this disclosure.

[0078] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0079] The above are merely embodiments of this disclosure and are not intended to limit the scope of this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of the claims of this disclosure.

Claims

1. A method for networked intelligent control of drainage pumps in a photovoltaic power station, characterized in that, Acquire multi-source data and fuse the multi-source data based on a weighted fusion algorithm to obtain effective water level values; Based on the effective water level value, a water accumulation risk prediction model is constructed to obtain the predicted water level value; Based on multi-source data, effective water level values, and predicted water levels, select the current operating mode; Based on preset constraints, construct an optimization model for the objective; Based on the selected current operating mode, the optimal drainage pump power setting value under the target is obtained through the target optimization model.

2. The intelligent control method for networked drainage pumps in a photovoltaic power station according to claim 1, characterized in that, The multi-source data includes water level data, meteorological data, photovoltaic power generation data, power grid data, and drainage pump status data; The water level data includes the measured water level value H_actual(t) at time t; The meteorological data includes the measured rainfall intensity at time t, R_current(t), and the predicted rainfall over the next Δt period at time t, R_forecast(t,Δt). The photovoltaic power generation data includes the photovoltaic array power generation P_pv(t) at time t; The power grid data includes the power grid price C_grid(t) at time t; The drainage pump status data includes the drainage pump operating power P_pump(t) at time t; the drainage flow rate Q_pump(t) at time t; and the drainage pump operating temperature T_pump(t) at time t.

3. The intelligent control method for networked drainage pumps in a photovoltaic power station according to claim 2, characterized in that, The effective water level value H_effective(t) is expressed as: H_effective(t) =ɑ1*H_actual(t)+ɑ2*H_trend(t); in, H_effective(t) represents the effective water level value; H_trend(t) represents the predicted value based on historical water level change trends; a1 and a2 are weighting coefficients, and a1 + a2 = 1.

4. The intelligent control method for networked drainage pumps in a photovoltaic power station according to claim 2, characterized in that, The method for constructing a flood risk prediction model based on effective water level values ​​to obtain predicted water level values ​​includes: The flood risk prediction model is expressed as follows: ； in, H_predict(t+Δt) represents the predicted water level at time t+Δt; A represents the catchment area of ​​the photovoltaic field; R_forecast(t,Δt) represents the predicted total rainfall over the time period Δt; β represents the surface permeability coefficient; Q_drain(t) represents the drainage capacity at time t; V_pool represents the effective volume of the drainage pool; Δt represents the prediction time step.

5. The intelligent control method for networked drainage pumps in a photovoltaic power station according to claim 2, characterized in that, The current operating modes include: safety priority mode, economic optimization mode, predictive scheduling mode, and value creation mode; If H_effective(t) or H_predict(t+Δt) meets the triggering condition of the safety priority mode, then the safety priority mode is selected; If H_effective(t) and C_grid(t) satisfy the triggering condition of the economic optimization mode, then the economic optimization mode is selected; If \(R_{forecast}(t,\Delta t)\) satisfies the trigger condition of the predictive scheduling mode, then the predictive scheduling mode is selected; If \(H_{effective}(t)\) and \(P_{pv}(t)\) satisfy the trigger condition of the value creation mode, then the value creation mode is selected.

6. A method for intelligent control of drainage pump networking in a photovoltaic power station according to claim 5, characterized in that, The trigger condition of the safety - first mode is: \(H_{effective}(t)\geq H_{danger}\) or \(H_{predict}(t + \Delta t)\geq H_{danger}\); where \(H_{danger}\) represents the danger water level; The trigger condition of the economic optimization mode is: \(H_{warning}\leq H_{effective}(t)<H_{danger}\), and \(C_{grid}(t)\) is in the peak period; where \(H_{warning}\) represents the warning water level; The trigger condition of the predictive scheduling mode is: \(R_{forecast}(t,\Delta t)\geq R_{threshold}\); Among them, \(R_{threshold}\) is the rainfall threshold; The trigger condition of the value creation mode is: \(H_{effective}(t)<H_{low}\) and \(P_{pv}(t)>P_{min}\); Among them, \(H_{low}\) is the low water level threshold; \(P_{min}\) represents the minimum power of the drainage pump.

7. The intelligent control method for networked drainage pumps in a photovoltaic power station according to claim 5, characterized in that, Based on the preset constraint conditions, an optimization model of the target is constructed, including: The preset constraint conditions include: Power balance constraint condition: \(P_{pump}(t)=P_{pv}(t)+P_{grid}(t)\); where \(P_{grid}(t)\) represents the power purchased from the power grid at the \(t\) - th time period; Water level safety constraint condition: \(H_{effective}(t)<H_{danger}\); Equipment limit constraint condition: \(P_{min}\leq P_{pump}(t)\leq P_{max}\).

8. The intelligent control method for networked drainage pumps in a photovoltaic power station according to claim 3, characterized in that, Based on the selection of the current operation mode, the set value of the drainage pump power under the optimal target is obtained through the optimization model of the target, including: The objective function of the optimization model of the target is expressed as: \(Min(J)=w1*C_{total}+w2*T_{accumulated}+w3*\Delta P_{sum}\); Among them, \(C_{total}\) represents the total operation cost; \(T_{accumulated}\) represents the cumulative operation time of the equipment; \(\Delta P_{sum}\) represents the maximum power fluctuation; \(w1\), \(w2\), \(w3\) are weight coefficients, and \(w1 + w2+w3 = 1\).

9. A networked intelligent control system for drainage pumps in a photovoltaic power station, characterized in that... The system includes: Data fusion module: Obtain multi - source data, and fuse the multi - source data based on the weighted fusion algorithm to obtain the effective water level value; Model construction module: Based on the effective water level value, construct a waterlogging risk prediction model to obtain the water level prediction value; Based on the preset constraint conditions, construct an optimization model of the target; Operation mode selection module: Based on multi - source data, effective water level value, and water level prediction value, select the current operation mode; Target optimization module: Based on the selection of the current operation mode, obtain the set value of the drainage pump power under the optimal target through the optimization model of the target.

10. A computer-readable storage medium, characterized in that, It stores instructions that, when executed by one or more processors, cause the processors to perform the intelligent network control method for drainage pumps of a photovoltaic power station as described in any one of claims 1-8.

Images