A new energy power generation intelligent control method and system based on big data

Abstract

The application discloses a new energy power generation intelligent control method and system based on big data, and concretely relates to the field of new energy power generation control, and comprises the steps of multi-dimensional data acquisition, data preprocessing, model analysis, control decision generation and execution and feedback correction: through the collection of environment, equipment, power grid and historical data by Internet of Things sensors, after the elimination of outliers, space-time alignment and normalization processing, a multi-parameter coupling output model and an optimization objective function are constructed, the photovoltaic, wind power and energy storage control instructions are generated in combination with equipment safety and power grid constraints, and the weight coefficient is corrected when the deviation exceeds the threshold. The system comprises data acquisition, processing, mathematical modeling analysis, control execution and storage modules, the scheme improves the control precision and system stability, prolongs the equipment life, has strong compatibility, can be directly applied to various scale new energy power stations, does not need to be re-adjusted, and effectively supports the efficient and stable operation of new energy power generation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy power generation control, more particularly, the present application relates to a new energy power generation intelligent control method and system based on big data. BACKGROUND

[0002] New energy power generation such as photovoltaic and wind power is a core component of clean energy, and its large-scale application is of great significance to energy structure transformation. Existing new energy power generation control technology mainly falls into two categories: one is the traditional control method, which relies on single physical parameters such as light intensity and wind speed feedback regulation, and realizes device control through preset empirical formula or simple logic; the other is the intelligent control method based on machine learning, which uses massive data to train prediction model and generates control strategy to adapt to environmental and load fluctuations.

[0003] However, in actual use, there are still some shortcomings, such as the traditional method relying on single parameter, such as adjusting the angle of photovoltaic panel only by light intensity, without considering the coupling relationship of multiple parameters such as the influence of temperature on photovoltaic conversion efficiency and the synergistic effect of wind speed and wind turbine pitch angle, resulting in low control precision and inability to adapt to complex environmental changes. The control method based on machine learning needs massive sample data to train the model, and is strongly dependent on data quality. In data sparse or extreme conditions such as typhoon and snowstorm, the generalization ability is poor. At the same time, the model is a "black box" structure, and the logic of the control strategy is not interpretable, which is difficult to meet the rigid requirements of power system on control reliability such as fault tracing and safety audit. Moreover, the existing technology cannot balance multiple objectives such as "maximizing power generation", "minimizing power fluctuation" and "equipment safety protection", often pursuing power generation at the expense of equipment wear and tear such as wind turbine overload operation, or sacrificing power generation efficiency for stable output. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides the following technical scheme: a new energy power generation intelligent control method based on big data, comprising: S1: multi-dimensional data acquisition: collecting new energy power generation data through Internet of Things sensor array, smart meter and power grid dispatching interface;

[0005] S2: data preprocessing: removing outliers, aligning time and space, and normalizing the collected new energy power generation data;

[0006] S3: model analysis: based on the data processed in S2, a multi-parameter coupled output model and an optimization objective function are constructed, and device safety constraints and power grid constraints are set;

[0007] The multi-parameter coupled output model includes a photovoltaic output model and a wind power output model;

[0008] The device safety constraints include the relevant parameter limits of photovoltaic panels, wind turbines and energy storage batteries.

[0009] The grid constraints include a rated voltage and a frequency range;

[0010] S4: Control decision generation and execution: generate photovoltaic control instructions, wind power control instructions and energy storage control instructions based on the results of S3 model analysis combined with constraint conditions and execute them;

[0011] S5: Feedback correction: calculate the control deviation value, and when the control deviation value is greater than the preset threshold, correct the weight coefficient in the optimization model.

[0012] Preferably, a new energy power generation intelligent control system based on big data comprises a data acquisition module composed of an illumination sensor, a temperature sensor, a wind speed sensor, a voltage and current sensor and an intelligent electric meter, which is connected with a data processing module through an RS485 bus;

[0013] The data processing module adopts an FPGA chip and is used for realizing data outlier elimination, time-space alignment and normalization processing;

[0014] The mathematical modeling analysis module is used for deploying a multi-parameter coupled output model and an optimization objective function based on an industrial control computer;

[0015] The control execution module comprises a PLC controller, a servo driver and a relay, and is used for receiving instructions from the analysis module and driving a photovoltaic light tracking motor, a fan variable pitch mechanism and an energy storage inverter;

[0016] The data storage module adopts an HBase distributed database and is used for supporting historical data query and model parameter update.

[0017] Preferably, the photovoltaic output model is as follows:

[0018] Pp=I×S×η0×[1-α×(Tp-25)]×kshade, wherein S is the total area of the photovoltaic panel, which is determined by the photovoltaic array design parameters, η0 is the conversion efficiency under standard working conditions, α is the temperature coefficient, which is 0.004 / ℃, kshade is the shading coefficient, which is obtained by fitting historical shading data: kshade=0.98-0.02×D, and D is the shading area ratio.

[0019] Preferably, the wind power output model is as follows:

[0020] When v

[0021] When vci≤v≤vN, Pw=PN×(v³-vci³) / (vN³-vci³);

[0022] When vN<v≤vco, Pw=PN;

[0023] Wherein vci is the cut-in wind speed, taking the value of 3 m / s, vN is the rated wind speed, taking the value of 12 m / s, vco is the cut-out wind speed, taking the value of 25 m / s, PN is the rated power of the fan, and a pitch angle correction term is introduced: Pw'=Pw×(1-0.01×β), to obtain the corrected actual output.

[0024] Preferably, the optimization objective function is as follows:

[0025] minJ=λ1×(1-Pactual / Pmax)+λ2×ΔP+λ3×(ΔT / Tmax), wherein Pactual is the current actual output of wind power generation or photovoltaic power generation, Pmax is the theoretical maximum output calculated by the wind power output model or the photovoltaic output model, ΔP is the output fluctuation value, ΔP=|Pactual-Pavg|, Pavg is the average output in the previous 10 minutes, ΔT is the equipment temperature deviation, ΔT=Tp-45℃ for a photovoltaic system and ΔT=Tg-80℃ for a wind power system, Tmax is the maximum allowable temperature deviation of the equipment, Tmax=40℃ for a photovoltaic system and Tmax=20℃ for a wind power system, λ1, λ2, and λ3 are weight coefficients, satisfying λ1+λ2+λ3=1, wherein:

[0026] When the grid load L≥80% of the rated load, λ1=0.6, λ2=0.3, and λ3=0.1;

[0027] When the grid load 50%≤L<80% of the rated load, λ1=0.4, λ2=0.4, and λ3=0.2;

[0028] When the grid load L<50% of the rated load, λ1=0.2, λ2=0.5, and λ3=0.3.

[0029] Preferably, the equipment safety constraints are as follows:

[0030] Photovoltaic panel: Tp≤85℃, Up≤1.1×UpN, wherein UpN is the rated voltage, Tp is the working temperature of the photovoltaic panel, and Up is the output voltage of the photovoltaic panel;

[0031] Fan: n≤1800 r / min, β≤30°, wherein n is the fan speed, and excessive speed may cause accelerated wear of mechanical parts or even safety accidents, and β is the pitch angle of the fan, used to adjust the efficiency and stability of wind energy absorption by the fan;

[0032] Energy storage battery: SOC≥20%, and the charging and discharging power≤0.5×PbN, wherein PbN is the rated power of the battery, and SOC is the state of charge of the battery;

[0033] The grid constraints are as follows:

[0034] Ug∈[0.95×UN,1.05×UN], wherein UN is the rated voltage, Ug is the grid voltage;

[0035] f∈[49.5Hz,50.5Hz], wherein f is the grid frequency.

[0036] Preferably, the photovoltaic control instruction comprises:

[0037] Light tracking angle adjustment: the real-time light tracking angle θt of the photovoltaic panel is calculated by the formula θt=arcsin(Idirect / Itotal), wherein Idirect represents the direct light intensity (unit: W / m2), and Itotal is the total light intensity.

[0038] Inverter MPPT parameter: the voltage reference value Uref is determined according to the formula Uref=UpN×(Popt / Pmax), wherein Popt is the optimal output of the photovoltaic system solved by the optimization model.

[0039] The wind power control instruction comprises:

[0040] Pitch angle adjustment: the pitch angle β is calculated by the formula β=0.5×(v-vN), but this adjustment strategy is only enabled when the wind speed v>vN.

[0041] Fan start-stop control: when the wind speed v<vci and lasts for 10 minutes, a stop command is issued at this time, and when the wind speed v>vco, the equipment is immediately stopped for protection.

[0042] The energy storage control instruction comprises:

[0043] Charging control: when the actual power generation power is greater than the optimal output calculated by the optimization model, and the state of charge SOC of the energy storage system is less than 80%, the charging operation is started, and the charging power Pch=Pactual-Popt.

[0044] Discharge control: when Pactual<Popt and SOC≥20%, the energy storage system performs discharge operation, and the discharge power Pdis=Popt-Pactual.

[0045] Technical effects and advantages of the present application:

[0046] The present application constructs a multi-parameter coupled output model, quantifies the influence of multiple factors on the output, reduces the output prediction error, and at the same time, the model is based on physical laws and mathematical formulas, does not need to rely on samples, and can adapt to extreme environment and complex working conditions.

[0047] The application adopts a deterministic mathematical model to replace a machine learning ''black box'' model, logic of a control strategy is clear and traceable, requirements of a power system on reliability and safety are met, and the application can still stably operate in a data sparse scene without massive sample training;

[0048] The application adjusts a weight coefficient in real time according to power grid load through an optimization objective function of a dynamic weight, and considers power generation, fluctuation suppression and equipment protection. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 A method flowchart of the application is shown in the figure;

[0050] Figure 2 A system structure schematic diagram of the application is shown in the figure. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0052] As shown in the figure, a new energy power generation intelligent control method based on big data includes: Figure 1

[0053] S1: Multi-dimensional data acquisition: through an Internet of Things sensor array, a smart meter and a power grid dispatching interface, new energy power generation data is acquired;

[0054] The new energy power generation data includes:

[0055] Environmental parameters: including light intensity (I, unit: W / m², sampling frequency: 1Hz) of a photovoltaic area, environmental temperature (T, unit: ℃, sampling frequency: 1Hz), wind speed (v, unit: m / s, sampling frequency: 1Hz) of a wind power area, wind direction (θ, unit: °, sampling frequency: 1Hz);

[0056] Device parameters: including working temperature (Tp, unit: ℃, sampling frequency: 1Hz) of a photovoltaic panel, output voltage (Up, unit: V, sampling frequency: 1Hz), output current (Ip, unit: A, sampling frequency: 1Hz) of the photovoltaic panel, rotating speed (n, unit: r / min, sampling frequency: 1Hz) of a fan, pitch angle (β, unit: °, sampling frequency: 1Hz), output power (Pw, unit: kW, sampling frequency: 1Hz) of the fan, state of charge (SOC, unit: %, sampling frequency: 0.1Hz) of an energy storage battery;

[0057] ​Grid parameters: including grid-connected point voltage (Ug, unit kV, sampling frequency 1 Hz), frequency (f, unit Hz, sampling frequency 1 Hz), real-time load (L, unit kW, sampling frequency 1 Hz);

[0058] Historical data: including hourly power generation output, environmental parameters and equipment operation parameters in the past 3 years, stored in a distributed database.

[0059] S2: data preprocessing: the collected new energy power generation data is subjected to outlier elimination, space-time alignment and normalization processing;

[0060] It should be specifically pointed out that the operation method of the outlier elimination is as follows: 3σ criterion is used to process each parameter, for the parameters subject to normal distribution, the data deviating from the mean value by 3 times the standard deviation is eliminated, for the parameters not subject to normal distribution, the quartile method is used to eliminate outliers, and the calculation formula is:

[0061] The outlier determination range is [Q1-1.5*IQR, Q3+1.5*IQR]

[0062] Wherein Q1 is the lower quartile, Q3 is the upper quartile, and IQR=Q3-Q1.

[0063] The operation method of the space-time alignment is as follows: linear interpolation method is used to unify all parameters into 1Hz time series data, and the interpolation formula is:

[0064] x(t)=x(t0)+(x(t1)-x(t0))×(t-t0) / (t1-t0)

[0065] Wherein t0, t1 are adjacent sampling time, and t is the interpolation time;

[0066] The operation method of the normalization processing is as follows: the data is mapped to the interval [0, 1], and the formula is:

[0067] x norm =(x-x min ) / (xmax-x min );

[0068] Wherein x is the original data, x min , x max is the minimum value and the maximum value in the historical data.

[0069] S3: model analysis: based on the data processed by S2, a multi-parameter coupled output model and an optimization objective function are constructed, and device safety constraints and power grid are set as constraint conditions;

[0070] The multi-parameter coupled output model includes a photovoltaic output model and a wind power output model.

[0071] The device safety constraints include photovoltaic panel, fan, energy storage battery related parameter limits;

[0072] The grid constraints include rated voltage and frequency range;

[0073] It should be specifically pointed out that the photovoltaic output model considers the total area of the photovoltaic panel, the standard working condition conversion efficiency, the temperature coefficient, and the shading coefficient parameter, wherein the shading coefficient is related to the shading area ratio; the wind power output model is calculated according to the wind speed size, and a pitch angle correction term is introduced to obtain the actual output;

[0074] The photovoltaic output model is as follows:

[0075] Pp=I×S×η0×[1-α×(Tp-25)]×kshade;

[0076] Wherein S is the total area of the photovoltaic panel (unit m²), which is determined by the photovoltaic array design parameters, η0 is the conversion efficiency under the standard working condition (light intensity 1000 W / m², temperature 25℃), α is the temperature coefficient, and the value is 0.004 / ℃, kshade is the shading coefficient, which is obtained by fitting the historical shading data: kshade=0.98-0.02×D, D is the shading area ratio (unit %);

[0077] The wind power output model is as follows:

[0078] When v

[0079] When vci≤v≤vN, Pw=PN×(v³-vci³) / (vN³-vci³);

[0080] When vN

[0081] Wherein vci is the cut-in wind speed (unit m / s), the value is 3 m / s, vN is the rated wind speed (unit m / s), the value is 12 m / s, vco is the cut-out wind speed (unit m / s), the value is 25 m / s, PN is the rated power of the fan (unit kW), and a pitch angle correction term is introduced: Pw'=Pw×(1-0.01×β), to obtain the corrected actual output.

[0082] The optimization objective function comprehensively considers the ratio of the actual output of photovoltaic power generation and wind power generation to the theoretical maximum output, the output fluctuation value, and the device temperature deviation, and the weight coefficient is dynamically adjusted according to the grid load;

[0083] The optimization objective function is as follows:

[0084] minJ=λ1×(1-Pactual / Pmax)+λ2×ΔP+λ3×(ΔT / Tmax);

[0085] Wherein Pactual is the current actual output of wind power generation or photovoltaic power generation (unit: kW), Pmax is the theoretical maximum output calculated by the wind power output model or the photovoltaic output model (unit: kW), ΔP is the output fluctuation value, ΔP = |Pactual-Pavg|, Pavg is the average output in the previous 10 minutes (unit: kW), ΔT is the equipment temperature deviation, the photovoltaic system ΔT = Tp-45℃, the wind power system ΔT = Tg-80℃ (Tg is the generator temperature), Tmax is the maximum allowable temperature deviation of the equipment, the photovoltaic system Tmax = 40℃, the wind power system Tmax = 20℃, λ1, λ2, λ3 are weight coefficients, satisfying λ1+λ2+λ3=1, wherein:

[0086] When the grid load L≥80% rated load, λ1=0.6, λ2=0.3, λ3=0.1;

[0087] When the grid load 50%≤L<80% rated load, λ1=0.4, λ2=0.4, λ3=0.2;

[0088] When the grid load L<50% rated load, λ1=0.2, λ2=0.5, λ3=0.3;

[0089] The equipment safety constraints are as follows:

[0090] Photovoltaic panel: Tp≤85℃, Up≤1.1×UpN, wherein UpN is the rated voltage, Tp is the working temperature of the photovoltaic panel, and Up is the output voltage of the photovoltaic panel;

[0091] Fan: n≤1800r / min, β≤30°, wherein n is the fan speed, too high speed may cause mechanical parts to wear out and even cause safety accidents, and β is the fan pitch angle, which is used to adjust the efficiency and stability of the fan absorbing wind energy;

[0092] Energy storage battery: SOC≥20%, charge and discharge power≤0.5×PbN, wherein PbN is the rated power of the battery, and SOC is the state of charge of the battery, to avoid excessive discharge of the battery affecting the service life;

[0093] The grid constraints are as follows:

[0094] Ug∈[0.95×UN,1.05×UN], wherein UN is the rated voltage, and Ug is the grid voltage, to ensure that the grid voltage is stable in a reasonable range and avoid damage to electrical equipment or grid collapse;

[0095] f∈[49.5Hz,50.5Hz], wherein f is the grid frequency, and frequency stability is a key indicator to ensure the safe and stable operation of the power system, and abnormal frequency will affect the normal work of power equipment.

[0096] S4: Control decision generation and execution: generate photovoltaic control instructions, wind power control instructions and energy storage control instructions based on the results of S3 model analysis combined with constraint conditions and execute them;

[0097] It is particularly pointed out that the photovoltaic control instructions include:

[0098] Light chasing angle adjustment: calculate the real-time light chasing angle θt of the photovoltaic panel by the formula θt=arcsin(Idirect / Itotal), where Idirect represents the direct light intensity (unit: W / m2) and Itotal is the total light intensity. This angle adjustment can make the photovoltaic panel always receive light in the best posture, improving the power generation efficiency. For example, when the sunlight is oblique, by calculating and adjusting the angle dynamically, the light energy absorption per unit area can be effectively increased.

[0099] Inverter MPPT parameter: determine the voltage reference value Uref according to the formula Uref=UpN×(Popt / Pmax), where Popt is the optimal output of the photovoltaic system obtained by the optimization model. This parameter setting can make the inverter track the maximum power point in real time, ensuring that the photovoltaic system can stably output maximum power under different light and temperature conditions;

[0100] The wind power control instructions include:

[0101] Pitch angle adjustment: calculate the pitch angle β using the formula β=0.5×(v-vN), but this adjustment strategy is only enabled when the wind speed v>vN (rated wind speed). By adjusting the pitch angle, the wind turbine blade angle can be controlled to avoid damage to the wind turbine due to overload in strong wind environment, while reasonably utilizing wind energy to achieve efficient power generation;

[0102] Fan start-stop control: when the wind speed v<vci and lasts for 10 minutes, it means that the current wind energy is not enough to maintain effective power generation of the wind turbine, so the stop command is issued; when the wind speed v>vco, the high wind speed may cause serious damage to the wind turbine structure, so the equipment is immediately stopped to protect it;

[0103] The energy storage control instructions include:

[0104] Charging control: when the actual power generation is greater than the optimal output calculated by the optimization model, and the state of charge SOC of the energy storage system is less than 80%, start the charging operation, the charging power Pch=Pactual-Popt. This strategy can store excess energy to avoid the phenomenon of abandoned electricity and improve energy utilization.

[0105] Discharge control: when Pactual<Popt and SOC≥20%, the energy storage system performs discharge operation, the discharge power Pdis=Popt-Pactual. In this way, when the power generation is insufficient, the power supply is supplemented to maintain the power balance of the power grid.

[0106] S5: Feedback correction: calculate the control deviation value, when the control deviation value is greater than the preset threshold value, correct the weight coefficient in the optimization model;

[0107] It should be specifically pointed out that the calculation method of the control deviation value is as follows:

[0108] e=|Pactual-Popt|, when e>5%, the weight coefficient in the optimization model is corrected: λ2=λ2+0.05, λ1=λ1-0.05, and λ1≥0 is ensured.

[0109] Specifically, taking a 100MW photovoltaic power station as an example, the implementation steps of the application are as follows:

[0110] One light sensor and one temperature sensor are arranged every 500m² in the photovoltaic array area, voltage and current sensors are installed at the bus box, I, T, Tp, Up and Ip data are collected in real time, Ug, f and L data are obtained through the power grid scheduling interface, and the sampling frequency is 1Hz;

[0111] The FPGA chip pre-processes the collected data, eliminates abnormal values with light intensity>2000W / m², interpolates the SOC data to 1Hz, and normalizes all parameters to the interval [0,1];

[0112] The industrial control computer runs the photovoltaic output model, wherein S=500000m², η0=18%, α=0.004 / ℃, kshade is calculated according to the real-time image recognition D value; the optimization objective function is set according to the power grid load L=90% rated load, λ1=0.6, λ2=0.3, λ3=0.1, and Popt=95MW is obtained by solving;

[0113] The PLC controller adjusts the light tracking angle θtrack=42°, and the inverter voltage reference value Uref=800V; when the actual output Pactual=92MW, the energy storage system discharges 3MW, and ensures Pactual=Popt;

[0114] The e value is continuously monitored, when e=2%<5%, the weight coefficient remains unchanged; when e=6% due to cloud cover, λ2 is adjusted to 0.35, and λ1 is adjusted to 0.55, until e≤5%.

[0115] As shown in a new energy power generation intelligent control system based on big data, comprising: Figure 2

[0116] ​Data acquisition module: composed of light sensor (measurement range 0-2000W / m2, accuracy ±5%), temperature sensor (measurement range-40℃-125℃, accuracy ±0.5℃), wind speed sensor (measurement range 0-50m / s, accuracy ±0.3m / s), voltage and current sensor (measurement range 0-1000V / 0-500A, accuracy ±0.2%), and smart meter, connected with data processing module through RS485 bus;

[0117] Data processing module: using FPGA chip, for realizing data outlier rejection, time and space alignment and normalization processing, processing delay ≤10ms;

[0118] Mathematical modeling analysis module: for deploying multi-parameter coupling output model and optimization objective function based on industrial control computer (CPU is Intel Xeon E5-2690, memory 32GB), calculation period 1 second;

[0119] Control execution module: including PLC controller, servo driver, relay, for receiving analysis module instructions, driving photovoltaic light tracking motor (positioning accuracy ±0.5°), fan variable pitch mechanism (response time ≤500ms), energy storage inverter (conversion efficiency ≥96%);

[0120] Data storage module: using HBase distributed database, storage capacity ≥10TB, for supporting historical data query and model parameter update.

[0121] Secondly: the drawings in the disclosed embodiments of the application only involve the structures involved in the disclosed embodiments of the application, other structures can refer to the usual design, and in the case of no conflict, the same embodiment and different embodiments of the application can be combined with each other;

[0122] Finally: the above only describes the preferred embodiments of the application, and is not used to limit the application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application should be included in the protection scope of the application.

Claims

1. A smart control method for new energy power generation based on big data, characterized in that, It includes: S1: Multi-dimensional data collection: Collect new energy power generation data through an Internet of Things sensor array, smart meters, and grid dispatching interfaces. S2: Data preprocessing: Perform outlier removal, spatio-temporal alignment, and normalization on the collected new energy power generation data. S3: Model analysis: Based on the data processed in S2, construct a multi-parameter coupled output model and an optimization objective function, and at the same time set equipment safety constraints and grid constraints as constraint conditions. The multi-parameter coupled output model includes a photovoltaic output model and a wind power output model. The photovoltaic output model is constructed with core parameters such as light intensity, total photovoltaic panel area, standard condition conversion efficiency, temperature coefficient, and shading coefficient. The wind power output model is constructed with core parameters such as wind speed, cut-in wind speed, rated wind speed, cut-out wind speed, rated power of the wind turbine, and pitch angle. The equipment safety constraints include parameter limitations related to photovoltaic panels, wind turbines, and energy storage batteries. The grid constraints include the rated voltage and frequency range. S4: Control decision generation and execution: Generate photovoltaic control instructions, wind power control instructions, and energy storage control instructions based on the results of S3 model analysis combined with constraint conditions and execute them. S5: Feedback correction: Calculate the control deviation value. When the control deviation value is greater than the preset threshold, correct the weight coefficient in the optimization model.

2. The intelligent control method for new energy power generation based on big data according to claim 1, characterized in that: The new energy power generation data includes: Environmental parameters: including light intensity and ambient temperature in the photovoltaic area, wind speed and wind direction in the wind power area. Equipment parameters: including the working temperature, output voltage, and output current of the photovoltaic panel; the rotation speed, pitch angle, and output power of the wind turbine, and the state of charge of the energy storage battery. Grid parameters: including the grid connection point voltage, frequency, and real-time load. Historical data: including hourly power generation output, environmental parameters, and equipment operation parameters in the past 3 years, stored in a distributed database.

3. The intelligent control method for new energy power generation based on big data according to claim 1, characterized in that: The photovoltaic output model is as follows: Pp = I × S × η0 × [1 - α × (Tp - 25)] × kshade, where S is the total photovoltaic panel area, determined by the photovoltaic array design parameters, η0 is the conversion efficiency under standard conditions, α is the temperature coefficient, with a value of 0.004 / °C, kshade is the shading coefficient, obtained by fitting historical shading data: kshade = 0.98 - 0.02 × D, D is the proportion of the shaded area, Pp is the photovoltaic output, I is the light intensity in the photovoltaic area, and Tp is the working temperature of the photovoltaic panel.

4. The intelligent control method for new energy power generation based on big data according to claim 1, characterized in that: The wind power output model is as follows: When v < vci or v > vco, Pw = 0; When vci ≤ v ≤ vN, Pw = PN × (v³ - vci³) / (vN³ - vci³); When vN < v ≤ vco, Pw = PN; Where vci is the cut-in wind speed, with a value of 3 m / s, vN is the rated wind speed, with a value of 12 m / s, vco is the cut-out wind speed, with a value of 25 m / s, PN is the rated power of the wind turbine, introducing a pitch angle correction term: Pw' = Pw × (1 - 0.01 × β), to obtain the corrected actual output. v is the wind speed in the wind power area, Pw is the theoretical wind power output, Pw' is the corrected actual wind power output, and β is the pitch angle of the wind turbine.

5. The intelligent control method for new energy power generation based on big data according to claim 1, characterized in that: The optimization objective function is as follows: minJ = λ1×(1 - Pactual / Pmax) + λ2×ΔP + λ3×(ΔT / Tmax), where Pactual is the current actual output of wind power generation or photovoltaic power generation, Pmax is the theoretical maximum output calculated by the wind power output model or photovoltaic power output model, ΔP is the output fluctuation value, ΔP = |Pactual - Pavg|, Pavg is the average output in the previous 10 minutes, ΔT is the equipment temperature deviation, for the photovoltaic system, ΔT = Tp - 45°C, for the wind power system, ΔT = Tg - 80°C, Tmax is the maximum allowable temperature deviation of the equipment, for the photovoltaic system, Tmax = 40°C, for the wind power system, Tmax = 20°C, and λ1, λ2, λ3 are weight coefficients satisfying λ1 + λ2 + λ3 = 1, where: When the grid load L ≥ 80% of the rated load, λ1 = 0.6, λ2 = 0.3, λ3 = 0.1; When the grid load 50% ≤ L < 80% of the rated load, λ1 = 0.4, λ2 = 0.4, λ3 = 0.2; When the grid load L < 50% of the rated load, λ1 = 0.2, λ2 = 0.5, λ3 = 0.3; Where, Tg is the generator temperature.

6. The intelligent control method for new energy power generation based on big data according to claim 1, characterized in that: The equipment safety constraints are as follows: Photovoltaic panel: Tp ≤ 85°C, Up ≤ 1.1×UpN, where UpN is the rated voltage, Tp is the working temperature of the photovoltaic panel, and Up is the output voltage of the photovoltaic panel; Wind turbine: n ≤ 1800 r / min, β ≤ 30°, where n is the rotational speed of the wind turbine, too high rotational speed will lead to increased wear of mechanical components and even cause safety accidents, and β is the pitch angle of the wind turbine, which is used to adjust the efficiency and stability of the wind turbine in absorbing wind energy; Energy storage battery: SOC ≥ 20%, charge-discharge power ≤ 0.5×PbN, where PbN is the rated power of the battery and SOC is the state of charge of the battery; The grid constraints are as follows: Ug ∈ [0.95×UN, 1.05×UN], where UN is the rated voltage and Ug is the grid voltage; f ∈ [49.5 Hz, 50.5 Hz], where f is the grid frequency.

7. The intelligent control method for new energy power generation based on big data according to claim 1, characterized in that: The photovoltaic control instructions include: Tracking angle adjustment: Calculate the real-time tracking angle θt of the photovoltaic panel through the formula θt = arcsin(Idirect / Itotal), where Idirect represents the direct sunlight intensity and Itotal is the total sunlight intensity; Inverter MPPT parameter: Determine the voltage reference value Uref according to the formula Uref = UpN×(Popt / Pmax), where Popt is the optimal output of the photovoltaic system obtained by solving the optimization model; The wind power control instructions include: Pitch angle adjustment: Calculate the pitch angle β using the formula β = 0.5×(v - vN), but this adjustment strategy is only enabled when the wind speed v > vN; Wind turbine start-stop control: When the wind speed v < vci and lasts for 10 minutes, a shutdown instruction is issued at this time. When the wind speed v > vco, the equipment is immediately shut down for protection; The energy storage control instructions include: Charging control: When the actual power generation is greater than the optimal output calculated by the optimization model and the state of charge (SOC) of the energy storage system is less than 80%, the charging operation is started, and the charging power Pch = Pactual - Popt; Discharging control: When Pactual < Popt and SOC ≥ 20%, the energy storage system performs a discharging operation, and the discharging power Pdis = Popt - Pactual; Where UpN is the rated voltage of the photovoltaic panel, Pmax is the theoretical maximum output, vN is the rated wind speed, vci is the cut-in wind speed, vco is the cut-out wind speed, Pactual is the current actual output, and Popt is the optimal output solved by the optimization model.

8. The intelligent control method for new energy power generation based on big data according to claim 1, characterized in that: The calculation method of the control deviation value is as follows: e = |Pactual - Popt|. When e > 5%, the weight coefficients in the optimization model are corrected: λ2 = λ2 + 0.05, λ1 = λ1 - 0.05, ensuring that λ1 ≥ 0; Where e is the control deviation value, Pactual is the current actual output, Popt is the optimal output solved by the optimization model, and λ1, λ2 are the weight coefficients in the optimization objective function.

9. A big data-based intelligent control system for new energy power generation, used to implement the big data-based intelligent control method for new energy power generation as described in any one of claims 1-8, characterized in that, It includes: Data acquisition module: Composed of a light sensor, a temperature sensor, a wind speed sensor, a voltage and current sensor, and a smart meter, and is connected to the data processing module through the RS485 bus; Data processing module: Adopts an FPGA chip and is used to implement the elimination of data outliers, spatio-temporal alignment, and normalization processing; Mathematical modeling and analysis module: Used to deploy a multi-parameter coupled output model and an optimization objective function based on an industrial control computer; Control execution module: Includes a PLC controller, a servo driver, and a relay, and is used to receive instructions from the analysis module and drive the photovoltaic light-tracking motor, the fan pitch mechanism, and the energy storage inverter; Data storage module: Adopts an HBase distributed database and is used to support historical data query and model parameter update.

Images