A model predictive control method for photovoltaic and photo-thermal integrated greenhouse

Abstract

The application provides a model predictive control system for a photovoltaic and light heat integrated greenhouse, and belongs to the technical field of intelligent control of agricultural greenhouse environment temperature. The system comprises the following steps: a concentrated photovoltaic and light heat integrated device converts solar energy into electric energy and heat energy; a storage battery stores the photovoltaic converted electric energy and supplies power to the greenhouse electric equipment when the photovoltaic power generation is insufficient; when the photovoltaic converted electric energy is excessive or insufficient, power interaction is performed through an external power grid; heat energy is supplied to the greenhouse through a heat storage water tank, a heating circulating water pipeline and a heat sink; a data acquisition module acquires greenhouse environment parameters and operation data of each actuator of the concentrated photovoltaic and light heat integrated module in real time and transmits the data to an intelligent decision module; the intelligent decision module receives data of the data acquisition module, outputs a greenhouse air temperature control instruction and a power dispatching strategy, and sends the instruction and the strategy to corresponding actuators of the concentrated photovoltaic and light heat integrated module. The system effectively integrates photovoltaic power generation, light heat recovery and microgrid dispatching, and significantly improves energy utilization efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent temperature control in agricultural greenhouses and energy optimization scheduling of photovoltaic microgrids, and in particular to a model predictive control method for photovoltaic-thermal integrated greenhouses. Background Technology

[0002] Currently, traditional greenhouses generally rely on fossil fuels such as coal-fired boilers for heating, supplemented by manual experience in ventilation and curtain operation. This results in large temperature fluctuations inside the greenhouse during winter and persistently high energy consumption per unit area. In terms of power management, traditional greenhouses lack coordinated scheduling of self-generated photovoltaic power, grid power purchases, and energy storage systems, making it impossible to effectively optimize energy costs during peak and off-peak electricity periods. This "extensive" temperature control and energy management model not only increases carbon emission intensity but also fails to meet the core demands of facility agriculture for precise temperature control and low-cost power supply. New energy smart greenhouses, by integrating photovoltaic and solar thermal integrated equipment with intelligent control systems, represent a key path to solving these problems.

[0003] However, existing technologies still have significant shortcomings in the coupling and regulation of temperature and electrical energy:

[0004] 1. Insufficient adaptability to nonlinear systems: For example, fuzzy control algorithms rely on manually set empirical rule bases. When dealing with complex characteristics such as the nonlinear change of heat collection efficiency of concentrated photovoltaic and solar thermal modules with irradiance and the thermal inertia of hot water storage tanks, control delays and strategy rigidity are likely to occur. Especially during the critical period of nighttime energy storage and grid power complementarity, it is difficult to achieve a balance between economy and stability.

[0005] 2. Limitations of single-variable control strategies: For example, while relying solely on PID algorithms can achieve single-loop temperature control, it cannot incorporate photovoltaic forecasts, energy consumption data, etc., into a unified framework for analysis. It is also impossible to dynamically optimize the charging and discharging power of energy storage and the power exchange between the grid and the grid, making it difficult to achieve the economy and efficiency of multi-energy coordinated operation while ensuring the stability of the greenhouse environment.

[0006] Therefore, a model-based predictive control method is needed for photovoltaic-thermal integrated greenhouses. Summary of the Invention

[0007] In view of this, the present invention provides a model predictive control system for photovoltaic-thermal integrated greenhouses, which effectively integrates photovoltaic power generation, solar thermal heat recovery and microgrid scheduling, and achieves the triple optimization goals of improving energy utilization, reducing operating costs and improving the temperature stability of the greenhouse environment.

[0008] Therefore, the present invention provides the following technical solution:

[0009] A model predictive control system for photovoltaic-thermal integrated greenhouses includes:

[0010] Concentrated photovoltaic and solar thermal integrated module, data acquisition module, and intelligent decision-making module;

[0011] The concentrated photovoltaic and solar thermal integrated module includes: a concentrated photovoltaic and solar thermal integrated device, a hot water storage tank, a temperature regulation device, a battery, and a grid interaction unit;

[0012] The concentrated photovoltaic-thermal integrated equipment converts solar energy into electrical energy and thermal energy;

[0013] The storage battery stores the electrical energy generated by photovoltaic conversion and supplies power to the greenhouse equipment when photovoltaic power generation is insufficient; the grid interaction unit is used to interact with the external power grid when there is a surplus or shortage of electrical energy.

[0014] Heat energy is supplied to the greenhouse through a hot water storage tank; a temperature regulation device assists the concentrating photovoltaic-thermal integrated equipment to achieve temperature control inside the greenhouse;

[0015] The data acquisition module collects greenhouse environmental parameters and operational data from each actuator of the integrated photovoltaic and solar thermal module in real time, and transmits them to the intelligent decision-making module.

[0016] The intelligent decision-making module outputs greenhouse air temperature control commands and power dispatch strategies based on the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic and solar thermal integrated module, and sends them to the corresponding actuators of the concentrated photovoltaic and solar thermal integrated module.

[0017] Furthermore, the integrated photovoltaic and solar thermal concentrator includes:

[0018] Photovoltaic panels, interference filters, vacuum collector tubes, and auxiliary supports;

[0019] The photovoltaic panel is parabolic in shape, with the open end of the parabola facing the direction of solar incidence;

[0020] The interference filter film is applied to the light-facing surface of the photovoltaic panel, selectively transmitting and reflecting the full spectrum of light irradiated onto the surface of the interference filter film.

[0021] The vacuum heat collection tubes are arranged parallel to the focal line of the parabolic surface. The light reflected by the interference filter film is focused onto the surface of the vacuum heat collection tubes to collect photothermal energy. The vacuum heat collection tubes are frustum bodies with several water inlets on the bottom surface and one water outlet on the top surface. The heating circulating water flows in from the water inlets and flows out from the water outlet.

[0022] The auxiliary support is installed on the non-opening side of the photovoltaic panel to support the photovoltaic panel.

[0023] Furthermore, it also includes a visualization monitoring module;

[0024] The visualization monitoring module presents photovoltaic power generation forecast data and greenhouse environment data in a visual format.

[0025] A model predictive control method for photovoltaic-thermal integrated greenhouses, based on a model predictive control system for such greenhouses, includes:

[0026] Collect greenhouse environmental parameters and operational data from each actuator of the concentrated photovoltaic-thermal integrated module;

[0027] Based on the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic and solar thermal integrated module, the PID controller outputs greenhouse air temperature control commands.

[0028] Based on solar irradiance and greenhouse air temperature, a machine learning model is used to output predicted values ​​for photovoltaic power generation.

[0029] The model predictive control algorithm uses the photovoltaic power generation prediction value and the heating energy consumption value of the temperature regulation device to output an energy dispatch strategy.

[0030] The corresponding actuators are controlled by commands for controlling greenhouse air temperature and energy dispatch strategies.

[0031] Furthermore, based on the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic-thermal integrated module, the PID controller outputs greenhouse air temperature control commands, including:

[0032] The Kalman filter algorithm is used to reduce noise in the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic and solar thermal integrated module to obtain noise-reduced data.

[0033] The state equation and observation equation are constructed based on noise-reduced data to characterize the photothermal output of the concentrated photovoltaic-thermal integrated equipment;

[0034] The solar thermal output is compared with the preset temperature using PID calculation to determine whether the solar thermal output meets the heat load requirements. If the requirements are met, the mixing ratio parameters and flow rate control parameters of the hot water storage tank are output to regulate the greenhouse air temperature. If the requirements are not met, the heat output power of the temperature regulation device is calculated, and the mixing ratio parameters and flow rate control parameters of the hot water storage tank are output to maintain the greenhouse air temperature within the target range.

[0035] Furthermore, the machine learning model is a long short-term memory network.

[0036] Furthermore, the step of using the photovoltaic power generation prediction value and the heating energy consumption value of the temperature regulation device to output an energy dispatch strategy through the model predictive control algorithm includes:

[0037] Using the predicted value of photovoltaic power generation and the heating energy consumption value of the temperature regulation device as inputs, and the operating cost ratio, photovoltaic absorption rate, grid power dependence and air source heat pump energy consumption ratio as optimization objectives, a model predictive control framework is established; and an energy dispatch strategy is generated through rolling time domain optimization, while constraints are applied to ensure the safety boundary of the system.

[0038] Furthermore, the optimization objective function is:

[0039]

[0040] Among them, C cost For operating costs, R pv For photovoltaic power absorption rate, D grid For mains power dependence, C max E represents the maximum operating cost. max This represents the total energy consumption of an air source heat pump operating continuously at maximum power.

[0041] Furthermore, the mixing ratio parameters and flow rate control parameters of the hot water storage tank include:

[0042]

[0043] Where: u(t) is the control quantity generated by the PID algorithm; γ(t) is the mixing ratio, γ min γ max These are the minimum and maximum mixing ratios, respectively; u low u high These are the lower and upper threshold values ​​for the PID output, respectively; k γ V is the proportionality coefficient; v(t) is the circulation velocity of the solar thermal heating system. min v max These are the minimum and maximum flow velocities, respectively; k v is the curve steepness coefficient, used to control response sensitivity; u0 is the center threshold of the PID output.

[0044] Furthermore, the formula for calculating the photothermal output power is as follows:

[0045] Q PV / T =G solar ·A eff ·η th ·γ(t)·(1-e -βv(t) )

[0046] The formula for calculating the heat output power of the temperature regulating device is as follows:

[0047]

[0048] Where, η th β is the photothermal conversion efficiency coefficient, β is the flow rate attenuation coefficient, and A is the photothermal conversion efficiency coefficient.eff The effective heat collection area of ​​the photovoltaic thermal receiver; COP is the coefficient of performance of the air source heat pump, η′ is the efficiency factor of the air source heat pump; Q load This indicates the heat load demand.

[0049] Advantages and positive effects of the present invention:

[0050] This system collects environmental parameters and operational data from the integrated photovoltaic and solar thermal power generation equipment in real time through a data acquisition module, and employs Kalman filtering to eliminate noise interference. The collected data is input into a proportional-integral-derivative (PID) controller to calculate the solar thermal output and dynamically optimize the flow rate and mixing ratio of the heat dissipation pipes. By integrating heat dissipation pipes and radiators, heat energy is directionally delivered into the greenhouse. Combined with a hot water storage tank, heat energy is stored and released on demand, ensuring a stable indoor temperature. A photovoltaic power generation prediction model is constructed based on historical data, and the predicted data is input into Model Predictive Control (MPC) to adjust the charging and discharging of the energy storage and the interaction with the mains power, balancing energy supply and demand, reducing operating costs, and ensuring the stability of greenhouse heating. This system effectively integrates photovoltaic power generation, solar thermal recovery, and microgrid scheduling, significantly improving energy utilization efficiency and providing a highly efficient, low-carbon, intelligent control solution for modern agricultural greenhouses.

[0051] In this integrated photovoltaic and solar thermal system, the vacuum collector tubes employ a truncated cone-shaped, gradually decreasing diameter design. By optimizing the pipe geometry, a larger cross-section is formed at the water inlet to reduce fluid resistance and ensure sufficient water flow. Simultaneously, the gradually decreasing pipe diameter during water convergence allows the flow velocity to adapt to the distribution of solar and thermal energy, significantly improving heat exchange efficiency. This structure buffers the impact of inlet water pressure, optimizes system pressure balance, and reduces the risk of localized stress concentration. The changing pipe diameter enables cascaded energy utilization, promoting uniform axial transfer of solar and thermal energy and improving the uniformity and stability of heat collection. Furthermore, its gradually decreasing profile effectively disperses thermal stress, enhances the structural strength and durability of the tube, strengthens fatigue and thermal shock resistance, and ensures long-term, efficient system operation. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a structural diagram of the model predictive control system for a photovoltaic-thermal integrated greenhouse in Example 1.

[0054] Figure 2 This is a schematic diagram of the integrated photovoltaic and solar thermal equipment in Example 1;

[0055] Among them, 1. Photovoltaic panel; 2. Interference filter film; 3. Vacuum collector tube; 31. Water inlet; 32. Water outlet; 4. Auxiliary support.

[0056] Figure 3 This is a flowchart of the model prediction and control method for photovoltaic-thermal integrated greenhouses in Example 2. Detailed Implementation

[0057] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0058] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0059] Example 1

[0060] Combination Figure 1 As shown, the present invention provides a model prediction and control system for photovoltaic-thermal integrated greenhouses, including: a concentrating photovoltaic-thermal integrated module, a data acquisition module, an intelligent decision-making module, and a visualization monitoring module.

[0061] The concentrated photovoltaic and solar thermal integrated module includes: a concentrated photovoltaic and solar thermal integrated equipment, a hot water storage tank, a temperature control device, a battery, and a grid interaction unit;

[0062] Concentrated photovoltaic (PV) and solar thermal integrated systems convert solar energy into electricity and heat. When the electricity generated by the PV system cannot meet the greenhouse's load demand, it prioritizes using the electricity stored in batteries. The system dynamically adjusts its power interaction with the external power grid through an intelligent decision-making module: if an energy gap still exists, it obtains power from the grid through a grid interaction unit to ensure stable equipment operation; when PV power generation is excessive, the surplus electricity is stored in batteries or fed back to the grid, optimizing energy dispatch strategies and reducing operating costs. The heat energy converted by the PV system is transported to the greenhouse as hot water through an integrated heat dissipation pipe system via a hot water storage tank. This hot water storage tank enables heat storage and on-demand release, maintaining a stable temperature inside the greenhouse and improving overall energy utilization.

[0063] The data acquisition module includes: greenhouse room temperature sensor, water temperature sensor, solar radiation sensor, and vacuum collector tube sensor. Each sensor transmits the collected greenhouse environmental parameters and equipment operation data to the intelligent decision-making module in real time and stably via a ZigBee network. The intelligent decision-making module receives the data from the acquisition module and analyzes it based on a model predictive control method for photovoltaic-thermal integrated greenhouses. It generates precise control commands for the mixing valve and flow valve to ensure the greenhouse air temperature is maintained. When solar thermal energy is insufficient, it activates the air source heat pump to provide additional heating. Simultaneously, the intelligent decision-making module outputs optimized energy storage charging and discharging power and mains power interaction power to balance energy supply and demand, reduce operating costs, and ensure the stability of greenhouse heating.

[0064] The visualization monitoring module presents photovoltaic power generation forecast data and greenhouse environmental data in a visual format. As a human-computer interaction hub, the visualization monitoring module receives greenhouse environmental parameters and equipment operation data from the intelligent decision-making module, and intuitively displays energy supply and demand and equipment operating status through charts, curves, and real-time dashboards. It also provides visual support for anomaly warnings and strategy optimization.

[0065] Combination Figure 2 As shown, the concentrated photovoltaic and solar thermal integrated equipment includes:

[0066] 1. Photovoltaic panel; 2. Interference filter film; 3. Vacuum heat collection tube; and 4. Auxiliary support.

[0067] The photovoltaic panel 1 is parabolic in shape, with its open end facing the direction of solar incidence to collect and concentrate solar energy. An interference filter 2 covers the light-facing surface of the photovoltaic panel 1, forming a parabolic shape along with the panel. It selectively transmits and reflects the full spectrum of light irradiated onto the surface: selectively filtering the photovoltaic-sensitive wavelength band (380–1100 nm) so that it directly irradiates the photovoltaic power generation panel below, where the panel efficiently converts this wavelength of light energy into electrical energy; and reflecting the ultraviolet (<380 nm) and infrared (>1100 nm) spectra. The photothermal energy reflected by the interference filter 2 is directly focused by its parabolic shape onto the surface of the vacuum heat collection tube located above, achieving directional collection of photothermal energy. This avoids overheating losses caused by the photovoltaic panel absorbing non-sensitive wavelengths of light energy and improves the efficiency of concentrated utilization of photothermal energy. The vacuum heat collection tube 3 is arranged parallel to the focal line of the parabolic surface, employing a truncated cone-shaped tapered tube design, and is equipped with four inlets 31 and one outlet 32. Heating circulating water flows in through four inlets 31, flows along the frustum-shaped pipe, and changes in flow velocity as the pipe diameter changes, finally flowing out through one outlet 32. This efficiently absorbs the photothermal energy reflected and concentrated by the interference filter film 2, completing the collection and transfer of heat energy. An auxiliary support 4 is installed on the open side of the photovoltaic panel to support it.

[0068] The intelligent decision-making module receives real-time data from the data acquisition module and uses a Kalman filter algorithm to denoise the data, filtering out noise interference to obtain denoised data. A PID controller is constructed to calculate the solar thermal output based on the denoised data, determining whether the solar thermal output meets the greenhouse air temperature requirements. If the solar thermal output is sufficient, optimized mixing ratio and flow rate control parameters are output to precisely adjust the solar thermal output. If the solar thermal output is insufficient, the heat consumption of the air source heat pump is calculated, and the solar thermal output and auxiliary heating of the air source heat pump are coordinated and controlled to ensure that the greenhouse air temperature environment remains stable within the target range, achieving precise control of the solar thermal and auxiliary heating equipment. A photovoltaic power generation prediction model is constructed based on historical data, and real-time data is imported every 15 minutes for rolling updates, dynamically tracking the trend of photovoltaic power generation changes and effectively addressing the intermittency and uncertainty of photovoltaic energy. Finally, the predicted photovoltaic power generation value and air source heat pump energy consumption data are integrated into the model predictive control algorithm denoised data framework, outputting optimized energy storage charging and discharging power and grid power interaction power to achieve efficient energy dispatch.

[0069] The advantage of this integrated photovoltaic and solar thermal system lies in the deep synergy between enhanced concentration and spectral optimization. Traditional PV / T equipment, limited by its flat-plate structure, can only utilize natural irradiance and suffers from low spectral efficiency. In contrast, the SCAPV / T (Integrated Photovoltaic and Solar Thermal System) utilizes a parabolic trough-shaped concentrating design, integrating an interference filter film 2 onto the surface of the photovoltaic panel 1 to form an integrated structure, increasing the incident light energy density by 3-5 times. Simultaneously, the interference filter film 2 intelligently divides the entire spectrum: selectively transmitting the 380-1100nm photovoltaic-sensitive wavelength band to the underlying photovoltaic cell layer, fully covering the main response range of monocrystalline silicon cells, thus improving photoelectric conversion efficiency compared to traditional photovoltaic modules; reflecting the <380nm ultraviolet and >1100nm infrared spectra and focusing them through the parabolic structure, converging them to the solar thermal collection unit with 3-5 times the light energy density, achieving comprehensive photovoltaic and solar thermal utilization. The improved concentrated photovoltaic and solar thermal integrated equipment has a significantly higher overall energy efficiency than traditional PV / T equipment, breaking through the dual bottlenecks of light energy density and spectral utilization of traditional equipment.

[0070] Example 2

[0071] A model predictive control method for photovoltaic-thermal integrated greenhouses, based on a model predictive control system for such greenhouses, includes the following steps:

[0072] S1. Multi-source data acquisition and dynamic preprocessing.

[0073] The system collects greenhouse environmental parameters and operating data of the integrated photovoltaic and solar thermal equipment in real time through a sensor network. It then uses a Kalman filter algorithm to reduce noise interference, obtains noise-reduced data, and stores it in a database to provide high-quality input for subsequent modeling.

[0074] S11. Through the deployed sensor network, real-time data collection of greenhouse environmental parameters and operating data of each actuator of the concentrated photovoltaic and solar thermal integrated module is used as raw data.

[0075] In this embodiment, the sensor network includes a room temperature sensor, a water temperature sensor, a solar irradiance sensor, and a vacuum collector tube temperature sensor.

[0076] In this embodiment, the greenhouse environmental parameters include: greenhouse air temperature, indoor light intensity, and hot water storage tank temperature. The operating data of each actuator in the concentrated photovoltaic-thermal integrated module include: photovoltaic power generation data and solar thermal output power of the concentrated photovoltaic-thermal integrated equipment.

[0077] S2. Based on the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic and solar thermal integrated module, the PID controller outputs the greenhouse air temperature control command.

[0078] S21. The Kalman filter algorithm is used to denoise the original data, filter out random noise interference, obtain denoised data, and improve the convergence of subsequent models.

[0079] S22. Construct the data state equation and observation equation, using the following formula:

[0080]

[0081] Among them, T air (t) represents the greenhouse air temperature at time t, I light (t) represents the light intensity inside the greenhouse at time t, where T is the light intensity. tank (t) represents the temperature of the hot water tank at time t; T air (t-1) represents the greenhouse air temperature at time t-1, I light (t-1) represents the light intensity inside the greenhouse at time t-1, where T tank (t-1) represents the temperature of the hot water storage tank at time t-1; Q heat Q represents the heating power of the circulating heating water to the environment. loss For heating heat loss power, G solar Q represents solar irradiance. PV / T Q′ represents the solar thermal output power of a concentrated photovoltaic (PV) solar thermal integrated device. loss The heat loss power of the water tank; α is the thermal inertia coefficient; η is the light transmittance attenuation coefficient of the greenhouse film; C water This is the specific heat capacity of water; This indicates the actual greenhouse air temperature detected by the temperature sensor. This indicates the actual light intensity detected by the light sensor. This indicates the actual temperature of the hot water storage tank detected by the water temperature sensor.

[0082] The state equation describes the state at the previous time step and takes into account external control inputs and process noise ω. t To dynamically estimate the true values ​​of environmental parameters; the observation equation is used to describe the relationship between sensor measurements and the true state, and observation noise v is introduced. t The predicted value is corrected using actual sensor measurements to reduce the impact of noise. The weights of the prediction and observation are dynamically adjusted using the covariance matrix (Q,R) to make the estimated value approximate the true value. Q represents the process noise ω. t Covariance matrix, R is the observation noise v t Covariance matrix. By appropriately setting Q and R, Kalman filtering can achieve optimal estimation in the complex environment of a greenhouse, providing accurate data support for the "solar-thermal-assisted heating coordinated control algorithm". At the same time, the processed data is stored in a historical database and used to construct a photovoltaic power generation prediction model, thereby improving the accuracy of system control and prediction in multiple ways.

[0083] S23. Define the error as the difference between the target temperature and the greenhouse air temperature:

[0084] e(t) = T set -T air (t)

[0085] Among them, T set T represents the target temperature. air (t) represents the greenhouse air temperature.

[0086] S24. Generate the control quantity u(t) using the PID algorithm:

[0087]

[0088] Where e(t) is the temperature error; K p K i and K d The controller parameters are determined through empirical tuning, and u(t) is the control quantity generated by the PID algorithm.

[0089] S25. Dynamically adjust the mixing ratio and flow rate:

[0090]

[0091] Where γ(t) is the mixing ratio, γ min γ is the minimum mixing ratio. max This represents the maximum mixing ratio; u low u is the lower limit threshold of the PID output. high k is the upper limit threshold for the PID output. γ V is the proportionality coefficient; v(t) is the circulation velocity of the solar thermal heating system. min v is the minimum circulating velocity for solar thermal heating. max The maximum circulating flow rate for solar thermal heating; u0 is the center threshold of the PID output; k v This is the curve steepness coefficient, used to control response sensitivity.

[0092] S26. Determine the solar thermal output power of the integrated photovoltaic and solar thermal equipment:

[0093] Q PV / T =G solar ·A eff ·η th ·γ(t)·(1-e -βv(t) )

[0094] Where, η th β is the photothermal conversion efficiency coefficient, β is the flow rate attenuation coefficient, and A is the photothermal conversion efficiency coefficient. eff This refers to the effective heat collection area of ​​the photovoltaic thermal receiver.

[0095] When the solar thermal output power cannot meet the heat load demand, calculate the power provided by the air source heat pump:

[0096]

[0097] Where COP is the coefficient of performance of the air source heat pump, and η′ is the efficiency factor of the air source heat pump; Q load This indicates the heat load demand.

[0098] S3. Based on solar irradiance and greenhouse air temperature, a machine learning model is used to output predicted values ​​for photovoltaic power generation.

[0099] Historical solar irradiance and greenhouse air temperature were extracted from the database as input features. The machine learning model architecture was designed as a two-layer LSTM network, with 64 neurons in the first layer and 32 neurons in the second layer. Dropout regularization was added between layers to suppress overfitting. The Adam optimizer was used to dynamically adjust the parameters, and the mean squared error loss function was selected. An early stopping mechanism was introduced to optimize the training process. After the input data was normalized and time-periodicly encoded, the model could output the photovoltaic power generation curve for the next 24 hours, significantly alleviating the challenges of the intermittency and volatility of photovoltaic energy.

[0100] To achieve dynamic updates to the forecast results, the system inputs the latest measured solar irradiance and ambient temperature data into the model every 15 minutes to regenerate the photovoltaic power generation forecast curve for the next 24 hours. This periodic refresh mechanism ensures that the forecast results are synchronized with real-time environmental changes, providing reliable support for subsequent energy optimization and scheduling.

[0101] S4. The model predictive control algorithm is used to output an energy dispatch strategy based on the photovoltaic power generation prediction value and the heating energy consumption value of the temperature regulation device.

[0102] Using the photovoltaic power generation forecast and air source heat pump energy consumption as inputs, an MPC framework is established. With the operating cost ratio, photovoltaic absorption rate, grid power dependence, and air source heat pump energy consumption ratio as optimization objectives, an energy dispatch strategy is generated through rolling time-domain optimization, and multi-dimensional constraints such as power balance, energy storage dynamics, and carbon emissions are applied to ensure the system's safety boundary.

[0103] S41. Construct an MPC framework with the objective function as follows:

[0104]

[0105] Among them, C cost Operating costs:

[0106]

[0107] R pv Photovoltaic grid integration rate:

[0108]

[0109] D grid Dependence on mains power:

[0110]

[0111] C max E represents the maximum operating cost. max This represents the total energy consumption of an air source heat pump operating continuously at maximum power.

[0112] S42. Construct a multi-objective rolling optimization and real-time scheduling model.

[0113] 1) Determine the surplus photovoltaic power sold as the state variable x(k) by using the battery state of charge, hot water tank temperature, real-time photovoltaic output, and grid power purchase:

[0114] x(k) = [SOC(k), T tank (k),P pv (k),P grid (k),P sell (k)] T

[0115] The control variable u(k) is constructed using battery charging power, battery discharging power, mains power purchase power, and photovoltaic surplus power sold:

[0116] u(k)=[P cahrge (k),P discharge (k),P grid (k),P sell (k)] T

[0117] Using load demand and solar irradiance as perturbation variables w(k):

[0118] w(k)=[P load G solar ] T

[0119] 2) Establish a state-space model:

[0120] x(k+1)=Ax(k)+Bu(k)+Ew(k)

[0121] in η loss Let be the energy loss coefficient of the energy storage system, h be the heat transfer coefficient, S be the heat dissipation area, and m be the mass of water.

[0122]

[0123] Matrix A represents the autocorrelation of state variables, matrix B reflects the direct influence of control variables on the state, and matrix E describes the coupling effect of disturbance variables.

[0124] S43. Perform a linear transformation on the multi-objective rolling optimization and real-time scheduling model, including:

[0125] 1) Construct the objective function:

[0126]

[0127] Where U is the sequence of control variables: N represents the prediction time-domain step number. In this embodiment, 24 hours is divided into 15-minute intervals, for a total of N = 96 steps.

[0128] f is a linear term in the objective function:

[0129] 2) Setting constraints includes:

[0130] ① System power balance constraint: P pv (t)+P bat (t)+P grid (t)=P load (t)+P sell (t)

[0131] ② Dynamic constraints on energy storage:

[0132] SOC Update:

[0133] SOC range: SOC min ≤SOC(t)≤SOC max

[0134] ③ Charge and discharge power constraints:

[0135] Charging power constraints:

[0136] Discharge power constraint:

[0137] ④ Mains power interaction constraints:

[0138] Power purchase limits:

[0139] Feedback power limit: 0≤P sell (t)≤P pv (t)

[0140] ⑤ Air source heat pump operating constraints:

[0141] Power limits:

[0142] Thermal power balance: Q solar (t)+Q pump (t)=Q load (t)+Q loss (t)+Q′ loss (t)

[0143] Among them, c grid For real-time electricity pricing, c sell To sell electricity at real-time prices, P grid For the purchased power capacity, P sell k represents the electricity sold. bat P is the energy storage loss coefficient. charge (t) represents the energy storage charging power, P discharge (t) represents the energy storage discharge power, k charge k is the charging loss coefficient. discharge P is the discharge loss coefficient; pv Photovoltaic output power; P load For the total load power, C rated For the battery's rated capacity, SOC min Minimum battery state of charge (SOC) max This represents the maximum state of battery charge. For minimum and maximum energy storage charging power, For minimum and maximum energy storage discharge power, For maximum power purchase capacity, This represents the maximum heating power of the air source heat pump.

[0144] S44. Solve the multi-objective rolling optimization and real-time scheduling model using the interior point method, including:

[0145] Introducing a slack variable s≥0 transforms the inequality constraint into an equality, and adding a logarithmic barrier term:

[0146]

[0147] Where μ is the barrier parameter, which gradually decreases to zero during Newton's iterations.

[0148] Finally, the optimal solution U is obtained. * ,

[0149] Among them, U * Indicates power dispatching strategies, including: Charge power command for the battery. This is a battery discharge power command. For power purchase orders, This is a power sales instruction.

[0150] Based on real-time environmental data, the system uses a PID controller, multi-objective rolling optimization, and real-time scheduling model to output the optimal control sequence for the next 24 hours. This sequence includes: energy storage charging and discharging power, mixing valve opening, solar thermal heating circulation velocity, and mains power interaction power. The intelligent decision-making module sends control commands in real-time to the integrated photovoltaic and solar thermal equipment, air source heat pump, and mains power transmission actuators. The system also employs a 15-minute rolling update and re-optimization process.

[0151] In summary, this system effectively integrates photovoltaic power generation, solar thermal recovery, and microgrid dispatch, significantly improving energy utilization efficiency.

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A model predictive control method for photovoltaic-thermal integrated greenhouses, characterized in that, This is implemented based on a model predictive control system for photovoltaic-thermal integrated greenhouses; the model predictive control system for photovoltaic-thermal integrated greenhouses includes: Concentrated photovoltaic and solar thermal integrated module, data acquisition module, and intelligent decision-making module; The concentrated photovoltaic-thermal integrated module includes: a concentrated photovoltaic-thermal integrated device, a hot water storage tank, a temperature regulation device, a battery, and a grid interaction unit; the concentrated photovoltaic-thermal integrated device converts solar energy into electrical energy and thermal energy; The storage battery stores the electrical energy generated by photovoltaic conversion and supplies power to the greenhouse equipment when photovoltaic power generation is insufficient; the grid interaction unit is used to interact with the external power grid when there is a surplus or shortage of electrical energy. Heat energy is supplied to the greenhouse through a hot water storage tank; a temperature regulation device assists the concentrating photovoltaic-thermal integrated equipment to achieve temperature control inside the greenhouse; The data acquisition module collects greenhouse environmental parameters and operational data from each actuator of the integrated photovoltaic and solar thermal module in real time, and transmits them to the intelligent decision-making module. The intelligent decision-making module outputs greenhouse air temperature control commands and power dispatch strategies based on the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic and solar thermal integrated module, and sends them to the corresponding actuators of the concentrated photovoltaic and solar thermal integrated module. The control method includes: Collect greenhouse environmental parameters and operational data from each actuator of the concentrated photovoltaic-thermal integrated module; Based on the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic and solar thermal integrated module, the PID controller outputs greenhouse air temperature control commands. Based on solar irradiance and greenhouse air temperature, a long short-term memory network is used to output photovoltaic power generation prediction values; The model predictive control algorithm uses the predicted photovoltaic power generation value and the heating energy consumption value of the temperature regulation device to output an energy dispatch strategy, including: Using the predicted value of photovoltaic power generation and the heating energy consumption value of the temperature regulation device as inputs, and the operating cost ratio, photovoltaic absorption rate, grid power dependence and air source heat pump energy consumption ratio as optimization objectives, a model predictive control framework is established; and an energy dispatch strategy is generated through rolling time domain optimization, while constraints are applied to ensure the system safety boundary. The optimization objective function is: in, For operating costs, For photovoltaic power absorption rate, Dependence on mains power To maximize operating costs, This represents the total energy consumption of the air source heat pump during continuous operation at maximum power. Energy consumption of air source heat pumps; The corresponding actuators are controlled by commands for controlling greenhouse air temperature and energy dispatch strategies.

2. The method according to claim 1, characterized in that, The integrated photovoltaic and solar thermal concentrator includes: Photovoltaic panels, interference filters, vacuum collector tubes, and auxiliary supports; The photovoltaic panel is parabolic in shape, with the open end of the parabola facing the direction of solar incidence; The interference filter film is applied to the light-facing surface of the photovoltaic panel, selectively transmitting and reflecting the full spectrum of light irradiated onto the surface of the interference filter film. The vacuum heat collection tubes are arranged parallel to the focal line of the parabolic surface. The light reflected by the interference filter film is focused onto the surface of the vacuum heat collection tubes to collect photothermal energy. The vacuum heat collection tubes are frustum bodies with several water inlets on the bottom surface and one water outlet on the top surface. The heating circulating water flows in from the water inlets and flows out from the water outlet. The auxiliary support is installed on the non-opening side of the photovoltaic panel to support the photovoltaic panel.

3. The method according to claim 1, characterized in that, It also includes a visualization monitoring module; The visualization monitoring module presents photovoltaic power generation forecast data and greenhouse environment data in a visual format.

4. The method according to claim 1, characterized in that, Based on the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic-thermal integrated module, the PID controller outputs greenhouse air temperature control commands, including: The Kalman filter algorithm is used to reduce noise in the greenhouse environmental parameters and the operating data of each actuator of the concentrated photovoltaic and solar thermal integrated module to obtain noise-reduced data. The state equation and observation equation are constructed based on noise-reduced data to characterize the photothermal output of the concentrated photovoltaic-thermal integrated equipment; The solar thermal output is compared with the preset temperature using PID calculation to determine whether the solar thermal output meets the heat load requirements. If the requirements are met, the mixing ratio parameters and flow rate control parameters of the hot water storage tank are output to regulate the greenhouse air temperature. If the requirements are not met, the heat output power of the temperature regulation device is calculated, and the mixing ratio parameters and flow rate control parameters of the hot water storage tank are output to maintain the greenhouse air temperature within the target range.

5. The method according to claim 4, characterized in that, The mixing ratio parameters and flow rate control parameters of the hot water storage tank include: in: It is the control quantity generated by the PID algorithm; For mixing ratio, These are the minimum and maximum mixing ratios, respectively. These are the lower and upper threshold values ​​for the PID output, respectively. This is the proportionality coefficient; For solar thermal heating circulation velocity, These are the minimum and maximum flow velocities, respectively. This is the curve steepness coefficient, used to control response sensitivity; This is the center threshold of the PID output.

6. The method according to claim 4, characterized in that, The formula for calculating the photothermal output power is: The formula for calculating the heat output power of the temperature regulating device is as follows: in, The photothermal conversion efficiency coefficient. The velocity attenuation coefficient is... The effective heat collection area of ​​the photovoltaic thermal receiver; The coefficient of performance (COP) of an air source heat pump. Efficiency factor for air source heat pumps; This indicates the heat load demand.

7. The method according to claim 1, characterized in that, Operating costs: Photovoltaic grid integration rate: Dependence on mains power: in, For real-time electricity prices, To provide real-time electricity pricing, For the power purchase capacity, For electricity sales capacity, For energy storage charging power, For energy storage discharge power, This is the charging loss coefficient. This is the discharge loss coefficient; Photovoltaic output power; Total load power, This represents the total time.

Images