Photovoltaic cooling equipment control methods, devices, photovoltaic cooling equipment and storage media

Abstract

This invention relates to a control method, apparatus, photovoltaic cooling device, and storage medium for photovoltaic cooling equipment. The method involves inputting first operating parameters of the photovoltaic cooling device into a trained first model, which outputs a predicted value of photovoltaic power generation within a first time period. Second operating parameters of the photovoltaic cooling device are input into a trained second model, which outputs a predicted value of the cooling demand of the target object within the first time period. The operating mode of the photovoltaic cooling device is controlled based on the predicted photovoltaic power generation and cooling demand. Therefore, photovoltaic power generation and cooling demand can be predicted using different operating parameters, and the operating mode can be adjusted accordingly to match photovoltaic power generation and cooling demand, avoiding insufficient or excessive cooling storage and improving the operating efficiency and energy utilization rate of the photovoltaic air conditioning system under different operating parameters.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic air conditioning technology, and in particular to a photovoltaic refrigeration equipment control method, device, photovoltaic refrigeration equipment and storage medium. Background Technology

[0002] With the widespread application of renewable energy, photovoltaic (PV) air conditioning systems have attracted considerable attention due to their energy-saving characteristics. However, existing PV air conditioning systems generally suffer from low energy utilization efficiency during operation. Specifically, under different regional and seasonal conditions, the varying operating parameters of PV air conditioners lead to significant differences between the air conditioning load and PV energy output. This results in the system operating at high load when energy is insufficient or at low load when energy is sufficient, severely impacting system operating efficiency. Furthermore, traditional cold storage strategies rely on fixed settings or experience-based judgments, lacking dynamic response mechanisms to weather changes, user behavior, and equipment status, leading to insufficient or excessive cold storage and resulting in resource waste.

[0003] Therefore, how to improve the operating efficiency of photovoltaic air conditioning systems under different operating parameters has become an urgent problem to be solved. Summary of the Invention

[0004] In view of this, in order to solve the above-mentioned technical problem of how to improve the operating efficiency of photovoltaic air conditioning systems under different operating parameters, the present invention provides a photovoltaic cooling equipment control method, device, photovoltaic cooling equipment and storage medium.

[0005] In a first aspect, embodiments of the present invention provide a photovoltaic cooling equipment control method, comprising:

[0006] The first operating parameters of the photovoltaic cooling equipment are input into the trained first model so that the photovoltaic power generation of the photovoltaic cooling equipment in the first time period is output through the first model. The first operating parameters represent the parameters that affect the power generation of the photovoltaic cooling equipment.

[0007] The second operating parameters of the photovoltaic cooling equipment are input into the trained second model so that the second model can output the predicted cooling demand of the target object in the first time period. The second operating parameters represent the parameters that affect the cooling demand of the target object.

[0008] The operating mode of the photovoltaic cooling equipment is controlled based on the predicted photovoltaic power generation and the predicted cooling demand during the first time period.

[0009] In one possible implementation, inputting the first operating parameters of the photovoltaic cooling device into the trained first model includes:

[0010] The system obtains the real-time light intensity for the photovoltaic cooling device as light intensity information, the historical power generation sequence of the photovoltaic cooling device per hour in the second time period as historical power generation information, the seasonal solar radiation coefficient of the current season, and the equipment aging coefficient of the photovoltaic cooling device.

[0011] The light intensity information, the historical power generation information, the seasonal sunshine coefficient, and the equipment aging coefficient are used as the first operating parameters and input into the first model.

[0012] In one possible implementation, inputting the second operating parameters of the photovoltaic cooling device into the trained second model includes:

[0013] The system obtains the usage time of the photovoltaic cooling equipment within the third time period as historical usage information, as well as the current weather and future weather as weather information, and the temperature difference information between indoors and outdoors.

[0014] The historical usage information, the weather information, and the temperature difference information are input into the second model as the second operating parameters.

[0015] In one possible implementation, controlling the operating mode of the photovoltaic cooling equipment during the first time period based on the predicted photovoltaic power generation and the predicted cooling demand includes:

[0016] When the predicted value of photovoltaic power generation is greater than the product of the predicted value of cooling demand and the first coefficient, the working mode of the photovoltaic cooling equipment is switched to pure photovoltaic drive mode.

[0017] Alternatively, when the predicted value of photovoltaic power generation is less than the product of the predicted value of cooling demand and the second coefficient, and the current cooling capacity of the photovoltaic cooling equipment is greater than or equal to the product of the predicted value of cooling demand and the third coefficient, the operating mode of the photovoltaic cooling equipment is switched to the phase change material driven mode.

[0018] Alternatively, when the predicted value of photovoltaic power generation is less than the product of the predicted value of cooling demand and the fourth coefficient, and the current cooling capacity of the photovoltaic cooling equipment is less than the product of the predicted value of cooling demand and the second coefficient, the operating mode of the photovoltaic cooling equipment is switched to a hybrid driving mode of photovoltaic and phase change materials, wherein the first coefficient is greater than the third coefficient, the second coefficient is greater than the fourth coefficient.

[0019] In one possible implementation, the method further includes:

[0020] When the photovoltaic cooling equipment operates in hybrid drive mode, the supply-demand difference of the photovoltaic cooling equipment is calculated using the following formula:

[0021] Supply-demand difference = Forecasted demand for cooling capacity - (Current cooling capacity of the photovoltaic cooling equipment + Forecasted photovoltaic power generation × Fifth coefficient × Energy efficiency ratio of the photovoltaic cooling equipment);

[0022] When the supply-demand difference is greater than the first threshold, the operating mode of the photovoltaic cooling equipment is switched to photovoltaic, phase change material and mains power drive mode.

[0023] In one possible implementation, the method further includes:

[0024] When the predicted value of photovoltaic power generation is greater than the product of the predicted value of cooling demand and the sixth coefficient, and the predicted value of cooling demand in the fourth time period in the future shows an upward trend, the phase change material of the photovoltaic cooling equipment is controlled to store cold, and the sixth coefficient is greater than the first coefficient.

[0025] In one possible implementation, the method further includes:

[0026] The refrigerant flow rate of the photovoltaic refrigeration device is adjusted according to the cold storage efficiency of the phase change material so that the cold storage efficiency is greater than the second threshold.

[0027] In a second aspect, embodiments of the present invention provide a photovoltaic cooling equipment control device, comprising:

[0028] The first control module is used to input the first operating parameters of the photovoltaic cooling equipment into the trained first model, so as to output the predicted value of the photovoltaic power generation of the photovoltaic cooling equipment in the first time period through the first model. The first operating parameters represent the parameters that affect the power generation of the photovoltaic cooling equipment.

[0029] The second control module is used to input the second operating parameters of the photovoltaic cooling equipment into the trained second model, so as to output the predicted value of the cooling demand of the target object in the first time period through the second model. The second operating parameters represent the parameters that affect the cooling demand of the target object.

[0030] The third control module is used to control the operating mode of the photovoltaic cooling equipment during the first time period based on the predicted photovoltaic power generation and the predicted cooling demand.

[0031] Thirdly, embodiments of the present invention provide a photovoltaic cooling device, including: a processor and a memory, wherein the processor is configured to execute a photovoltaic cooling device control program stored in the memory to implement the photovoltaic cooling device control method described in any one of the first aspects above.

[0032] Fourthly, embodiments of the present invention provide a storage medium storing one or more programs, which can be executed by one or more processors to implement the photovoltaic cooling device control method described in any one of the first aspects.

[0033] The photovoltaic cooling equipment control scheme provided in this invention involves inputting a first operating parameter of the photovoltaic cooling equipment into a trained first model, which outputs a predicted value of photovoltaic power generation of the photovoltaic cooling equipment within a first time period. The first operating parameter represents a parameter affecting the power generation of the photovoltaic cooling equipment. A second operating parameter of the photovoltaic cooling equipment is input into a trained second model, which outputs a predicted value of the cooling demand of a target object within the first time period. The second operating parameter represents a parameter affecting the cooling demand of the target object. The operating mode of the photovoltaic cooling equipment within the first time period is controlled based on the predicted photovoltaic power generation and the predicted cooling demand. Therefore, photovoltaic power generation and cooling demand can be predicted using different operating parameters of the equipment, and the operating mode of the equipment can be adjusted according to these parameters to match photovoltaic power generation and cooling demand, avoiding insufficient or excessive cooling storage, and improving the operating efficiency and energy utilization rate of the photovoltaic air conditioning system under different operating parameters. Attached Figure Description

[0034] Figure 1 A schematic flowchart of a photovoltaic cooling equipment control method provided in an embodiment of the present invention;

[0035] Figure 2 A schematic flowchart of another photovoltaic cooling equipment control method provided in an embodiment of the present invention;

[0036] Figure 3 This is a schematic diagram of the structure of a photovoltaic cooling equipment control device provided in an embodiment of the present invention;

[0037] Figure 4 This is a structural schematic diagram of a photovoltaic cooling device provided in an embodiment of the present invention. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] To facilitate understanding of the embodiments of the present invention, further explanations and descriptions will be provided below with reference to the accompanying drawings and specific embodiments. These embodiments do not constitute a limitation on the embodiments of the present invention.

[0040] Figure 1 This is a flowchart illustrating a photovoltaic cooling equipment control method provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method specifically includes:

[0041] S11. Input the first operating parameters of the photovoltaic cooling equipment into the trained first model, so as to output the predicted value of photovoltaic power generation of the photovoltaic cooling equipment in the first time period through the first model. The first operating parameters represent the parameters that affect the power generation of the photovoltaic cooling equipment.

[0042] The photovoltaic cooling equipment control method provided in this invention is applied to photovoltaic cooling equipment, which may include, but is not limited to, photovoltaic air conditioners. It is suitable for air conditioning systems in residential and commercial buildings (such as office buildings and shopping malls) that require photovoltaic power supply in different regions and seasons. Specifically, it predicts the photovoltaic power generation and cooling demand through different operating parameters of the photovoltaic cooling equipment, and adjusts the equipment working mode according to the photovoltaic power generation and cooling demand to match the photovoltaic power generation and cooling demand.

[0043] In this embodiment, the first operating parameter is the core parameter that characterizes the power generation of the photovoltaic cooling equipment, and may include, but is not limited to, the following: 1. Real-time environmental parameters: real-time value of photovoltaic panel irradiance (e.g., collected once every minute by a local sensor), outdoor temperature and humidity (which, together with irradiance, affect power generation efficiency), etc.; 2. Historical operating parameters: power generation curve of the photovoltaic cooling equipment within a historical time period (e.g., a historical 72-hour period) (one sampling point per hour; the power generation curve characterizes the sequence of photovoltaic power generation per hour and can be used as a feature input to help the model predict the short-term trend of future photovoltaic power generation (e.g., daytime peak, nighttime decline, etc.)), covering different historical time periods (e.g., within one year). 3. Equipment characteristic parameters: 1. Historical power generation data of weather conditions (e.g., sunny, cloudy, rainy days, etc.) (used as basic data for model training); 2. Preset seasonal coefficient of photovoltaic panels (e.g., divided into 4 levels according to spring, summer, autumn and winter, with values ​​ranging from 0.8 to 1.2, reflecting the impact of seasons on sunlight, used to distinguish the photovoltaic energy available in different seasons, which can be obtained by obtaining the seasonal radiation of the region through the meteorological database, and then comparing it with the sunlight of different seasons in the standard year (the meteorological data of the standard year can be publicly obtained), and setting the ratio between them as the coefficient); 3. Equipment aging coefficient (calculated based on the photovoltaic panel attenuation model, with an annual attenuation rate ≤1.5%, reflecting the impact of equipment loss on power generation).

[0044] The first model is a photovoltaic energy prediction model, which can be constructed using the XGBoost algorithm (extreme gradient boosting algorithm). The core objective of the model is to output the "predicted value of photovoltaic power generation in the first time period". The first time period is usually the same day (within 24 hours), and the prediction results are dynamically updated every hour.

[0045] During the first model training, historical operating data of photovoltaic cooling equipment (including power generation data covering different weather scenarios within a historical time period) can be selected as training data. The training data can be divided into a training set (for model parameter learning), a validation set (for adjusting model hyperparameters), and a test set (for evaluating the final model performance) according to a preset ratio (e.g., 8:1:1). The daily actual photovoltaic power generation in the historical data is used as a label, associated with the first operating parameter of the corresponding date to form training sample pairs. The preprocessed operating parameters are used as model input features. Based on the XGBoost algorithm framework, the model parameters are iteratively optimized with the goal of minimizing the difference between predicted and actual power generation (MAPE). The model performance is verified using the validation set, and hyperparameters such as the learning rate and tree depth are adjusted to ensure that the MAPE error on the validation set is controlled within ±10%. The first operating parameters of the test set are input into the trained model, and the predicted photovoltaic power generation value corresponding to the test set is output. The MAPE error of the test set is calculated. If the test set error is ≤ ±10%, the model training is complete; if the error exceeds the threshold, the hyperparameters are readjusted or training data is supplemented until the error requirement is met.

[0046] The trained first model is embedded into the local edge computing node of the photovoltaic cooling equipment to receive newly collected first operating parameters in real time. At midnight every day, the predicted power generation and the actual power generation are automatically compared. If the error is greater than ±5%, incremental training is triggered (updating the model parameters with the new data of the day). Full retraining is performed every month, and the operating data of the latest month is added to the training set to ensure the long-term robustness of the model.

[0047] During prediction, the first operating parameters of the photovoltaic cooling equipment are preprocessed and then input into the trained first model. Based on the input parameters, the first model calls the learned power generation laws to calculate and output the predicted value of photovoltaic power generation in the first time period. At the same time, the predicted value can be updated once an hour to adapt to real-time parameter changes (such as fluctuations in light intensity).

[0048] In one possible implementation, the first operating parameters of the photovoltaic cooling device are input into the trained first model, including:

[0049] The system acquires real-time irradiance for the photovoltaic cooling equipment as irradiance information, historical power generation sequence of the photovoltaic cooling equipment per hour in the second time period as historical power generation information, seasonal solar irradiance coefficient of the current season, and equipment aging coefficient of the photovoltaic cooling equipment. The irradiance information, historical power generation information, seasonal solar irradiance coefficient, and equipment aging coefficient are used as first operating parameters and input into the first model.

[0050] In this embodiment, real-time light intensity is obtained as follows: a high-precision light sensor is installed near the photovoltaic panel array of the photovoltaic cooling equipment to obtain the real-time light intensity. Real-time light intensity data can be automatically collected at a frequency of once per minute. The hourly historical power generation sequence within the second time period (within the historical 72 hours) is obtained as follows: historical power generation data is extracted from the inverter data interface of the photovoltaic cooling equipment. The inverter automatically records the cumulative power generation per hour. The extracted data is arranged in chronological order to obtain the hourly historical power generation sequence. The seasonal solar irradiance coefficient of the previous season is obtained as follows: Seasonal division: it is divided into four levels according to the natural seasons: spring, summer, autumn, and winter. The edge computing node has a built-in season judgment program, which automatically identifies the season according to the current date, presets a seasonal solar irradiance coefficient reference table (for example, spring: 1.0, summer: 1.2, autumn: 1.0, winter: 0.8), and automatically matches the coefficient corresponding to the current season. The equipment aging factor is obtained in the following ways: based on the degradation model of the photovoltaic panel of the photovoltaic cooling equipment (for example, the formula is: equipment aging factor = 1 - (years of use of photovoltaic panel × annual degradation rate), where the annual degradation rate is set to ≤1.5% according to industry standards), the manufacturing date of the photovoltaic panel is retrieved from the asset management system of the photovoltaic cooling equipment, and the current years of use are calculated.

[0051] S12. Input the second operating parameters of the photovoltaic cooling equipment into the trained second model, so as to output the predicted value of the cooling demand of the target object in the first time period through the second model. The second operating parameters represent the parameters that affect the cooling demand of the target object.

[0052] In this embodiment, the second operating parameter is a core parameter characterizing the cooling demand of the target object (such as a residential or office building covered by photovoltaic air conditioning). It may include, but is not limited to: 1. Parameters of the target object's historical air conditioning usage (e.g., the target object's users' historical 24-hour air conditioning usage time periods, preferred air conditioning temperatures, etc.). Historical operation logs can be extracted from the photovoltaic air conditioning system's controller or user APP, and the data can be organized according to "hourly usage duration" (e.g., 45 minutes of usage in an hour is recorded as 0.75 hours) to form a 24-hour user behavior sequence. If it is a new system without historical data, a user behavior template of the same type of building can be imported by default (e.g., office buildings default to "high usage frequency from 9:00 to 18:00"). 2. Temperature parameters, which may include: the real-time temperature difference between the indoor and outdoor areas of the target object, the temperature change trend recorded over a historical period, and the predicted temperature change trend over a future period. This can be obtained and calculated by installing temperature and humidity sensors near the photovoltaic panels indoors and outdoors of the target object and acquiring network data. 3. Weather parameters, which may include: the weather trend (probability of rainfall, probability of cloudy weather) for the target area in the near future, historical weather trends, and the current season type, etc., which can be obtained by calling a third-party meteorological API. 4. Equipment-related parameters, which may include: the current operating status of the photovoltaic cooling equipment (e.g., whether the compressor is on), the spatial area of ​​the target object (e.g., a 100㎡ residence), and the building insulation coefficient (e.g., the external wall thermal resistance R-value), etc.

[0053] The second model is a cooling demand forecasting model, which can use LSTM (Long Short-Term Memory Network). The core objective is to output the predicted cooling demand of the target object within the first time period (e.g., the current day or the next 24 hours).

[0054] During training, the second model collects historical data of the target object over a specific period, which may include: daily operating parameters (user behavior sequences, temperature difference sequences, meteorological trend data, etc.) and actual daily cooling energy consumption (obtained from the air conditioner meter or system energy consumption statistics module) as training data. The training data is divided into training, validation, and test sets according to a preset ratio (e.g., 7:2:1); all data can be aligned by hour to form hourly operating parameters and hourly cooling energy consumption sample pairs. This data is then input into the initial model for training (e.g., input sequence length = 24 hours, output cooling energy demand curve for the next 24 hours).

[0055] The trained LSTM model is embedded into the edge computing node of the photovoltaic air conditioning system (sharing hardware such as Raspberry Pi with the first model), supporting real-time parameter reception and fast inference.

[0056] The second operating parameters are preprocessed and then integrated into a 24-hour sequence in chronological order to form the model input array, which is then input into the second model. This allows the second model to output hourly and overall demand cooling capacity forecasts for the first time period; the forecast results are simultaneously stored in the local database.

[0057] In one possible implementation, after obtaining the first and second operating parameters, preprocessing is required. Specifically, all collected first and second operating parameters need to be uploaded to a local edge computing node (such as a Raspberry Pi) or a cloud data processing center, and the following preprocessing operations are performed to ensure parameter validity: 1. Data cleaning: Using the Laida criterion (3σ criterion), gross errors (such as abrupt changes caused by sensor malfunctions) in parameters such as light intensity and power generation are removed; 2. Normalization: Using the Min-Max normalization method, all parameters are mapped to the [0,1] interval to eliminate dimensional differences (such as light intensity in lux and power generation in kWh); 3. Feature extraction: Further extracting derived features from the preprocessed data, such as calculating the light intensity slope (reflecting the trend of light change), ultimately forming a standardized dataset. After preprocessing, the data is then input into the first and second models for further processing.

[0058] In one possible implementation, the second operating parameters of the photovoltaic cooling device are input into the trained second model, including: obtaining the usage time of the photovoltaic cooling device in a third time period as historical usage information, obtaining the current weather and future weather as weather information, and obtaining the temperature difference information between indoors and outdoors; and inputting the historical usage information, weather information and temperature difference information as the second operating parameters into the second model.

[0059] In this embodiment, when the second operating parameters include historical usage information, weather information, and temperature difference information, they can be obtained in the following ways: 1. Historical usage information represents the usage time of the photovoltaic cooling equipment within the third time period (e.g., 24 hours), reflecting the temporal distribution pattern of cooling demand. This can be obtained from the operation log of the photovoltaic air conditioner controller or the usage records of the user's APP. 2. Weather information represents the current and future weather, covering key meteorological factors affecting cooling demand to ensure that the fluctuation of cooling demand in the first time period can be predicted. The current weather can be obtained by acquiring real-time meteorological data at the prediction start time, which may include: current rainfall status, current outdoor temperature and humidity, etc., used as the basic benchmark for calibrating cooling demand. Future weather can be obtained by acquiring hourly meteorological forecast data for the first time period, which may include: hourly rainfall probability, hourly cloudy probability, outdoor temperature and humidity, etc., directly related to the adjustment coefficient of cooling demand (e.g., cooling demand +15% when rainfall probability > 30%). 3. Temperature difference information is the difference between indoor and outdoor temperatures, which is a key physical factor affecting the intensity of cooling demand and can be obtained through temperature sensors.

[0060] S13. Control the working mode of the photovoltaic cooling equipment during the first time period based on the predicted value of photovoltaic power generation and the predicted value of cooling demand.

[0061] In this embodiment, the predicted value of photovoltaic power generation and the predicted value of cooling demand are compared to determine the current working mode of the photovoltaic cooling equipment. Different working modes adopt different driving methods, which may include: pure photovoltaic drive, phase change material drive, hybrid drive, etc. When the predicted value of photovoltaic power generation is high, pure photovoltaic drive can be adopted. When the predicted value of cooling demand is high and the cooling capacity of the photovoltaic cooling equipment is high, since the cooling storage system of the photovoltaic cooling equipment stores cold through phase change materials, phase change material drive can be adopted. When the predicted value of cooling demand is high and both the predicted value of photovoltaic power generation and the cooling capacity are low, hybrid drive of photovoltaic and phase change materials can be adopted.

[0062] Specifically, this may include: when the predicted value of photovoltaic power generation is greater than the product of the predicted value of cooling demand and the first coefficient, controlling the working mode of the photovoltaic cooling equipment to switch to pure photovoltaic drive mode;

[0063] Alternatively, when the predicted value of photovoltaic power generation is less than the product of the predicted value of cooling demand and the second coefficient, and the current cooling capacity of the photovoltaic cooling equipment is greater than or equal to the product of the predicted value of cooling demand and the third coefficient, the operating mode of the photovoltaic cooling equipment is switched to the phase change material driven mode.

[0064] Alternatively, when the predicted value of photovoltaic power generation is less than the product of the predicted value of cooling demand and the fourth coefficient, and the current cooling capacity of the photovoltaic cooling equipment is less than the product of the predicted value of cooling demand and the second coefficient, the operating mode of the photovoltaic cooling equipment is switched to a hybrid drive mode of photovoltaic and phase change materials, where the first coefficient is greater than the third coefficient, the second coefficient is greater than the fourth coefficient.

[0065] In this embodiment, coefficients can be preset, for example, the first coefficient = 1.1, the third coefficient = 0.9, the second coefficient = 0.8, and the fourth coefficient = 0.7. The operating mode of the photovoltaic cooling equipment is controlled according to the following formula after calculation and judgment: If the predicted photovoltaic power generation > the predicted cooling demand × the first coefficient, the control equipment switches to pure photovoltaic drive mode, shuts down the cold storage system, and only uses photovoltaic cooling; if the predicted photovoltaic power generation < the predicted cooling demand × the second coefficient, and the current cold storage capacity ≥ the predicted cooling demand × the third coefficient, the system switches to phase change material drive mode, using the cold storage capacity generated by the phase change material to drive the equipment cooling, with photovoltaic power used for auxiliary power supply; if the predicted photovoltaic power generation < the predicted cooling demand × the fourth coefficient, and the current cold storage capacity < the predicted cooling demand × the second coefficient, the photovoltaic and phase change material hybrid drive mode is activated, with photovoltaic cooling and phase change material cooling performed simultaneously, dynamically adjusting the energy supply ratio; the above parameters are scanned every 10 minutes, and the mode is switched in real time according to the value changes, with mode change notifications and expected energy saving rates pushed to the user's APP.

[0066] In one possible implementation, when the photovoltaic cooling device operates in a hybrid drive mode, the supply-demand difference of the photovoltaic cooling device is calculated using the following formula: supply-demand difference = predicted demand cooling capacity - (current cooling capacity of the photovoltaic cooling device + predicted photovoltaic power generation × fifth coefficient × energy efficiency ratio of the photovoltaic cooling device); when the supply-demand difference is greater than the first threshold, the operating mode of the photovoltaic cooling device is switched to photovoltaic, phase change material and grid power drive mode.

[0067] In this embodiment, the fifth coefficient is used to correct the photovoltaic power generation forecast to the actual energy conversion efficiency that can be used for cooling. Considering factors such as photovoltaic power transmission loss and equipment standby power consumption, it can be set to 0.95. For older photovoltaic systems, where power transmission loss is higher, it can be lowered to 0.92-0.94; for newly built high-efficiency photovoltaic systems, it can be raised to 0.96-0.97. The first threshold is used to determine whether the supply-demand difference triggers a mode switch. When the supply-demand difference is greater than the first threshold, it indicates that the energy supply from photovoltaics and phase change materials can no longer meet the cooling demand, and mains power needs to be introduced to supplement it.

[0068] When the photovoltaic cooling equipment is in a hybrid drive mode of photovoltaic and phase change material, data is collected by the temperature sensor of the cold storage tank and the phase change material status monitoring equipment (infrared thermal imager). The current cold storage capacity is calculated as "remaining capacity of the cold storage tank + current cold storage rate of the phase change material × total cold storage capacity of the phase change material". For example, if the remaining capacity of the cold storage tank is 1.2 kWh and the cold storage rate of the phase change material is 60% (total cold storage capacity is 2 kWh), then the current cold storage capacity = 1.2 + 0.6 × 2 = 2.4 kWh.

[0069] The energy efficiency ratio can be read from the nameplate or controller of the photovoltaic air conditioning equipment (e.g., summer cooling energy efficiency ratio COP=3.2). If the equipment is operating under partial load (e.g., the compressor is not at full load), it should be corrected according to "rated COP × current load rate".

[0070] The supply-demand difference is calculated as: Demand for cooling capacity - (Current cooling capacity + Forecasted photovoltaic power generation × Fifth coefficient × Energy efficiency ratio). For example, if the forecasted cooling capacity is 2.5 kWh, current cooling capacity is 1.8 kWh, forecasted photovoltaic power generation is 0.6 kWh, the fifth coefficient is 0.95, and the energy efficiency ratio is 3.0, the supply-demand difference is 2.5 - (1.8 + 0.6 × 0.95 × 3.0) = 2.5 - (1.8 + 1.71) = 2.5 - 3.51 = -1.01 kWh (in this case, the supply-demand difference is negative). (No mode switching required). For example, if the predicted cooling demand is 3.0 kWh, the current cooling storage is 1.5 kWh, the predicted photovoltaic power generation is 0.5 kWh, the fifth coefficient is 0.95, the energy efficiency ratio is 3.0, and the supply-demand difference is 3.0 - (1.5 + 0.5 × 0.95 × 3.0) = 3.0 - (1.5 + 1.425) = 3.0 - 2.925 = 0.075 kWh, and the first threshold is set to 0.05 kWh, then the supply-demand difference (0.075) is greater than the threshold, triggering a mode switch. That is, when the supply-demand difference is greater than the first threshold, the photovoltaic, phase change material, and mains-driven modes are immediately triggered to simultaneously supply power for cooling.

[0071] After connecting to the grid, the power supply ratio is adjusted every 5 minutes based on the real-time supply-demand difference: if the supply-demand difference is still greater than the first threshold, the grid power supply ratio is gradually increased (e.g., 5% each time, up to 50%), while the photovoltaic power supply ratio is reduced (minimum 20%); if the supply-demand difference drops below the first threshold, the grid power supply ratio is gradually reduced (e.g., 5% each time, down to 10%), and the photovoltaic power supply ratio is prioritized to increase (if the photovoltaic power generation forecast increases); the phase change material power supply ratio is always maintained at 30%-50% to avoid excessive release leading to insufficient cooling storage in subsequent periods. Meanwhile, the edge computing node continuously monitors the mains power consumption, actual photovoltaic power generation, and phase change material cold storage rate. It records the power supply duration and energy consumption data of each module every 10 minutes. If the mains power consumption exceeds the expected value for one consecutive hour, it prompts to check whether the air conditioner is faulty. When any of the following conditions are met, it automatically restores to the hybrid drive mode: the supply-demand difference is ≤ the first threshold for two consecutive calculations (and there is no sudden increase in demand for cooling); the predicted value of photovoltaic power generation increases significantly (e.g., from 0.5 kWh / hour to 1.2 kWh / hour), and the increase of "predicted value of photovoltaic power generation × fifth coefficient × energy efficiency ratio" is ≥ the supply-demand difference; the phase change material cold storage rate is replenished to more than 80%, and the current cold storage can cover part of the demand for cooling.

[0072] In one possible implementation, when the predicted value of photovoltaic power generation is greater than the product of the predicted value of cooling demand and the sixth coefficient, and the predicted value of cooling demand in the fourth time period in the future shows an upward trend, the phase change material of the photovoltaic cooling equipment is controlled to store cold, and the sixth coefficient is greater than the first coefficient.

[0073] In this embodiment, the product of the sixth coefficient and the predicted cooling demand is used to determine whether the photovoltaic power generation is sufficient, ensuring that the photovoltaic power supply not only covers the current cooling demand but also has surplus energy available for cooling storage. When the first coefficient is 1.1, the sixth coefficient can be set to 1.3. The fourth time period is used to determine whether the predicted cooling demand shows an upward trend, and can be set to the next 2-4 hours. Every 15 minutes, the edge computing node performs a dual determination on "photovoltaic power generation sufficiency" and "future cooling demand trend." Only when both conditions are met is phase change material cooling storage triggered. The specific determination steps are as follows:

[0074] The predicted cooling demand is multiplied by the sixth coefficient. If the predicted photovoltaic power generation is greater than the calculated result, the photovoltaic power supply is considered sufficient, meeting the basic conditions for cold storage energy. If not, the current cold storage determination is terminated, and the process waits for the next cycle. Hourly data for the fourth time period is extracted from the predicted cooling demand for the next 24 hours output by the second model. The linear regression slope of the predicted cooling demand for the fourth time period is calculated. If the slope is greater than 0, an upward trend is identified. Alternatively, the difference between the predicted value of the last hour and the predicted value of the first hour in the fourth time period is calculated. If the difference is greater than 0 and the proportion of the difference to the predicted value of the first hour is ≥10%, a significant upward trend is identified. If both sufficient photovoltaic power and an upward trend are met, a cold storage start command is generated to initiate the phase change material cold storage operation.

[0075] In one possible implementation, the refrigerant flow rate of the photovoltaic cooling device is adjusted according to the cold storage efficiency of the phase change material so that the cold storage efficiency is greater than a second threshold.

[0076] In this embodiment, the second threshold is the minimum standard for cold storage efficiency, used to determine whether the cold storage efficiency of the phase change material meets the standard. It needs to be set in conjunction with the characteristics of the phase change material (e.g., latent heat of phase change, thermal conductivity) and the system's energy-saving target to ensure efficient energy utilization during the cold storage process and avoid resource waste. The second threshold can be set to 85% to ensure that, at this threshold, the refrigerant flow rate can be adjusted quickly to meet the standard without excessive energy consumption.

[0077] During the phase change material (PCM) cold storage process, edge computing nodes calculate the cold storage efficiency in real time. The formula and data source are as follows: Cold storage efficiency = (actual cold storage increment / theoretical cold storage energy input) × 100%. The actual cold storage increment is calculated by monitoring the change in the cold storage rate of the PCM using an infrared thermal imager (e.g., from 60% to 62% within 5 minutes) and combining it with the total cold storage capacity of the PCM (e.g., 2 kWh), i.e., "cold storage rate increment × total cold storage capacity" (e.g., (62%-60%) × 2 = 0.04 kWh); Theoretical cold storage energy input is calculated based on the refrigerant flow rate, refrigerant specific heat capacity, refrigerant inlet and outlet temperature difference, and time. The formula is "refrigerant flow rate × specific heat capacity × temperature difference × time" (e.g., refrigerant flow rate 0.3 kg / 5 min, theoretical energy input = 0.3 × 4.2 × (3-1) × 5 / 60 ≈ 0.21 kWh).

[0078] Calculation example: If the actual increase in cold storage capacity is 0.04 kWh and the theoretical cold storage energy input is 0.21 kWh, then the cold storage efficiency = (0.04 / 0.21) × 100% ≈ 19.05% (at this point, it is far below the second threshold, and the refrigerant flow rate needs to be adjusted immediately).

[0079] The cold storage controller aims to achieve a cold storage efficiency greater than the second threshold. Based on the difference between the real-time calculated cold storage efficiency and the second threshold, it dynamically adjusts the refrigerant flow rate using a combination of step-wise regulation and PID fine-tuning. Specifically: The cold storage controller first obtains the current refrigerant flow rate, real-time cold storage efficiency, and the second threshold, calculating the efficiency difference as the difference between the second threshold and the real-time cold storage efficiency. If the real-time cold storage efficiency is greater than the second threshold, the current refrigerant flow rate is maintained, and the efficiency is recalculated every 5 minutes without adjustment. If the real-time cold storage efficiency is less than or equal to the second threshold, the adjustment range is determined based on the efficiency difference (the larger the difference, the larger the adjustment range, avoiding prolonged inefficient operation). Adjustment ranges are divided based on the efficiency difference, corresponding to different flow rate adjustment ranges to ensure rapid reduction of the efficiency gap. Specific ranges and adjustment rules may include: when 0% < efficiency difference ≤ 5%, increase the current flow rate by 10%; when 5% < efficiency difference ≤ 10%, increase the current flow rate by 15%; when the efficiency difference > 10%, increase the current flow rate by 20%.

[0080] If the cold storage efficiency still does not stabilize above the second threshold after step-by-step adjustment, PID fine-tuning is initiated. By optimizing the proportional (P), integral (I), and derivative (D) parameters, the efficiency can be accurately achieved.

[0081] The control method for photovoltaic cooling equipment provided in this invention integrates multi-source data such as light intensity, sunshine duration, seasonal sunshine coefficient, and weather conditions. It utilizes machine learning to train a prediction model, reducing the prediction error of the total energy of the photovoltaic panels. This provides a quantitative basis for the dynamic allocation of air conditioning operation modes, thereby avoiding energy waste or insufficiency. By introducing a machine learning-based model for calculating the difference between cooling supply and demand, the error is controlled within a very small range. This scientifically determines the amount of cooling that can be stored in advance, ensuring precise matching between cooling storage operations and actual demand, thus improving the overall energy efficiency of the system. Furthermore, by combining phase change material-driven and photovoltaic-driven modes, flexible switching of the air conditioning operation mode at night is achieved, further enhancing the system's adaptability and energy-saving performance.

[0082] As an example Figure 2 A schematic flowchart of another photovoltaic cooling equipment control method provided in an embodiment of the present invention is shown below. Figure 2 As shown, the method specifically includes:

[0083] 1. Data Acquisition and Fusion:

[0084] Real-time operational data is collected via local sensors, including photovoltaic panel output power, irradiance, outdoor temperature and humidity, cold storage tank temperature, and phase change material (PCM) status (infrared thermal imager monitors the PCM interface and outputs the percentage of PCM progress in real time). The PCM status reflects the PCM's cold storage progress. If 40% of the PCM remains untransformed, it has absorbed 60% of its total heat capacity, leaving 40% of its cold storage capacity remaining. The phase change progress can be determined by detecting the solid-liquid interface position using an infrared thermal imager. Furthermore, by monitoring this parameter, the refrigerant flow rate can be adjusted. For example, the flow rate can be lower at the beginning of cold storage, increased appropriately in the middle stage, and lowered at the end to prevent overcooling and ensure efficient cold storage.

[0085] External data such as regional average sunshine duration (30-year historical monthly average), seasonal sunshine coefficient (divided into 4 levels according to spring / summer / autumn / winter, coefficient 0.8–1.2), and future weather change trends (such as probability of cloudy days and probability of rainfall) are obtained through meteorological databases.

[0086] All collected data is uploaded to a local edge computing node (such as a Raspberry Pi) or a cloud data processing center, and undergoes data cleaning (using the Laida criterion (3σ criterion) to remove gross errors), normalization (Min-Max normalization to [0,1]), and feature extraction (e.g., calculating the slope of light intensity, the rate of change of temperature and humidity, etc.) to form a standardized dataset (data is recorded once every 1 minute).

[0087] 2. Construction of photovoltaic energy prediction model:

[0088] Input features include, but are not limited to: real-time light intensity, historical 72-hour power generation curve (sampled every hour), daily seasonal coefficient, and equipment aging coefficient (calculated based on the photovoltaic panel degradation model, with an annual degradation rate ≤1.5%).

[0089] Model training:

[0090] Algorithm: Commonly used XGBoost can be selected;

[0091] Training data: 3 years of historical data (covering sunny / cloudy / rainy days), divided into 8:1:1 sets for training / validation / test;

[0092] Output: Daily photovoltaic power generation forecast (the error (MAPE) threshold can be set to ±10%; if this is not met, the model will be retrained), dynamically updated hourly.

[0093] Deployment: The model is embedded in edge nodes, and outputs prediction results after receiving new data in real time.

[0094] 3. Cold air demand forecasting and supply-demand gap calculation:

[0095] Cooling demand forecast:

[0096] Input: User's historical 24-hour air conditioner usage time, indoor-outdoor temperature difference (demand surges when ΔT≥5℃), and future weather trends (demand increases by 15% when the probability of rainfall is >30%).

[0097] Model: LSTM neural network (input sequence length = 24 hours, output cooling demand curve for the next 24 hours).

[0098] Supply and demand difference calculation:

[0099] Current status: Remaining capacity of the cold storage tank (kWh), current cold storage rate of the phase change material (%), and equipment insulation efficiency (calculated based on thermal resistance).

[0100] The formula for the difference is: Supply-demand difference = Forecasted cooling demand - (Current cooling storage capacity + Forecasted photovoltaic power generation × 0.95 × Air conditioning energy efficiency ratio);

[0101] Optimized model: SVR (key parameters can be selected as follows: kernel function is RBF, C=1.0, ε=0.05), prediction error threshold can be set to ±10%.

[0102] 4. Dynamic switching of nighttime operation mode:

[0103] Example as follows:

[0104] 1. When the mode type is pure photovoltaic drive, the triggering condition is: photovoltaic predicted power generation > cooling demand × 1.1, and the execution action is: shut down the cold storage system and the photovoltaic directly powers the air conditioner compressor;

[0105] 2. When the mode type is phase change material driven, the triggering condition is: photovoltaic power generation < cooling demand × 0.8, and cold storage capacity ≥ cooling demand × 0.9. The execution action is: release the cooling capacity of the phase change material, and the photovoltaic power supply assists the compressor.

[0106] 3. The mode type is: when running in a hybrid mode, the triggering condition is: photovoltaic power generation < cooling demand × 0.7 and cold storage capacity < cooling demand × 0.8. The execution action is: photovoltaic power supply 60% + phase change material power supply 40%, dynamically adjusted.

[0107] The current driving mode can also be determined by the relationship between the supply and demand difference and the preset threshold. The execution mechanism is as follows: the controller scans the difference every 10 minutes and switches the execution mode accordingly, and pushes mode change notifications (which may include the expected energy saving rate) through the user's APP.

[0108] 5. Coordinated optimization of cold storage strategies:

[0109] Cooling decision-making logic:

[0110] Pre-emptive cooling: For example, when the predicted photovoltaic power generation is greater than the cooling demand × 1.3 and the demand is predicted to increase in the next 24 hours, the cooling storage controller is activated, and the refrigerant flow rate is adjusted according to the PID algorithm (the target can be set as the phase change material cooling storage rate = excess photovoltaic power generation × 0.85).

[0111] Delayed cold storage: For example, when the photovoltaic power generation is less than 0.9 of the demand or the probability of cloudy days is greater than 70%, cold storage is stopped and the cold storage tank remains in its current state.

[0112] Execution Guarantee: The cold storage controller monitors the phase change material's phase change progress in real time (using infrared imaging analysis) and dynamically adjusts the refrigerant flow rate to ensure the cold storage efficiency of the phase change material is ≥90%.

[0113] 6. Feedback and Model Updates:

[0114] Feedback data:

[0115] This includes, but is not limited to, actual power generation, actual cooling consumption, and cooling storage efficiency.

[0116] Optimize the process:

[0117] The predicted value is automatically compared with the actual value at 0:00 every day, and the error is calculated.

[0118] When the error is greater than ±5%, incremental training is triggered (XGBoost / SVR model is updated using the new data of the day).

[0119] Full retraining is performed monthly (including the latest real-world data) to ensure the model's robustness and stability.

[0120] This embodiment can achieve accurate estimation of the total energy of photovoltaic panels based on a photovoltaic energy prediction model using multi-source data fusion and machine learning algorithms; optimize the calculation method of cooling supply and demand difference based on predicted energy and cooling demand, combined with weather changes, user behavior habits and equipment status; dynamically allocate nighttime air conditioning operation modes (pure photovoltaic drive and phase change material drive); and achieve precise matching of cooling storage operations and efficient utilization of resources based on the supply and demand difference optimization results of a collaborative control method for cooling storage strategies.

[0121] Figure 3 This is a schematic diagram of the structure of a photovoltaic cooling equipment control device provided in an embodiment of the present invention, as shown below. Figure 3 As shown, the device specifically includes:

[0122] The first control module 31 is used to input the first operating parameters of the photovoltaic cooling equipment into the trained first model, so as to output the predicted value of the photovoltaic power generation of the photovoltaic cooling equipment in the first time period through the first model. The first operating parameters represent the parameters that affect the power generation of the photovoltaic cooling equipment.

[0123] The second control module 32 is used to input the second operating parameters of the photovoltaic cooling equipment into the trained second model, so as to output the predicted value of the cooling demand of the target object in the first time period through the second model. The second operating parameters represent the parameters that affect the cooling demand of the target object.

[0124] The third control module 33 is used to control the working mode of the photovoltaic cooling equipment during the first time period based on the predicted value of photovoltaic power generation and the predicted value of cooling demand.

[0125] In one possible implementation, the first control module is specifically used to acquire the real-time light intensity of the photovoltaic cooling device as light intensity information, acquire the historical power generation sequence of the photovoltaic cooling device per hour in the second time period as historical power generation information, acquire the seasonal solar radiation coefficient of the current season, and acquire the equipment aging coefficient of the photovoltaic cooling device.

[0126] The light intensity information, the historical power generation information, the seasonal sunshine coefficient, and the equipment aging coefficient are used as the first operating parameters and input into the first model.

[0127] In one possible implementation, the second control module is specifically used to acquire the usage time of the photovoltaic cooling equipment in the third time period as historical usage information, acquire the current weather and future weather as weather information, and acquire the temperature difference information between indoors and outdoors.

[0128] The historical usage information, the weather information, and the temperature difference information are input into the second model as the second operating parameters.

[0129] In one possible implementation, the third control module is specifically used to control the photovoltaic cooling equipment to switch its operating mode to pure photovoltaic drive mode when the predicted value of photovoltaic power generation is greater than the product of the predicted value of cooling demand and the first coefficient.

[0130] Alternatively, when the predicted value of photovoltaic power generation is less than the product of the predicted value of cooling demand and the second coefficient, and the current cooling capacity of the photovoltaic cooling equipment is greater than or equal to the product of the predicted value of cooling demand and the third coefficient, the operating mode of the photovoltaic cooling equipment is switched to the phase change material driven mode.

[0131] Alternatively, when the predicted value of photovoltaic power generation is less than the product of the predicted value of cooling demand and the fourth coefficient, and the current cooling capacity of the photovoltaic cooling equipment is less than the product of the predicted value of cooling demand and the second coefficient, the operating mode of the photovoltaic cooling equipment is switched to a hybrid driving mode of photovoltaic and phase change materials, wherein the first coefficient is greater than the third coefficient, the second coefficient is greater than the fourth coefficient.

[0132] In one possible implementation, the third control module is further configured to calculate the supply-demand difference of the photovoltaic cooling device using the following formula when the photovoltaic cooling device operates in a hybrid drive mode:

[0133] Supply-demand difference = Forecasted demand for cooling capacity - (Current cooling capacity of the photovoltaic cooling equipment + Forecasted photovoltaic power generation × Fifth coefficient × Energy efficiency ratio of the photovoltaic cooling equipment);

[0134] When the supply-demand difference is greater than the first threshold, the operating mode of the photovoltaic cooling equipment is switched to photovoltaic, phase change material and mains power drive mode.

[0135] In one possible implementation, the third control module is further configured to control the phase change material of the photovoltaic cooling equipment to store cold when the predicted value of photovoltaic power generation is greater than the product of the predicted value of cooling demand and the sixth coefficient, and the predicted value of cooling demand in the fourth time period in the future shows an upward trend, wherein the sixth coefficient is greater than the first coefficient.

[0136] In one possible implementation, the third control module is further configured to adjust the refrigerant flow rate of the photovoltaic cooling device according to the cold storage efficiency of the phase change material, so that the cold storage efficiency is greater than a second threshold.

[0137] The photovoltaic cooling equipment control device provided in this embodiment can be as follows: Figure 3 The apparatus shown can perform, as Figure 1-2 All steps of the control method for photovoltaic cooling equipment, thereby achieving Figure 1-2 For details on the technical effects of the photovoltaic cooling equipment control method shown, please refer to [link / reference]. Figure 1-2 The relevant descriptions are presented concisely and will not be elaborated upon here.

[0138] Figure 4 This is a schematic diagram of the structure of a photovoltaic cooling device provided in an embodiment of the present invention. Figure 4The photovoltaic cooling device 400 shown includes at least one processor 401, a memory 402, at least one network interface 404, and other user interfaces 403. The various components in the photovoltaic cooling device 400 are coupled together via a bus system 405. It is understood that the bus system 405 is used to enable communication between these components. In addition to a data bus, the bus system 405 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 4 The general designated all buses as Bus System 405.

[0139] The user interface 403 may include a display, keyboard, or clicking device (e.g., mouse, trackball, touchpad, or touchscreen).

[0140] It is understood that the memory 402 in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate Synchronous DRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The memory 402 described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0141] In some implementations, memory 402 stores elements, executable units or data structures, or subsets thereof, or extended sets thereof: operating system 4021 and application program 4022.

[0142] The operating system 4021 includes various system programs, such as the framework layer, core library layer, and driver layer, used to implement various basic business functions and handle hardware-based tasks. The application program 4022 includes various applications, such as a media player and a browser, used to implement various application functions. The program implementing the method of this embodiment can be included in the application program 4022.

[0143] In this embodiment of the invention, by calling the program or instructions stored in the memory 402, specifically the program or instructions stored in the application program 4022, the processor 401 executes the method steps provided in each method embodiment, including, for example:

[0144] The first operating parameters of the photovoltaic cooling equipment are input into the trained first model so that the photovoltaic power generation of the photovoltaic cooling equipment in the first time period is output through the first model. The first operating parameters represent the parameters that affect the power generation of the photovoltaic cooling equipment.

[0145] The second operating parameters of the photovoltaic cooling equipment are input into the trained second model so that the second model can output the predicted cooling demand of the target object in the first time period. The second operating parameters represent the parameters that affect the cooling demand of the target object.

[0146] The operating mode of the photovoltaic cooling equipment is controlled based on the predicted photovoltaic power generation and the predicted cooling demand during the first time period.

[0147] The methods disclosed in the above embodiments of the present invention can be applied to processor 401, or implemented by processor 401. Processor 401 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 401 or by instructions in the form of software. The processor 401 may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software units in the decoding processor. The software units may be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 402. Processor 401 reads the information in memory 402 and, in conjunction with its hardware, completes the steps of the above method.

[0148] It is understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described herein, or combinations thereof.

[0149] For software implementation, the techniques described herein can be implemented by units that perform the functions described herein. The software code can be stored in memory and executed by a processor. The memory can be implemented in the processor or external to the processor.

[0150] The photovoltaic cooling equipment provided in this embodiment can be as follows: Figure 4 The device shown can perform, for example Figure 1-2 All steps of the control method for photovoltaic cooling equipment, thereby achieving Figure 1-2 For details on the technical effects of the photovoltaic cooling equipment control method shown, please refer to [link / reference]. Figure 1-2 The relevant descriptions are presented concisely and will not be elaborated upon here.

[0151] This invention also provides a storage medium (computer-readable storage medium). This storage medium stores one or more programs. The storage medium may include volatile memory, such as random access memory; it may also include non-volatile memory, such as read-only memory, flash memory, hard disk, or solid-state drive; and it may also include combinations of the above types of memory.

[0152] One or more programs in the storage medium can be executed by one or more processors to implement the photovoltaic cooling equipment control method described above, which is executed on the device side.

[0153] The processor is used to execute a photovoltaic cooling device control program stored in the memory to implement the following steps of the photovoltaic cooling device control method executed on the device side:

[0154] The first operating parameters of the photovoltaic cooling equipment are input into the trained first model so that the photovoltaic power generation of the photovoltaic cooling equipment in the first time period is output through the first model. The first operating parameters represent the parameters that affect the power generation of the photovoltaic cooling equipment.

[0155] The second operating parameters of the photovoltaic cooling equipment are input into the trained second model so that the second model can output the predicted cooling demand of the target object in the first time period. The second operating parameters represent the parameters that affect the cooling demand of the target object.

[0156] The operating mode of the photovoltaic cooling equipment is controlled based on the predicted photovoltaic power generation and the predicted cooling demand during the first time period.

[0157] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0158] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0159] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A photovoltaic refrigeration device control method, characterized by, The application comprises the following steps: inputting first operation parameters of the photovoltaic refrigeration device into a trained first model to output a photovoltaic power generation amount prediction value of the photovoltaic refrigeration device in a first time period through the first model, wherein the first operation parameters represent parameters affecting the power generation amount of the photovoltaic refrigeration device; inputting second operation parameters of the photovoltaic refrigeration device into a trained second model to output a demand cold amount prediction value of the target object in the first time period through the second model, wherein the second operation parameters represent parameters affecting the demand cold amount of the target object; controlling the working mode of the photovoltaic refrigeration device in the first time period according to the photovoltaic power generation amount prediction value and the demand cold amount prediction value; The step of inputting the first operation parameters of the photovoltaic refrigeration device into the trained first model comprises the following steps: obtaining real-time illumination intensity of the photovoltaic refrigeration device as illumination intensity information, obtaining a historical power generation amount sequence of the photovoltaic refrigeration device in each hour in a second time period as historical power generation information, obtaining a seasonal daylight coefficient of a current season, and obtaining a device aging coefficient of the photovoltaic refrigeration device; inputting the illumination intensity information, the historical power generation information, the seasonal daylight coefficient, and the device aging coefficient into the first model as the first operation parameters; when the working mode of the photovoltaic refrigeration device is a hybrid driving mode, calculating a supply-demand difference value of the photovoltaic refrigeration device through the following formula: supply-demand difference value = demand cold amount prediction value - (current cold storage amount of the photovoltaic refrigeration device + photovoltaic power generation amount prediction value * fifth coefficient * energy efficiency ratio of the photovoltaic refrigeration device), wherein the fifth coefficient is a coefficient for correcting the energy conversion efficiency of the photovoltaic power generation amount prediction value to the actual energy available for refrigeration; when the supply-demand difference value is greater than a first threshold value, controlling the working mode of the photovoltaic refrigeration device to switch to a photovoltaic, phase change material, and mains driving mode; when any of the following conditions is met, automatically returning to the hybrid driving mode: the supply-demand difference value is less than or equal to the first threshold value for two consecutive times, there is no demand cold amount surge trend, the photovoltaic power generation amount prediction value rises, and the increase of the photovoltaic power generation amount prediction value * fifth coefficient * energy efficiency ratio is greater than or equal to the supply-demand difference value.

2. The method of claim 1, wherein, The step of inputting the second operation parameters of the photovoltaic refrigeration device into the trained second model comprises the following steps: obtaining a use time period of the photovoltaic refrigeration device in a third time period as historical use information, obtaining current weather and future weather as weather information, and obtaining temperature difference information between indoor and outdoor; inputting the historical use information, the weather information, and the temperature difference information into the second model as the second operation parameters.

3. The method of claim 1, wherein, The step of controlling the working mode of the photovoltaic refrigeration device in the first time period according to the photovoltaic power generation amount prediction value and the demand cold amount prediction value comprises the following steps: when the photovoltaic power generation amount prediction value is greater than the product of the demand cold amount prediction value and a first coefficient, controlling the working mode of the photovoltaic refrigeration device to switch to a pure photovoltaic driving mode. or, when the photovoltaic power generation amount prediction value is less than the product of the demand cold amount prediction value and a second coefficient, and the current cold storage amount of the photovoltaic refrigeration device is greater than or equal to the product of the demand cold amount prediction value and a third coefficient, controlling the working mode of the photovoltaic refrigeration device to switch to a phase change material driving mode; or, when the photovoltaic power generation amount prediction value is less than the product of the demand cold amount prediction value and a fourth coefficient, and the current cold storage amount of the photovoltaic refrigeration device is less than the product of the demand cold amount prediction value and the second coefficient, controlling the working mode of the photovoltaic refrigeration device to switch to a photovoltaic and phase change material hybrid driving mode, the first coefficient being greater than the third coefficient, the third coefficient being greater than the second coefficient, and the second coefficient being greater than the fourth coefficient.

4. The method of claim 3, wherein, The method further comprises: when the photovoltaic power generation amount prediction value is greater than the product of the demand cold amount prediction value and a sixth coefficient, and the demand cold amount prediction value in a future fourth time period shows an upward trend, controlling the phase change material of the photovoltaic refrigeration device to store cold, the sixth coefficient being greater than the first coefficient.

5. The method of claim 4, wherein, The method further comprises: adjusting the refrigerant flow rate of the photovoltaic refrigeration device according to the cold storage efficiency of the phase change material, so that the cold storage efficiency is greater than a second threshold.

6. A photovoltaic refrigeration apparatus control device characterized by comprising: comprises: a first control module configured to input first operation parameters of the photovoltaic refrigeration device into a trained first model to output, by the first model, a photovoltaic power generation amount prediction value of the photovoltaic refrigeration device in a first time period, the first operation parameters representing parameters affecting the power generation amount of the photovoltaic refrigeration device; a second control module configured to input second operation parameters of the photovoltaic refrigeration device into a trained second model to output, by the second model, a demand cold amount prediction value of a target object in the first time period, the second operation parameters representing parameters affecting the demand cold amount of the target object; a third control module configured to control the working mode of the photovoltaic refrigeration device in the first time period according to the photovoltaic power generation amount prediction value and the demand cold amount prediction value; the first control module is specifically configured to obtain real-time illumination intensity of the photovoltaic refrigeration device as illumination intensity information, obtain a historical power generation sequence of the photovoltaic refrigeration device in a second time period per hour as historical power generation information, obtain a seasonal daylight coefficient of a current season, and obtain a device aging coefficient of the photovoltaic refrigeration device; input the illumination intensity information, the historical power generation information, the seasonal daylight coefficient, and the device aging coefficient into the first model as the first operation parameters; the third control module is further configured to, when the working mode of the photovoltaic refrigeration device is a hybrid driving mode, calculate a supply-demand difference value of the photovoltaic refrigeration device by the following formula: supply-demand difference value = demand cold amount prediction value - (current cold storage amount of the photovoltaic refrigeration device + photovoltaic power generation amount prediction value × fifth coefficient × energy efficiency ratio of the photovoltaic refrigeration device), the fifth coefficient being a coefficient for correcting the energy conversion efficiency of the photovoltaic power generation amount prediction value to the actual energy available for refrigeration. when the supply-demand difference is greater than a first threshold value, the working mode of the photovoltaic refrigeration device is switched to a photovoltaic, phase change material and mains power driven mode; when any one of the following conditions is met, the hybrid driven mode is automatically restored: the supply-demand difference is less than or equal to the first threshold value for two consecutive times, and there is no sudden increase trend of the required cooling capacity, the photovoltaic power generation capacity prediction value rises, and the increase of the photovoltaic power generation capacity prediction value multiplied by the fifth coefficient multiplied by the energy efficiency ratio is greater than or equal to the supply-demand difference.

7. A photovoltaic refrigeration apparatus, characterized by, comprising: a processor and a memory, the processor being configured to execute a photovoltaic refrigeration device control program stored in the memory to implement the photovoltaic refrigeration device control method of any one of claims 1-5.

8. A storage medium, characterized by The storage medium stores one or more programs, which can be executed by one or more processors to implement the photovoltaic refrigeration device control method of any one of claims 1-5.

Images