Photovoltaic energy storage and cooling coupled system and method based on copper-based composite phase change material

Abstract

The copper-based composite phase change material photovoltaic energy storage and cooling coupled system and method covers power generation, cooling energy storage, energy conversion and other modules. During the day, the solar photovoltaic panel generates heat, the copper-based composite phase change material absorbs heat and stores energy and cools the photovoltaic panel; at night, the phase change material releases heat, and the thermoelectric material converts thermal energy into electrical energy. The system uses the DDPG algorithm of deep reinforcement learning to optimize operation. The state space includes solar photovoltaic panel temperature, phase change material temperature and other parameters, and the action space involves controlling cooling energy storage, energy conversion and other operations. The reward function is designed from multiple aspects such as efficiency, energy storage, power output and stability. By constructing a policy network and a value network, the system can make decisions according to environmental changes. Data is extracted from the experience replay buffer to update network parameters, allowing the system to adaptively adjust. The system not only solves the problem of high temperature of photovoltaic panels and night power generation, but also uses algorithms to improve energy utilization efficiency, which is far superior to traditional photovoltaic systems in performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of photovoltaic energy storage and cooling, in particular to a photovoltaic energy storage and cooling coupling system and method based on copper-based composite phase change material. BACKGROUND

[0002] Under the background of global warming, photovoltaic power generation is increasingly widely used. However, the photoelectric conversion efficiency of photovoltaic panels is not fixed during the process of converting solar energy into electrical energy. When the photovoltaic panel is irradiated, part of the solar energy will be converted into electrical energy, and another part will inevitably be converted into heat energy, thereby causing the temperature of the photovoltaic panel to rise. Related studies have shown that under normal circumstances, the photoelectric conversion efficiency of the photovoltaic panel will decrease by about 4-5% for every 10℃ rise in temperature. Moreover, photovoltaic power generation has obvious intermittent characteristics, and its power generation efficiency is greatly affected by weather conditions. In rainy days and at night, the efficiency of photovoltaic power generation is very low, and effective power generation is almost impossible, so it needs to be matched with a high-efficiency energy storage system to achieve a balance between supply and demand of energy, thereby improving the reliability and stability of energy supply.

[0003] Phase change materials have such a characteristic that they can undergo phase change within a certain temperature range, and in the process of phase change, they will absorb or release a large amount of latent heat, while their own temperature will remain relatively stable. This unique performance makes phase change materials have potential application prospects in the field of photovoltaic panel cooling and energy storage. In the early research process, traditional phase change materials such as paraffin and fatty acids have been tried to be applied in photovoltaic systems. However, these traditional phase change materials have many problems, such as poor thermal conductivity and unsatisfactory thermal stability.

[0004] With the continuous development of material science, composite phase change materials have become a hot topic in current research. Compared with traditional aluminum-based phase change materials with a thermal conductivity of about 200W / (m·K), copper-based materials have a thermal conductivity of up to 400W / (m·K), so copper-based materials can conduct the heat generated by photovoltaic panels faster, thereby effectively reducing the temperature of photovoltaic panels and reducing the efficiency loss caused by high temperature. At the same time, the metal support adopts a three-period minimal surface structure. The unique feature of this structure is that the three-period minimal surface will exhibit periodic variation in three independent dimensions, and the crystal cell structures are interconnected, forming continuous channels inside, with super-high specific strength and specific stiffness, good energy absorption capacity and anisotropic mechanical properties, which endow the material with better thermal conductivity and thermal stability, so that it can maintain good performance in repeated heat absorption and release process, and is more suitable for use in photovoltaic energy storage and cooling coupling system.

[0005] The differences compared with the prior art are as follows:

[0006] Comparison with the technology of patent CN116120899A "A porous copper-based composite phase change heat storage material and its preparation method"

[0007] Material structure and composition: The porous copper-based composite phase change heat storage material prepared by patent CN116120899A is composed of a porous copper heat-conducting frame and a phase change material (such as paraffin, sodium acetate trihydrate, etc.), where the porous copper frame is prepared by copper powder, salt crystals and chlorinated salt, and the salt is used as a pore-forming agent to form a pore structure. The copper-based composite phase change material of this patent is composed of a phase change material and a metal bracket (aluminum alloy) with a three-period minimal surface structure. The unique three-period minimal surface structure makes the material have ultra-high specific strength, specific stiffness and good energy absorption capacity, etc. There are obvious differences in material structure and composition between the two.

[0008] Preparation method: The preparation method of CN116120899A includes preparing copper salt mixed tablets, porous copper heat-conducting frame, and then immersing the porous copper frame into liquid phase change material. It involves multiple steps such as dispersion of copper powder in organic solvent, mixing with salt crystals and chlorinated salt. The preparation method of the copper-based composite phase change material of this patent is to first prepare copper-based powder by mechanical crushing method, then use electroplating method to improve the compatibility with phase change material, and then mix, pour, solidify, heat treatment and processing, etc. The preparation process and specific operation are different from the former.

[0009] Application scenarios and function implementation: CN116120899A is mainly applied to the field of heat storage and heat conduction, aiming to solve the problems of traditional materials in the process of heat energy storage and utilization, and to realize efficient heat storage and heat conduction function. The photovoltaic energy storage and photovoltaic panel cooling coupled system based on copper-based composite phase change material of this patent is applied to the field of photovoltaic, focusing on solving the problem of low power caused by high temperature of solar photovoltaic panel and inability to generate electricity at night or in rainy weather. Through the phase change of the material, functions such as photovoltaic panel cooling, daytime waste heat storage and night heat energy and electric energy conversion are realized. The application scenarios and functions implemented are fundamentally different from the former.

[0010] Performance improvement focus: CN116120899A focuses on improving the heat storage and heat conduction performance of the material by increasing the porosity of the material, using the rough surface of the copper sheet to limit the leakage of the phase change material and improving the light-heat conversion efficiency. In addition to using the high thermal conductivity of the copper-based composite phase change material to improve the heat dissipation effect, this patent also relies on the three-period minimal surface structure to improve the thermal stability of the material, and focuses more on improving the power generation efficiency of the entire photovoltaic system, realizing efficient use of energy and enhancing the adaptability of the system in different environments. The performance improvement focus of the two is different.

[0011] Algorithm application: the patent CN116120899A does not mention the algorithm related content. This patent innovatively introduces the DDPG algorithm in deep reinforcement learning, defines the state space of each parameter, and includes the action space of the operation of the phase change material cooling and energy storage module, the thermoelectric material energy conversion module and the electric energy transmission module, designs the reward function considering efficiency, energy storage, electric energy output and stability, realizes the dynamic adjustment of the control strategy of the system according to the environment, improves the flexibility and adaptability of the system, and optimizes the performance of the system, which is the unique technical advantage of this patent distinguishing from CN116120899A. SUMMARY

[0012] In view of the above problems, the present application provides a copper-based composite phase change material photovoltaic energy storage and cooling coupling system and method, which aims to provide a system that is different from traditional photovoltaic systems, which couples photovoltaic energy storage and photovoltaic panel cooling systems inside the system, reducing the loss of solar photovoltaic panel power generation efficiency caused by high temperature. At the same time, the addition of thermoelectric materials and small electric energy collection and inversion systems allows the waste heat of the traditional photovoltaic water cooling system to continue to generate electricity at night, playing a role in converting thermal energy into electrical energy, improving the power generation efficiency of solar photovoltaic panels, slowing down the aging of solar photovoltaic panels, and playing a role in reducing the overall power generation of the system. A system and method for preparing some materials based on copper-based composite phase change material photovoltaic energy storage and photovoltaic panel cooling coupling.

[0013] To achieve the above purpose, the technical solution adopted by the present application is:

[0014] The copper-based composite phase change material photovoltaic energy storage and cooling coupling system comprises a cooling and energy storage module of copper-based composite phase change material and aluminum alloy support; a power generation module of solar photovoltaic panel installed at the top; an energy conversion module of thermoelectric material installed below, and a support module composed of stainless steel support at the bottom, an electric energy transmission module composed of energy collection and inversion system module and external lead wire, and the solar panel and copper-based composite phase change material, aluminum alloy support and thermoelectric material are closely arranged from top to bottom. The energy collection and inversion system is composed of electric energy collection module, control unit and inversion circuit.

[0015] As a further improvement of the system of the present application, the thermoelectric material used for the conversion of thermal energy and electrical energy is silicon-germanium alloy.

[0016] As a further improvement of the system of the present application, the preparation method of the copper-based composite phase change material comprises the following steps:

[0017] Step 1, prepare copper-based powder by mechanical crushing method;

[0018] Step 2, surface treatment of copper-based powder by electroplating method to improve its compatibility with phase change material;

[0019] Step 3, mix the surface treated copper-based powder with phase change material, etc., and the mass fraction of the copper-based powder is preferably 10%-30%;

[0020] Step 4, pour the mixed material into a mold, and solidify under certain temperature and pressure; for paraffin with a melting point of 50-60℃, liquid paraffin has better flowability, and the pouring temperature can be controlled at 70-80℃; the pouring pressure is controlled at 0.1-0.5MPa;

[0021] Step 5, heat treat the solidified material to improve its phase change performance and stability; the heat treatment temperature and time are determined according to the composition and performance requirements of the material, and the paraffin is fully melted and better fused with the copper-based powder at a temperature of 100-150℃, and the heat treatment time is about 2-6 hours;

[0022] Step 6, process the heat treated material, such as cutting, grinding, polishing, etc., to obtain the required shape and size.

[0023] The present application provides a control method for a photovoltaic energy storage and cooling coupled system based on a copper-based composite phase change material, and the specific steps are as follows:

[0024] First, define the state space, action space and reward function to facilitate the implementation of the algorithm:

[0025] State space:

[0026] The state space covers a plurality of parameters that have an important influence on the performance of the system, including but not limited to the temperature of the solar panel T solar , the temperature of the copper-based composite phase change material T pcm , the phase change degree of the phase change material α pcm , which is estimated by a temperature sensor and a material property formula, the temperature difference ΔT th ermo between the upper and lower surfaces of the thermoelectric material, the current energy storage capacity E storage , the real-time electricity price P grid of the power grid, and the light intensity I ligh t These parameters together constitute the state vector,

[0027] s=[T solar ,T pcm ,α pcm ,ΔT th ermo ,E storage ,P grid ,I ligh t ], (1)

[0029] Comprehensively reflects the performance of the system at a certain moment;

[0030] Action space:

[0031] The action space defines the set of operations that the system can take, including operations to control the phase change material cooling and energy storage module, such as adjusting the working intensity u of the heat dissipation device cool , which ranges from operations to control the thermoelectric material energy conversion module, such as adjusting the heat conduction efficiency η between the thermoelectric material and the phase change material, the heat dissipation environment th ermo , ranging from and the operation of the electric energy conduction module, determining the electric energy output power P cool , ranging from and the output timing, so the action vector can be represented as

[0032] [u cool ,η th ermo ,P cool ], (2)

[0034] Reward function:

[0035] The total reward includes four sub-rewards: efficiency, energy storage, electric energy output, and stability;

[0036] Efficiency reward is:

[0037] R efficiency = k1 × (Δη solar ), (3)

[0038] where Δη solar represents the change in photoelectric conversion efficiency, Δη solar > 0 when the effective work of the phase change material leads to an increase in photoelectric conversion efficiency, and k1 is the corresponding weight coefficient.

[0039] Energy storage reward is:

[0040] R storage = k2 × (E storage (t) - E storage (t-1)) - k3 × E loss , (4)

[0041] where E storage (t) and E storage (t-1) represent the current and previous energy storage capacity, respectively, and E loss represents the loss in the energy storage process, and k2 and k3 are weight coefficients. When the energy storage effect is good, R storage > 0;

[0042] Electric energy output reward is:

[0043] R output = k4 × P outX P grid -k5 X |AP out |, (5)

[0044] where k4 and k5 are weight coefficients, this function takes into account the economic benefits of output power and the stability of output power, and when the electricity price is high, output power and stable power will obtain higher rewards;

[0045] The stability reward is

[0046]

[0047] where x i (t) represents the key parameters of the system, n represents the number of parameters, the reward aims to keep the system parameters stable, and k6 is the weight coefficient.

[0048] Thus the total reward function can be calculated as

[0049] R stability = R efficiency + R storage + R output + R stability ,(7)

[0050] Then, the specific implementation algorithm includes the following steps:

[0051] Step one, algorithm initialization:

[0052] First, construct the policy network π(s|θ π ) and the value network Q(s,a|θ Q ), the task of the policy network is to generate actions according to the input state information, while the value network is to evaluate the value of the current state-action pair, these networks include input layer, multiple hidden layers and output layer, the input layer receives the state vector, and the output layer outputs the corresponding action or value evaluation result;

[0053] Then set up the parameters θ π and θ Q of the initialization network, these parameters are randomly generated, but it is necessary to ensure that the output range of the network meets the range requirements of the action space and the value evaluation;

[0054] At the same time, other parameters are initialized, the experience replay buffer D is created, its capacity is N, which is used to store the experience tuples in the process of interaction between the agent and the environment:

[0055] (s,a,R,s′),(8)

[0056] This helps to break the correlation between samples and improve the stability of learning;

[0057] Step two, interact with the algorithm and the environment and execute actions:

[0058] The sensor delivers the collected system state information to the agent, forming a state vector s t . The agent inputs s t into the policy network to obtain an action vector:

[0059] a t = π(s t | θ π ), (9)

[0060] The action vector contains control instructions for different modules of the system, including adjustments to the working intensity of the heat dissipation device u cool , adjustments to the thermal conductivity efficiency of the thermoelectric material η th ermo , and control of the electrical power output P out . These actions will be converted into specific control signals and sent to the corresponding actuators.

[0061] After the actions are executed, the system state will change, and the new state s t+1 will be collected by the sensor. At the same time, the system will calculate the corresponding reward based on the new state. If the photoelectric conversion efficiency of the system improves, a positive efficiency reward R t will be calculated according to the corresponding formula. If the energy storage capacity increases and the energy loss is small, a positive energy storage reward will be obtained.

[0062] Step three, learning and updating using algorithms:

[0063] First, randomly select a batch of experience tuples from the experience replay buffer D, denoted as (s i , a i , R i , s i+1 ).

[0064] Calculate the target value:

[0065] For each experience tuple, calculate the target value according to the Bellman equation:

[0066] y i = R i + γQ(s i+1 , π(s i+1 | θ π )| θ Q ), (10)

[0067] Based on the current reward and the expected value of the next state, estimate the target value of the current state-action pair;

[0068] Then update the value network by minimizing the temporal difference error, i.e.

[0069]

[0070] to update the parameters of the value network, which enables the value network to more accurately evaluate the value of state-action pairs.

[0071] Finally, the policy network is updated according to the deterministic policy gradient algorithm, that is,

[0072]

[0073] The action generated by the policy network can obtain a higher cumulative reward in the long term.

[0074] Advantages: Compared with the prior art, the present application has the following advantages:

[0075] Solve the problem of too cold or too hot temperature of solar photovoltaic panel;

[0076] With the high thermal conductivity of copper-based composite phase change material, it can quickly conduct the heat generated by the photovoltaic panel, thereby effectively reducing the temperature of the photovoltaic panel, and reducing the loss of photoelectric conversion efficiency caused by high temperature. In northern China, frost and freezing may occur at night in winter, and copper-based composite phase change material can release a large amount of heat accumulated during the day to prevent solar photovoltaic panels from being damaged by freezing.

[0077] Realize waste heat utilization and night power generation;

[0078] The traditional solar cooling system cannot fully utilize the waste heat, but in this patent, from the evening to the night, the copper-based composite phase change material releases heat when it solidifies, which is transferred to the silicon-germanium alloy thermoelectric material. The temperature difference between the upper and lower parts of the thermoelectric material makes the electrons move and generate potential difference, converting heat energy into electric energy. At the same time, the energy collection and inversion system can efficiently capture weak electric energy and accurately invert the collected electric energy into alternating current. The system realizes the storage of excess heat during the day and power generation at night, improves the power generation efficiency of the system, and plays a role in peak shaving.

[0079] Improve the performance of the material;

[0080] The metal support in the copper-based composite phase change material adopts a three-period minimal surface structure, which has super-high specific strength and specific stiffness, good energy absorption capacity and anisotropic mechanical properties, endows the material with better thermal conductivity and thermal stability, so that it can maintain good performance in repeated heat absorption and release process, and is more suitable for photovoltaic energy storage and cooling coupled system than traditional phase change materials (such as pure paraffin, fatty acid and the like with poor thermal conductivity and insufficient thermal stability). At the same time, the copper-based phase change material has higher thermal conductivity than the traditional aluminum-based phase change material, about 380-400 W / (m·K), which can conduct the heat generated by the photovoltaic panel more quickly, thereby effectively reducing the temperature of the photovoltaic panel and reducing the efficiency loss caused by high temperature.

[0081] The structure is reasonably designed;

[0082] In the overall design of the system, the solar photovoltaic panel, the copper-based composite phase change material, the aluminum alloy support and the thermoelectric material are in close contact from top to bottom, and the layout of each module is reasonable, which is conducive to efficient heat transfer and energy conversion. At the same time, the aluminum alloy support supports the copper-based composite phase change material to form a cooling and energy storage module, and the stainless steel support forms a support module, which is stable and can meet different functional requirements.

[0083] Algorithm innovation;

[0084] Traditional systems mostly use fixed rule control, which has poor adaptability, processing capacity and optimizability, while the DDPG algorithm introduced in the patent can dynamically adjust the control strategy according to the temperature of the solar photovoltaic panel, the phase change degree, the electricity price and other key parameters, such as adjusting the heat dissipation intensity and the electric energy output power, adapting to complex environment, improving the flexibility and adaptability of the system, and avoiding the limitations of traditional control. Through the well-designed reward function, the system performance is optimized from multiple dimensions such as improving energy conversion efficiency, optimizing electric energy storage process, realizing economic benefit maximization of electric energy output and ensuring system stability. BRIEF DESCRIPTION OF DRAWINGS

[0085] Figure 1 It is a system diagram of the photovoltaic energy storage and photovoltaic panel cooling coupled system of the application;

[0086] Figure 2 It is a flowchart of the energy collection and inversion system;

[0087] Figure 3 It is a phase change material of the three-period minimal surface structure of copper base;

[0088] Figure 4 It is a flowchart of algorithm implementation;

[0089] Attachment mark;

[0090] 1, solar photovoltaic panel; 2, copper-based composite phase change material; 3, aluminum alloy support; 4, thermoelectric material; 5, wire; 6, support; 7, energy collection and inversion system. DETAILED DESCRIPTION

[0091] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. The following embodiments are used to illustrate the present application but not to limit the scope of the present application.

[0092] Structurally, as Figure 1 , first, the solar photovoltaic panel 1 (using single silicon crystal material) is installed at the top as a power generation module to receive solar energy and convert it into electric energy. Then, the four corners of the solar photovoltaic panel 1 are fixed on the top of the aluminum alloy support 3 by detachable screws, so that the solar photovoltaic panel can be detached and maintained conveniently, and at the same time, a gap of about 10 cm is left between the solar photovoltaic panel 1 and the bottom plate of the aluminum alloy support 3, and a cooling and energy storage module composed of the copper-based composite phase change material 2 supported by the aluminum alloy support 3 is installed closely below the solar photovoltaic panel 1, as shown in the microstructure Figure 2 , the phase change material used in the copper-based composite phase change material 2 uses paraffin or fatty acid or inorganic compound, etc., and the copper-based composite phase change material 2 is prepared by a specific preparation method (such as preparing copper-based powder by atomization method, mixing with phase change material after electroplating treatment, and through steps of pouring, heat treatment, processing, etc.), and the aluminum alloy support 3 plays a role of stable support to ensure that the copper-based composite phase change material 2 is in close contact with the solar photovoltaic panel to effectively absorb the heat generated by the solar photovoltaic panel.

[0093] Then, the thermoelectric material 4 (using tellurium bismuth alloy) is installed below the copper-based composite phase change material 2 as an energy conversion module to convert heat into electric energy. The upper part of the thermoelectric material 4 is tightly attached to the aluminum alloy support 3 by brazing to maintain good heat conduction relationship, so that an effective temperature difference is formed between the upper and lower parts of the thermoelectric material 4, wires 5 are added on both sides of the thermoelectric material 4, and an energy collection and inversion system 7 is applied at both ends of the wires 5. Figure 3The electric energy collection module comprises high-sensitivity current transformers, voltage transformers and other sensors. The current transformer can convert a tiny AC current signal into a voltage signal that can be processed by the collection circuit. The voltage transformer collects and converts a tiny voltage. The control unit comprises multiple microcontrollers or digital signal processors, receives digital signals from the collection module, analyzes data according to a set program algorithm, generates control pulse signals according to the data, controls the working state of the inverter circuit, ensures stable and efficient operation of the entire system, and comprises power switches, drive circuits, filter circuits and the like. Under the driving of the pulse signal of the control unit, the power switch works in a high-frequency switching state to convert DC power into AC power, which can capture weak electric energy and convert it into 220V / 50Hz AC power directly connected to the power grid. The lowermost support module composed of a stainless steel support 6 provides a stable support structure for the entire system, ensuring that the system can be placed stably under various terrain conditions.

[0094] Overall, the solar photovoltaic panel 1, the copper-based composite phase change material 2, the aluminum alloy support 3 and the thermoelectric material 4 are closely arranged from top to bottom, which is conducive to efficient heat transfer and effective energy conversion. The close cooperation between the modules ensures the stability and efficiency of the system, for example, heat can be quickly transferred from the solar photovoltaic panel 1 to the copper-based composite phase change material 2, and at night it can be smoothly transferred from the phase change material to the thermoelectric material 4.

[0095] Working principle:

[0096] During the day, as the daytime light intensity becomes stronger, the solar photovoltaic panel 1 begins to work, converting solar energy into electrical energy while itself heating up due to the absorption of solar energy, with the temperature rising. The heat generated by the solar photovoltaic panel 1 is quickly transferred to the copper-based composite phase change material 2 cooling and energy storage module below. Since the phase change materials such as paraffin and fatty acid are encapsulated in the pores of the copper base, copper has good thermal conductivity, which can quickly transfer the external heat to the phase change materials such as paraffin and fatty acid, making the phase change materials such as paraffin and fatty acid change phase faster and store energy by absorbing heat. At the same time, after the paraffin melts, the pore structure of the copper base can also prevent the leakage of paraffin, ensuring the stability of the composite phase change material. Due to the high thermal conductivity of the copper-based composite phase change material 2, heat can be quickly conducted in the material. As the heat is absorbed, the phase change materials such as paraffin or fatty acid in the copper-based composite phase change material 2 gradually melt, absorbing a large amount of heat in the process, thereby cooling the solar photovoltaic panel 1, reducing its temperature, and reducing the loss of photoelectric conversion efficiency caused by high temperature. At the same time, the liquid phase change materials such as paraffin or fatty acid can carry a large amount of heat, and their metal supports ensure the high thermal conductivity and stability of the cooling and energy storage module during the heat absorption process, which can quickly transfer heat to the entire phase change material such as paraffin or fatty acid and prevent the material from becoming unstable due to temperature changes.

[0097] In the evening to night, the light intensity decreases to zero, and the solar photovoltaic panel 1 no longer generates new heat, and the temperature of the entire system gradually decreases. The copper-based composite phase change material 2 begins to solidify, releasing a large amount of heat during the solidification process. The released heat is transferred to the thermoelectric material 4 below, and the thermoelectric material utilizes the temperature difference between the upper and lower parts to make a large number of electrons move up and down in the material. By connecting wires to the upper and lower parts of the thermoelectric material 4, the directional movement of electrons produces a potential difference, which in turn converts heat energy into electrical energy. Using this potential difference and weak current, the energy collection and inversion system 7 can capture weak electrical energy and accurately invert it into alternating current. This realizes the storage of excess heat during the day and the conversion of heat energy into electrical energy at night, improves the overall power generation of the system, plays a role in peak shaving, and makes the system effectively utilize energy at different times, improving energy utilization efficiency.

[0098] Specific implementation of the algorithm:

[0099] Connect the solar photovoltaic panel to the maximum power point tracking controller through wires to ensure that the solar photovoltaic panel can output maximum power under different light conditions. The maximum power point tracking controller is responsible for adjusting the working voltage and current of the solar photovoltaic panel to match the load and improve the photoelectric conversion efficiency.

[0100] The phase change material container is arranged on the back or near the solar photovoltaic panel to effectively absorb the excess heat generated by the solar photovoltaic panel. Heat dissipation devices such as heat pipes or heat sinks can be installed around the phase change material container to assist heat transfer and regulation.

[0101] In each system of the device, various sensors are installed, including temperature sensors: multiple temperature sensors such as thermocouples or thermistors are arranged on the surface of the solar photovoltaic panel, inside and outside the phase change material container, at both ends of the thermoelectric material, and in the environment to monitor temperature changes. These sensors transmit temperature data to the controller through an analog-to-digital converter (ADC). Illumination sensors: photodiodes or phototransistors are selected as illumination sensors and placed in appropriate locations to measure light intensity. Similarly, the data of the illumination sensor is transmitted to the controller through the ADC. Power sensors: power sensors such as coulomb meters are installed in the power storage module to monitor the stored power, and the data is also transmitted to the controller.

[0102] Deployment and operation of algorithms in the system

[0103] Algorithm deployment process:

[0104] Processor selection: according to the calculation requirements and real-time requirements of the system, low-cost microcontrollers (such as ARM Cortex-M series) or embedded processors (such as STM32 series) can be selected.

[0105] Software environment: build a corresponding operating system (such as Linux) or real-time operating system (such as FreeRTOS) on the embedded processor, and install a lightweight version of the deep learning framework (such as TensorFlow Lite or PyTorch Mobile) to provide a platform for algorithm running. Use programming languages such as Python or C / C++ to write algorithms for continuous action space (DDPG) in deep reinforcement learning.

[0106] Code.

[0107] Combine the policy network and value network to realize a deep neural network, which can use the application programming interface (API) provided by the deep learning framework. For network structure, multi-layer perceptron (MLP) can be used, the input layer is determined according to the dimension of the state space, the hidden layer is set with appropriate number of neurons according to the calculation ability and performance requirements, and the output layer is designed according to the dimension of the action space. Store the initial parameters of the network in the storage device of the embedded processor (such as SD card or flash memory) for subsequent reading and updating.

[0108] Use the appropriate hardware interface (such as I2C, SPI, or GPIO) to read sensor data from the analog-to-digital converter. Write a driver program to convert the sensor data into the state vector required by the algorithm. For example, combine temperature sensor data, light intensity data, and battery data into a state vector.

[0109] The actuators are then controlled via GPIO or PWM (pulse width modulation) signals. For heat dissipation devices, PWM signals can be used to adjust their operating intensity, enabling continuous control of heat dissipation power. For power output, GPIOs are used to control relays or power switches, adjusting the power output according to the actions generated by the algorithm.

[0110] The operation process of the algorithm:

[0111] like Figure 4 The flowchart shown runs the algorithm. First, the algorithm is initialized to complete the network construction and initialization required for deep learning, preparing for the agent's decision-making and evaluation. Then, the environment interacts with the action execution. The agent generates and executes actions based on the current state, thereby changing the system state. At the same time, rewards are calculated based on the changes in the system state to reflect the quality of the action. Then, the experience replay buffer stores the information of each interaction, providing experience data for subsequent learning, which helps to improve the stability and robustness of the algorithm. To avoid overfitting and enhance the learning effect, a random portion of the stored experience is extracted for learning to prevent the algorithm from falling into a local optimum.

[0112] The Bellman equation is then used to calculate the target value, which is the key basis for updating the value network. Based on the calculated target value, the value network is updated by minimizing the temporal difference error to improve the accuracy of its value assessment of state-action pairs. The policy network is then updated based on the calculated results to guide the policy network to generate better actions to obtain higher long-term cumulative rewards. Finally, the system performance is evaluated to determine whether it has met expectations or stabilized, and to decide whether to continue optimization or end the algorithm.

[0113] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent variation based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A control method for a system of photovoltaic energy storage and cooling coupled with copper-based composite phase change materials, comprising a cooling and energy storage module of copper-based composite phase change materials (2) and an aluminum alloy bracket (3); a power generation module of a solar photovoltaic panel (1) installed at the top; an energy conversion module of a thermoelectric material (4) installed at the bottom; and an electric energy conduction module composed of a support module composed of a stainless steel bracket (6), an energy collection and inversion system (7) module and an external wire (5) at the bottom, wherein the solar photovoltaic panel (1) is closely connected to the copper-based composite phase change materials (2), the aluminum alloy bracket (3) and the thermoelectric material (4) from top to bottom, wherein the energy collection and inversion system (7) is composed of an electric energy collection module, a control unit and an inversion circuit, and wherein: The specific steps are as follows: First, define the state space, action space, and reward function to facilitate the implementation of the algorithm: State Space: The state space covers several parameters that have an important impact on system performance: solar panel temperature , copper-based composite phase change material temperature , the phase change degree of phase change material , the temperature difference between the upper and lower surfaces of the thermoelectric material , Current energy storage capacity , real-time electricity prices of power grids and light intensity , these parameters together constitute the state vector, (1) It fully reflects the performance of the system's operating status at a certain moment; Action Space: The action space defines the set of operations that the system can take, including controlling the operation of the phase change material cooling and energy storage modules, and adjusting the working intensity of the heat dissipation device. , whose range is ; Control the operation of the thermoelectric material energy conversion module and adjust the heat conduction efficiency between the thermoelectric material, phase change material and heat dissipation environment , the range is ; and the operation of the power conduction module to determine the power output power , the range is and output timing, so the action vector can be expressed as (2) Reward function: The total reward includes four sub-rewards: efficiency, energy storage, power output, and stability; The efficiency bonus is: = ( ), (3) in Indicates the change in photoelectric conversion efficiency. When the phase change material works effectively, the photoelectric conversion efficiency increases. > 0, is the corresponding weight coefficient; Energy storage rewards are: (4) in and Represent the energy storage capacity at the current and previous moments respectively, represents the loss in the energy storage process, and is the weight coefficient. When the energy storage effect is good >0; The power output rewards are; (5) in and is the weight coefficient. This function comprehensively considers the economic benefits of outputting electricity and the stability of output power. When the electricity price is high and the power output is stable, a higher reward will be obtained. Stability rewards are; (6) in Indicates the key parameters of the system. represents the number of parameters, and the reward aims to keep the system parameters stable. is the weight coefficient; The total reward function can be calculated as (7) Next, the specific implementation algorithm includes the following steps: Step 1: Initialize the algorithm: First build a policy network and value network The task of the policy network is to generate actions based on the input state information, while the value network is to evaluate the value of the current state-action pair. These networks include an input layer, multiple hidden layers, and an output layer. The input layer receives the state vector, and the output layer outputs the corresponding action or value evaluation result. Then set the parameters for initializing the network and ,These parameters are randomly generated, but it is necessary to ensure that the output range of the network meets the range requirements of the action space and value assessment; At the same time, other parameters are initialized and an experience replay buffer D is created with a capacity of N, which is used to store the experience tuples during the interaction between the agent and the environment: (8) This helps break the correlation between samples and improves the stability of learning; Step 2: Interact the algorithm with the environment and perform actions: The sensor transmits the collected system status information to the agent to form a state vector , the agent will Input the policy network and get the action vector: (9) The action vector contains the control instructions for different modules of the system and the working intensity of the heat dissipation device. Adjustment of the thermal conductivity of thermoelectric materials Regulation, and electrical output power These actions will be converted into specific control signals and sent to the corresponding actuators; After executing the action, the system state will change, and the new state It will be collected by the sensor, and the system will calculate the corresponding reward based on the new state. If the system's photoelectric conversion efficiency is improved, a positive efficiency reward will be calculated according to the corresponding formula. If the energy storage capacity increases and the energy loss is small, a positive energy storage reward will be obtained; Step 3: Use the algorithm to learn and update: First, a batch of experience tuples of size m are randomly extracted from the experience replay buffer D, denoted as , Calculate the target value: For each experience tuple, the target value is calculated according to the Bellman equation: (10) Estimate the target value of the current state-action pair based on the current reward and the expected value of the next state; Then update the value network by minimizing the temporal difference error, i.e. (11) to update the parameters of the value network, which enables the value network to more accurately evaluate the value of state-action pairs, Finally, update the policy network and update the parameters of the policy network according to the deterministic policy gradient algorithm, that is, (12) Encourage the policy network to generate actions that can achieve higher cumulative rewards in the long run.

2. The control method of the photovoltaic energy storage and cooling coupled system based on copper-based composite phase change materials according to claim 1, characterized in that: The thermoelectric material (4) used for the conversion of thermal energy and electrical energy is a silicon-germanium alloy.

3. The control method of the photovoltaic energy storage and cooling coupled system based on copper-based composite phase change materials according to claim 1, characterized in that: The preparation method of the copper-based composite phase change material (2) comprises the following steps: Step 1: preparing copper-based powder by mechanical pulverization; Step 2: Surface treatment of the copper-based powder is performed by electroplating to improve its compatibility with the phase change material; Step 3: Evenly mix the surface-treated copper-based powder with the phase change material, wherein the mass fraction of the copper-based powder is 10%-30%; Step 4: Pour the mixed material into the mold and solidify it under a certain temperature and pressure. For paraffin wax with a melting point of 50-60°C, liquid paraffin has better fluidity. The pouring temperature is controlled at 70-80°C; the pouring pressure is controlled at 0.1-0.5MPa. Step 5: Heat treat the solidified material. The temperature and time of heat treatment are determined according to the composition and performance requirements of the material. The temperature range is 100-150°C to fully melt the paraffin wax and better integrate it with the copper-based powder. The heat treatment time is about 2-6 hours. Step 6: Process the heat-treated material by cutting, grinding, and polishing to obtain the desired shape and size.

Images