An energy sharing economy system based on distributed home energy storage and AI

Abstract

The application relates to the technical field of energy regulation, and discloses an energy sharing economy system based on distributed household energy storage and AI, a distributed energy storage unit, which comprises a solar heat collecting device, a calcination reactor and a storage tank system; the solar heat collecting device is used for collecting solar energy and converting the solar energy into heat energy, driving calcium carbonate in the calcination reactor to generate a decomposition reaction, generating calcium oxide and carbon dioxide; the storage tank system comprises calcium carbonate and calcium oxide storage tanks and a carbon dioxide storage tank; the calcium carbonate and calcium oxide storage tanks adopt a double-bin structure design and respectively store calcium carbonate and calcium oxide. The calcium-based thermochemical energy storage technology is combined with the solar heat collecting device, efficient conversion and stable storage of renewable energy are realized, hidden dangers of traditional battery energy storage and cycle life limitations are avoided, and the safety and sustainability of the household energy system are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of energy regulation technology, and in particular to an energy-sharing economy system based on distributed home energy storage and AI. Background Technology

[0002] As the basic unit of energy consumption, the household's low-carbon transformation is of great significance to carbon emission reduction.

[0003] In rural areas of Northwest, North, and Northeast my country, individual heating systems are the primary method for self-built houses. Due to the dispersed nature of residences, low heat load density, and high pipeline costs, centralized heating is difficult to popularize. Heating fuels mainly consist of loose coal, straw, and firewood, which suffer from low heating efficiency, significant safety hazards, and high carbon emissions. These systems are prone to fires and carbon monoxide poisoning, and lack intelligent control, resulting in serious energy waste.

[0004] Currently, technology companies such as Xiaomi and Huawei are promoting low-carbon living in homes through smart home systems, but their energy storage solutions still mainly focus on electrochemical technologies such as lithium-ion batteries. While this type of technology can achieve energy time-shifting, it has several limitations: high safety risks, as the battery is prone to separator meltdown; limited cycle life, as battery life decreases with charging and discharging; in addition, the energy density of the batteries is low, and the purchase and replacement costs remain high.

[0005] Calcium-based thermochemical energy storage technology has been applied in the industrial field, boasting high energy density and system thermal conversion efficiency, as well as advantages such as strong safety, wide availability of materials, and low cost. However, there are currently no publicly available technical solutions for combining it with artificial intelligence systems and applying it to the home to achieve distributed energy sharing. Summary of the Invention

[0006] This invention provides an energy-sharing economy system based on distributed home energy storage and AI to solve existing technical problems, thereby addressing the issues of short equipment lifespan and high maintenance costs.

[0007] To address the aforementioned technical problems, according to one aspect of the present invention, more specifically, an energy-sharing economy system based on distributed home energy storage and AI, comprising a distributed energy storage unit including a solar thermal collector, a calcination reactor, and a storage tank system; the solar thermal collector is used to collect solar energy and convert it into thermal energy, driving the decomposition reaction of calcium carbonate in the calcination reactor to generate calcium oxide and carbon dioxide; the storage tank system includes a calcium carbonate and calcium oxide storage tank and a carbon dioxide storage tank, wherein the calcium carbonate and calcium oxide storage tank adopts a dual-compartment structure design to store calcium carbonate and calcium oxide respectively;

[0008] A heating unit, connected to the storage tank system, is used to cause calcium oxide and carbon dioxide to undergo a carbonization exothermic reaction, releasing heat; the heating unit is equipped with a first heat exchanger, which is used to transfer the heat released by the reaction to the circulating water medium, forming a primary heat exchange.

[0009] The heat exchange unit includes a second heat exchanger installed at the user end. The second heat exchanger is connected to the heating unit through an insulated pipe to form a closed water circuit. The second heat exchanger is used to receive the heated circulating water medium and release heat to the user through secondary heat exchange.

[0010] The intelligent AI control unit includes a sensor module, a data acquisition and communication module, and an AI control module. The sensor module is used to monitor the status of the energy storage unit and user energy consumption information in real time. The data acquisition and communication module is connected to the sensor module and is used to collect and upload data. The AI ​​control module runs a deep reinforcement learning algorithm model and dynamically controls the material entry and exit and reaction process of the heating unit based on the received data to achieve precise heating for individual households and optimized energy allocation.

[0011] The cloud platform is connected to the intelligent AI control unit and is used to collect the sensor data uploaded by each coordinator, run the trained deep reinforcement learning algorithm to make intelligent decisions, and transmit the running data and optimization decisions to the user terminal in real time.

[0012] Furthermore, in the intelligent AI control unit, the data acquisition and communication module includes a ZigBee communication unit, a coordinator with CC2530 as the main control chip, and an ESP8266-WiFi module; the coordinator is used to convert the format and adapt the protocol of the field data collected by the sensor to the cloud platform, and to forward the control commands issued by the cloud platform to the end execution device.

[0013] Furthermore, the intelligent AI control unit uses the DDQN algorithm as its deep reinforcement learning algorithm. This algorithm is trained offline in the MATLAB environment and decouples action selection from Q-value calculation by introducing a dual network structure of target network and update network.

[0014] Furthermore, the calcium carbonate and calcium oxide storage tanks are equipped with a material conveying device to precisely control the amount of solid materials entering and leaving the tank according to instructions from the cloud platform; the carbon dioxide storage tank is a high-pressure tank equipped with a pressure monitoring and safety relief device, and a heat tracing and insulation system is installed on the outside of the tank.

[0015] Furthermore, the reaction medium of the heating unit is a calcium oxide composite material doped with Zr and Y, and the effective conversion rate decreases by no more than 4.9% after 50 cycles, with an energy storage density of no less than 1950.69 kJ / kg.

[0016] Furthermore, the first and second heat exchangers are plate heat exchanger structures; the second heat exchanger supports multi-point parallel operation, with each user operating independently, enabling individual household control and heat metering.

[0017] Furthermore, the user terminal includes a visual display screen installed in the user's home to display real-time heating data and provide an interactive control interface; the user terminal also includes a mobile terminal application that interfaces with a cloud platform.

[0018] Furthermore, in the closed water circuit, the water that has released heat returns to the heating unit along the return pipe to enter the next cycle.

[0019] Furthermore, the energy sharing economy system adopts a modular distributed architecture, in which multiple distributed energy storage units and multiple heating units work together through the intelligent AI control unit to form a scalable energy sharing network.

[0020] An energy regulation method based on distributed home energy storage and AI includes the following steps:

[0021] Step S1: Collect solar energy through a solar thermal collector to drive the decomposition of calcium carbonate in the calcination reactor, generating calcium oxide and carbon dioxide, which are then stored in a dual-compartment structure of calcium carbonate and calcium oxide storage tanks and a carbon dioxide storage tank, respectively.

[0022] Step S2: The sensor module monitors the status of the energy storage unit and the user's energy consumption information in real time, transmits the data to the AI ​​control module via the data acquisition and communication module, and uploads it to the cloud platform by the coordinator.

[0023] Step S3: The cloud platform runs the trained deep reinforcement learning algorithm model to make intelligent decisions and sends control commands to the end execution devices through the coordinator;

[0024] Step S4: Based on the received decision instructions, the AI ​​control module dynamically regulates the material entry and exit and reaction process in the heating unit, so that calcium oxide and carbon dioxide undergo a carbonization exothermic reaction.

[0025] Step S5: In the heating unit, the heat released by the reaction is transferred to the circulating water medium through the first heat exchanger, forming a primary heat exchange.

[0026] Step S6: The heated circulating water medium is transported to the second heat exchanger at the user end through an insulated pipeline, and the heat is released to the user through secondary heat exchange.

[0027] Step S7: The water that has released heat returns to the heating unit along the return pipe and enters the next cycle; the calcium carbonate generated in the reaction is recovered to the calcium carbonate and calcium oxide storage tank, realizing the recycling of materials.

[0028] This invention provides an energy-sharing economy system based on distributed home energy storage and AI. Compared with existing technologies, the advantages achieved by this method are:

[0029] 1. This invention combines calcium-based thermochemical energy storage technology with solar thermal collectors to achieve efficient conversion and stable storage of renewable energy, avoiding the safety hazards and cycle life limitations of traditional battery energy storage, and significantly improving the safety and sustainability of home energy systems.

[0030] 2. By introducing an intelligent AI control unit and a cloud platform collaborative decision-making mechanism, this invention can perceive users' energy needs and energy storage status in real time, dynamically optimize the heating process, achieve precise heating for individual households and intelligent matching of energy supply and demand, and significantly improve energy utilization efficiency and system response capabilities.

[0031] 3. By constructing a distributed modular architecture, this invention supports the coordinated operation of multiple household energy storage units and heating units, forming a scalable energy sharing network, reducing the construction cost of centralized heating pipelines, and enhancing the autonomy and mutual assistance capabilities of regional energy systems.

[0032] 4. By using doped and modified calcium oxide composite material as the reaction medium and combining it with material recycling design, this invention significantly improves the conversion stability and energy storage density of the system in multiple heat storage and release cycles, extends the service life of the equipment, and reduces long-term operation and maintenance costs. Attached Figure Description

[0033] Figure 1 This is an overall view of the energy storage system in an embodiment of the present invention;

[0034] Figure 2 This is a unit overview diagram of the energy storage system in an embodiment of the present invention. Detailed Implementation

[0035] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Example 1

[0037] like Figure 1 and Figure 2 As shown, the system in this embodiment mainly includes a distributed energy storage unit, a heating unit, a heat exchange unit, an intelligent AI control unit, and a cloud platform.

[0038] The distributed energy storage unit includes solar panels, a calcining furnace, and a storage tank system deployed on a rooftop platform. The solar panels focus solar thermal energy to power the calcining furnace. Decompose, generate and The tank system consists of three independent tanks, each storing... , and ,in and The storage tank adopts a dual-compartment structure design.

[0039] The heating unit is equipped with a carbonization reactor and a first heat exchanger. During periods of heat demand, and An exothermic reaction occurs in the carbonization reactor, and the released heat is transferred to the circulating water medium through the first heat exchanger, forming a primary heat exchange.

[0040] The heat exchange unit includes a second heat exchanger (such as a radiator or underfloor heating coil) installed at each user end. The heated hot water is transported to each household through insulated pipes, and the heat is released to the user through the second heat exchanger. The water after releasing heat returns to the heating unit along the return pipe, forming a closed water circuit.

[0041] The intelligent AI control unit includes a sensor module, a ZigBee communication unit, a coordinator with a CC2530 as the main control chip, and an ESP8266-WiFi module. The sensors collect real-time data on the status of the energy storage device and household energy consumption, which is then transmitted to the coordinator via ZigBee communication. The coordinator is responsible for data collection and preliminary processing, and the ESP8266-WiFi module uploads the data to the cloud platform.

[0042] The cloud platform runs the trained DDQN algorithm model, makes intelligent decisions, and forwards control commands to the end-effectors via a coordinator. Simultaneously, the cloud platform transmits operational data and optimization decisions to the user's mobile terminal in real time, supporting remote monitoring and scheduling.

[0043] Example 2

[0044] A DDQN algorithm control strategy is proposed, wherein the intelligent AI control unit employs the DDQN algorithm as the reinforcement learning algorithm. This algorithm decouples action selection from Q-value calculation by introducing a dual-network structure of a target network and an update network. When calculating the target Q-value, the action corresponding to the maximum Q-value is first determined in the update network, and then this action is substituted into the target network to calculate the target Q-value.

[0045] To verify the applicability of the DDQN algorithm in residential energy storage systems, three algorithms—DQN, DDQN, and MTTP—were trained offline for 500 rounds in the MATLAB environment. The training results show that the DDQN algorithm achieved a heating quality rate of over 90% in the initial training phase, with a convergence speed approximately 25% faster than DQN. In the final 100 rounds of training, the standard deviation of the DDQN algorithm's heating quality index was only 1.8%, significantly better than DQN's 4.97% and MTTP's 5.13%.

[0046] In the 24-hour control strategy comparison, the action space was set to seven adjustment ranges for CO2 flow rate, namely 10%, 20%, 30%, 50%, 70%, 85%, and 100% of the rated flow rate. The total daily heat supply of the system optimized with DDQN was 727.7 MJ, which is about 200 MJ less than that of the unoptimized system (949.3 MJ), showing a significant energy-saving effect.

[0047] Taking into account factors such as heating quality, heating efficiency, CO2 savings rate, and stability, the DDQN algorithm achieved the highest overall score (85.2 points), and therefore was selected as the core control strategy.

[0048] Example 3

[0049] like Figure 2 As shown, a specific design of a storage tank system is provided; the storage tank system consists of two parts: a calcium carbonate / calcium oxide storage tank and a carbon dioxide storage tank.

[0050] The calcium carbonate / calcium oxide storage tank features a dual-compartment design, with one side storing the reactant calcium carbonate and the other side storing the product calcium oxide. The tank is equipped with a material conveying system that allows for precise control of the inflow and outflow of solid materials based on instructions from a cloud platform. The tank is constructed from a high-temperature resistant and corrosion-resistant special alloy, and an inner insulation layer minimizes heat loss.

[0051] Carbon dioxide storage tanks are used to temporarily store carbon dioxide gas generated in the forward reaction, which is then released to participate in the reverse reaction when exothermic reactions are required. The tanks are designed with a high-pressure tank (6MPa) and are equipped with pressure monitoring and safety relief devices to ensure operational safety. An external heat tracing and insulation system is installed on the tank to prevent phase change of carbon dioxide during depressurization and to ensure gas flowability.

[0052] The storage tank system is linked with the cloud platform intelligent control system to automatically adjust the material storage and release strategy based on the energy consumption forecast results: during the low energy consumption period, the excess heat is stored in solid and gaseous media in the form of chemical energy; during the high energy consumption period, the stored media is released to carry out reverse reaction heating.

[0053] Example 4

[0054] An optimization of the reaction medium. The reaction medium of the heating unit adopts a calcium oxide composite material doped with Zr and Y. The conversion rate of pure calcium oxide material decreases significantly after multiple cycles: the conversion rate is 70%-90% in the first cycle, drops to 40%-50% after 10 cycles, and is below 10% after 50 cycles, indicating that it is basically deactivated.

[0055] By doping with Zr and Y elements, the cycling stability was significantly improved. Experimental data showed that after 50 cycles, the effective conversion rate decreased by only 4.9%, and the energy storage density remained as high as 1950.69 kJ / kg, demonstrating a significant performance improvement over pure calcium oxide materials and greatly extending the system's service life.

[0056] Example 5

[0057] like Figure 1 and Figure 2 The figure shows performance data for an energy-sharing economy system based on distributed home energy storage and AI. The system using this invention can achieve the following performance indicators:

[0058] 1. Energy storage density: 3.2 GJ / m³, far exceeding conventional hot water storage;

[0059] 2. Increased solar energy utilization rate: approximately 25%;

[0060] 3. Dependence on fossil fuels reduced by over 60%;

[0061] 4. Heating costs reduced by 20%-40%;

[0062] 5. Carbon dioxide emission reduction: Taking a residential building with 100 residents as an example, the daily carbon dioxide emission is 11 kg, which is about 80% lower than that of traditional gas heating (64.7 kg / day), and about 19.6 tons of carbon dioxide emission reduction per year.

[0063] Cyclic stability: The conversion rate of the Zr and Y doped calcium oxide composite material decreased by ≤4.9% after 50 cycles.

[0064] Example 6

[0065] An intelligent control strategy based on a DDPG-MPC hybrid optimization algorithm. To further improve the system's response capability and control accuracy in complex dynamic environments, this embodiment introduces a hybrid optimization algorithm combining Deep Deterministic Strategy Gradient (DDPG) and Model Predictive Control (MPC) into the intelligent AI control unit, realizing refined dynamic control of material entry and exit and reaction processes in the heating unit.

[0066] The core of this algorithm lies in integrating the decision-making ability of the DDPG reinforcement learning algorithm with the prediction ability of MPC, forming a two-layer control architecture of "learning + prediction":

[0067] Upper-level DDPG controller: responsible for learning the global optimal scheduling strategy from historical energy consumption data and meteorological information, and outputting the ideal control target for a period of time in the future, such as CO2 flow setpoint and reaction temperature target value;

[0068] Lower-level MPC controller: Based on the current system state and the reference trajectory of DDPG output, combined with the system dynamics model, it continuously optimizes the control quantity and adjusts the opening of the material conveying device and reactor valve in real time to ensure that the system maximizes energy efficiency and response speed while meeting the user's energy needs.

[0069] The mathematical expression of this hybrid algorithm is as follows:

[0070] 1. Definition of state space:

[0071] System status Includes: Remaining energy in energy storage units Indoor and outdoor temperature difference for users User energy demand forecast Time characteristics , represented as:

[0072] ;

[0073] 2. Definition of Action Space:

[0074] action CO2 flow rate regulation coefficient of the heating unit , with reactor feed rate Coupling control:

[0075] ;

[0076] 3. DDPG target network update mechanism:

[0077] The DDPG algorithm uses an Actor network. Output actions, Critic network To assess the value of an action, the target network is updated as follows:

[0078] ;

[0079] ;

[0080] in, This is the soft update coefficient.

[0081] 4. MPC Optimization Objective Function:

[0082] In each control cycle, MPC solves the following finite-time optimization problem:

[0083] ;

[0084] in, This refers to the actual heating temperature. The reference temperature output by DDPG; Q is the control variable; Q and R are the weight matrices; N is the prediction time step.

[0085] 5. Scrolling optimization and feedback correction:

[0086] At each sampling time, MPC adjusts the current state accordingly. Solve the optimization problem to obtain the optimal control sequence. , ..., only the first step is executed. And then re-optimize in the next moment to form a closed-loop control.

[0087] Through this hybrid optimization algorithm, the system can predict and proactively adjust to avoid response lag when user energy consumption patterns change abruptly (such as extreme weather or temporary increases in heating demand); at the same time, it maintains high energy efficiency during steady-state operation. Experiments show that in scenarios with sudden changes in user behavior, the system response time using the DDPG-MPC algorithm is reduced by approximately 32% compared to the single DDQN algorithm, and the average energy efficiency is improved by approximately 8.7%, with no control oscillations or overshoot observed in 50 consecutive runs.

[0088] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. An energy-sharing economy system based on distributed home energy storage and AI, characterized in that: The distributed energy storage unit includes a solar thermal collector, a calcination reactor, and a storage tank system. The solar thermal collector is used to collect solar energy and convert it into thermal energy to drive the decomposition reaction of calcium carbonate in the calcination reactor to generate calcium oxide and carbon dioxide. The storage tank system includes a calcium carbonate and calcium oxide storage tank and a carbon dioxide storage tank. The calcium carbonate and calcium oxide storage tank adopts a dual-compartment structure design to store calcium carbonate and calcium oxide respectively. A heating unit, connected to the storage tank system, is used to cause calcium oxide and carbon dioxide to undergo a carbonization exothermic reaction, releasing heat; the heating unit is equipped with a first heat exchanger, which is used to transfer the heat released by the reaction to the circulating water medium, forming a primary heat exchange. The heat exchange unit includes a second heat exchanger installed at the user end. The second heat exchanger is connected to the heating unit through an insulated pipe to form a closed water circuit. The second heat exchanger is used to receive the heated circulating water medium and release heat to the user through secondary heat exchange. The intelligent AI control unit includes a sensor module, a data acquisition and communication module, and an AI control module. The sensor module is used to monitor the status of the energy storage unit and user energy consumption information in real time. The data acquisition and communication module is connected to the sensor module and is used to collect and upload data. The AI ​​control module runs a deep reinforcement learning algorithm model to dynamically control the material entry and exit and reaction process of the heating unit based on the received data. The cloud platform is connected to the intelligent AI control unit and is used to collect the sensor data uploaded by each coordinator, run the trained deep reinforcement learning algorithm to make intelligent decisions, and transmit the running data and optimization decisions to the user terminal in real time.

2. The energy-sharing economy system based on distributed home energy storage and AI according to claim 1, characterized in that: In the intelligent AI control unit, the data acquisition and communication module includes a ZigBee communication unit, a coordinator with CC2530 as the main control chip, and an ESP8266-WiFi module. The coordinator is used to convert the format and adapt the protocol of the field data collected by the sensors and upload it to the cloud platform, and to forward the control commands issued by the cloud platform to the end execution device.

3. The energy sharing economy system based on distributed home energy storage and AI according to claim 1, characterized in that: The intelligent AI control unit uses the DDQN algorithm, a deep reinforcement learning algorithm that is trained offline in the MATLAB environment. By introducing a dual-network structure of target network and update network, action selection and Q-value calculation are decoupled.

4. The energy-sharing economy system based on distributed home energy storage and AI according to claim 1, characterized in that: The calcium carbonate and calcium oxide storage tanks are equipped with a material conveying device, which is used to precisely control the amount of solid materials entering and leaving the tanks according to instructions from the cloud platform. The carbon dioxide storage tank is a high-pressure tank equipped with pressure monitoring and safety relief devices, and a heat tracing and insulation system is installed on the outside of the tank.

5. The energy-sharing economy system based on distributed home energy storage and AI according to claim 1, characterized in that: The reaction medium of the heating unit is a calcium oxide composite material doped with Zr and Y. After 50 cycles, the effective conversion rate decreases by no more than 4.9%, and the energy storage density is not less than 1950.69 kJ / kg.

6. The energy-sharing economy system based on distributed home energy storage and AI according to claim 1, characterized in that: The first and second heat exchangers are plate heat exchanger structures; the second heat exchanger supports multi-point parallel operation.

7. The energy sharing economy system based on distributed home energy storage and AI according to claim 1, characterized in that: The user terminal includes a visual display screen installed in the user's home to display real-time heating data and provide an interactive control interface; the user terminal also includes a mobile terminal application that interfaces with a cloud platform.

8. The energy-sharing economy system based on distributed home energy storage and AI according to claim 1, characterized in that: In the closed water circuit, the water that has released heat returns to the heating unit along the return pipe to enter the next cycle.

9. The energy-sharing economy system based on distributed home energy storage and AI according to any one of claims 1-8, characterized in that, The energy sharing economy system adopts a modular distributed architecture, in which multiple distributed energy storage units and multiple heating units work together through the intelligent AI control unit to form a scalable energy sharing network.

10. An energy regulation method based on distributed home energy storage and AI, characterized in that, The energy-sharing economy system based on distributed home energy storage and AI, applicable to any one of claims 1-8, comprises the following steps: Step S1: Collect solar energy through a solar thermal collector to drive the decomposition of calcium carbonate in the calcination reactor, generating calcium oxide and carbon dioxide, which are then stored in a dual-compartment structure of calcium carbonate and calcium oxide storage tanks and a carbon dioxide storage tank, respectively. Step S2: The sensor module monitors the status of the energy storage unit and the user's energy consumption information in real time, transmits the data to the AI ​​control module via the data acquisition and communication module, and uploads it to the cloud platform by the coordinator. Step S3: The cloud platform runs the trained deep reinforcement learning algorithm model to make intelligent decisions and sends control commands to the end execution devices through the coordinator; Step S4: Based on the received decision instructions, the AI ​​control module dynamically regulates the material entry and exit and reaction process in the heating unit, so that calcium oxide and carbon dioxide undergo a carbonization exothermic reaction. Step S5: In the heating unit, the heat released by the reaction is transferred to the circulating water medium through the first heat exchanger, forming a primary heat exchange. Step S6: The heated circulating water medium is transported to the second heat exchanger at the user end through an insulated pipeline, and the heat is released to the user through secondary heat exchange. Step S7: The water that has released heat returns to the heating unit along the return pipe and enters the next cycle; the calcium carbonate generated in the reaction is recovered to the calcium carbonate and calcium oxide storage tank.

Images