Solar low-calorific-value heat storage heating system

Abstract

The invention discloses a solar low-heat-value heat storage heating system. According to the invention, through the AI prediction sub-module of the intelligent heat energy allocation module, the matching precision of energy supply and demand is significantly improved. The system can collect various data such as meteorological parameters, historical loads and heat storage states in real time, heat supply requirements, solar input fluctuation and heat storage efficiency are accurately pre-judged through feature extraction and multi-model prediction, and prediction results are continuously corrected through error calibration. The system can know how much heat is needed and how much solar energy can be provided in advance, so that energy flow in heat collection, heat storage and heat supply links is more reasonably allocated, the situation of heat waste or insufficient supply is reduced, the self-adaptive learning sub-module effectively strengthens the long-term operation capacity of the system, the system achieves self-learning, and the energy utilization rate of the system is improved. The operation stability is kept, the energy efficiency can be continuously optimized along with the external environment, more energy is saved after long-term use, and more convenience and rapidness are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of solar thermal storage and heating, and specifically relates to a solar low calorific value thermal storage and heating system. Background Art

[0002] Solar thermal storage heating uses solar collectors to collect solar radiation and store the heat in a thermal storage medium. This energy is then used to provide indoor heating during periods of low sunlight, such as winter or on cloudy days. The system primarily consists of a solar collector, a thermal storage device (such as a water tank, phase change thermal storage material, or geothermal heat storage), a circulation pump, a control system, and terminal heating equipment (such as floor heating, radiators, or fan coil units). During daytime hours with ample sunlight, the collector converts solar energy into heat, heating the thermal storage medium and storing it in a dedicated container. At night or during rainy weather, the system automatically activates and releases the stored heat for heating. Solar thermal storage heating systems not only effectively reduce the consumption of traditional fossil fuels and lower carbon emissions, but also increase solar energy utilization and achieve efficient energy recycling. With advances in energy storage technology and the development of intelligent control, this system is gradually becoming more efficient, energy-saving, and environmentally friendly, and is widely used in clean heating projects in residential areas, agricultural greenhouses, and public buildings.

[0003] However, existing technologies have two major shortcomings: first, traditional systems lack the ability to conduct real-time analysis and accurate prediction of multi-source data, and energy allocation relies more on experience or simple timing control, which can easily lead to heat waste or insufficient supply; second, it is difficult to automatically mine data patterns and dynamically adjust parameters during long-term operation, requiring frequent manual intervention and adjustment. The stability and adaptability of the system will gradually decline with environmental changes (such as weather and electricity prices). Summary of the Invention

[0004] The purpose of the present invention is to provide a solar low calorific value heat storage heating system in order to solve the above-mentioned problems.

[0005] The technical solution adopted by the present invention is as follows: a solar low calorific value heat storage heating system, the system comprising: a solar heat collection module, a low calorific value heat storage module, an intelligent heat energy allocation module, a heating output module, a control system module and an energy monitoring and user interaction module;

[0006] The intelligent heat energy allocation module is internally configured with: an AI prediction submodule, a dynamic allocation decision submodule, a multi-objective optimization submodule, and an adaptive learning submodule;

[0007] The heat transfer medium circulation loop outlet of the solar heat collection module is connected to the inlet end of the heat exchange coil of the low calorific value heat storage module through a pipeline;

[0008] The outlet end of the heat exchange coil of the low calorific value heat storage module is connected to the inlet end of the heat exchange unit of the heating output module through a pipeline.

[0009] The communication interface of the control system module is respectively connected to the temperature sensor data output end of the solar thermal collection module, the temperature sensor data output end of the low calorific value heat storage module, and the flow sensor data output end of the heating output module;

[0010] The control signal output end of the actuator drive unit of the control system module is respectively connected to the motor control end of the solar thermal collector module tracking bracket, the inverter control end of the heating output module circulating pump and the servo mechanism control end of the flow control valve.

[0011] The data acquisition terminal signal input terminal of the energy monitoring and user interaction module is connected in parallel with the sensor data output terminal of each module to summarize the operation data;

[0012] The signal output end of the user command input module of the energy monitoring and user interaction module is connected to the command receiving end of the control system module through the communication interface, and the user settings are transmitted to the control system for execution.

[0013] The prediction data output terminal of the intelligent thermal energy allocation module is connected to the algorithm input terminal of the control system module to provide a prediction basis for the generation of the control strategy.

[0014] In a preferred embodiment, the solar thermal collector module is internally provided with a heat collecting plate array, a tracking bracket, a heat transfer medium circulation loop, and a temperature sensor. The heat collecting plate array is composed of a plurality of flat plate or vacuum tube type heat collecting plates connected in parallel, the surface of which is covered with a high-absorption coating, and is responsible for converting solar radiation energy into heat energy; the tracking bracket adjusts the angle of the heat collecting plate in real time through a dual-axis drive mechanism, and the dual-axis drive mechanism includes a horizontal rotation axis and a pitch adjustment axis to ensure that the plate surface remains perpendicular to the incident sunlight; the heat transfer medium circulation loop uses a mixed solution of antifreeze and water as a heat carrier, connects the heat collecting plate and the heat storage module through a pipe, and uses the temperature difference to drive natural circulation or pump forced circulation to extract heat; the temperature sensors are evenly distributed on the surface of the heat collecting plate and in the circulation pipe, and monitor the heat collection efficiency and medium temperature in real time, providing an adjustment basis for the control system.

[0015] In a preferred embodiment, the low-heat-value heat storage module is internally provided with a heat storage material layer, a phase change container, a heat insulation shell, and a heat exchange coil. The heat storage material layer is composed of organic phase change materials such as paraffin and fatty acids and expanded graphite, has a stable phase change temperature of 40-60°C, and has a high latent heat density. The phase change container is a sealed metal tank structure that encapsulates the heat storage material as an independent unit to prevent material leakage and aging. The heat insulation shell is composed of aerogel felt and polyurethane foam and covers the outer surface of the phase change container to reduce heat loss during heat storage. The heat exchange coil is spirally embedded in the phase change material layer, with one end connected to the heat transfer medium circuit of the solar heat collection module and the other end connected to the hot water circulation system of the heating output module, realizing heat storage and release through medium flow.

[0016] In a preferred embodiment, the AI prediction sub-module is internally provided with a data acquisition layer, a feature extraction unit, a prediction model cluster, and an error calibration module. The data acquisition layer is responsible for real-time acquisition of multi-source heterogeneous data, including meteorological parameters such as solar irradiance, ambient temperature, and cloud cover provided by weather stations, historical heating load data recorded by building energy consumption monitoring systems, and state information such as temperature and capacity of the heat storage module. All data are synchronized to the module database at a 15-minute cycle. The feature extraction unit extracts periodic features such as daily / weekly / monthly load fluctuation patterns, correlation features such as the negative correlation between solar irradiance and heating demand, and abnormal point features such as load surges caused by extreme weather from the original data based on time series analysis techniques, forming structured feature vectors that can be input into the model. The prediction model cluster contains multiple parallel running sub-models, among which the long short-term memory network is responsible for capturing the time dependence of heating demand, the random forest model is used to predict the nonlinear fluctuations of solar input, and the support vector regression is used to predict the heat storage efficiency in small sample scenarios. The fusion weights of each sub-model are dynamically adjusted to ensure that the prediction results continuously approach the true value as the system runs.

[0017] In a preferred embodiment, the dynamic allocation decision sub-module dynamically determines the allocation ratio of solar thermal energy between direct heating and heat storage charging based on multi-source data. The module's inputs include the solar input prediction value P sol (unit: kilowatts), the future period heating demand prediction curve D(t) (unit: kilowatts), and the real-time collected heat storage module state (such as the current heat storage capacity C cur , the maximum heat storage capacity C max , the heat storage material temperature T store , and the real-time electricity price of the power grid ρ(t) (unit: yuan / kilowatt-hour).

[0018] The operation flow of the module is divided into three key links: firstly, the basic demand matching, according to the real-time comparison of solar input and heating demand to determine the initial allocation ratio α base : if the solar energy is sufficient P sol ≥D(t), the current heating demand is met first, and the remaining heat energy is used for heat storage; if the solar energy is insufficient P sol <D(t), all solar energy is used for heating, and the insufficient part is supplemented by the heat storage module. Secondly, the economic correction, combined with the fluctuation of electricity price to adjust the allocation strategy: in the valley electricity price period ρ(t) is lower, the direct heating ratio is reduced, and the heat storage charging is increased to utilize the low-price energy; in the peak electricity price period ρ(t) is higher, the direct heating ratio is increased, and the dependence on high-price grid energy is reduced. Finally, the heat storage state correction, according to the characteristics of the heat storage material to optimize the charging behavior: if the heat storage temperature T store is close to the material phase change temperature T phase , the charging speed is reduced to avoid heat loss caused by overcharging; if the temperature is too low, T store <<T phase , the charging ratio is increased to improve the heat storage efficiency. Finally, the ratio α after the comprehensive correction of the basic matching, the economy and the heat storage state is taken as the output to guide the heat energy allocation execution.

[0019] Wherein:

[0020] The comprehensive correction formula of the dynamic allocation ratio is:

[0021]

[0022] Wherein:

[0023] α base represents the basic allocation ratio (determined by the matching relationship between solar energy and demand, range 0≤α base ≤1);

[0024] β represents the electricity price correction coefficient (empirical value, reflecting the influence intensity of electricity price on the allocation ratio, 0<β<1);

[0025] ρ(t) represents the real-time electricity price in the current period (yuan / kilowatt hour);

[0026] ρth represents the electricity price threshold (the upper limit of the valley electricity price, yuan / kilowatt hour);

[0027] γ represents the temperature correction coefficient (empirical value, reflecting the influence intensity of the heat storage temperature on the allocation ratio, 10<γ<1);

[0028] Tphase represents the heat storage material phase change temperature reference value (℃, such as 60℃ of paraffin);

[0029] Tstore represents the current temperature of the heat storage module (℃).

[0030] In a preferred embodiment, the multi-objective optimization submodule, based on the initial allocation strategy output by the dynamic allocation decision submodule, takes user comfort as the bottom line constraint, and simultaneously weighs the three key objectives of energy cost, equipment life, and system stability, to generate a comprehensive optimal heat energy allocation plan through quantitative analysis and a dynamic weighting mechanism;

[0031] The input of the module includes: initial parameters of the dynamic allocation strategy (such as direct heating ratio α, thermal storage charging ratio 1-α), real-time collected energy cost data (such as grid electricity price ρ(t), solar energy equivalent cost ρ sol ), equipment status data (such as the number of times the heat storage module is charged and discharged N, temperature fluctuation ΔT store ), and the user-set comfort threshold (such as the indoor temperature allowable range [T set -δ,T set +δ]).

[0032] The operation of the module is divided into three key links:

[0033] Target quantification: converting abstract targets into calculable indicators. Energy cost is quantified by multiplying the current total energy consumption by the electricity price (C cost ); the equipment life is reflected by the increment of the number of charge and discharge cycles of the heat storage module (ΔN) to reflect the material aging; the system stability is reflected by the fluctuation amplitude of the heat storage temperature (ΔT store )Measure operational risk.

[0034] Dynamic weight adjustment: Dynamically adjust the priority weight of each target w according to the system operation stage (such as heating peak / valley) or user preferences (such as "economic mode" and "comfort mode"). cost (cost weight), w wear (life weight), w stab (stability weight), the sum of the three is always 1. For example, when the user selects "Economy Mode", wcost increases; when heating demand is high at night in winter, wstab increases to ensure thermal storage stability.

[0035] Comprehensive optimization solution: Under the premise of satisfying the comfort constraint (indoor temperature does not exceed [Tset-δ, Tset+δ]), the allocation strategy that minimizes the comprehensive goal is found through iterative calculation to ensure that energy utilization achieves the optimal balance between cost, life and stability.

[0036] The quantitative expression of the comprehensive optimization objective is:

[0037]

[0038] in:

[0039] J represents the comprehensive optimization target value (the smaller the value, the better the solution);

[0040] w cost ,w wear ,w stab are the dynamic weight coefficients of energy cost, equipment life and system stability (0 <w cost ,w wear ,w stab <1, and the sum is 1);

[0041] C cost Indicates the total energy cost of the current period (yuan);

[0042] C base Indicates the historical benchmark energy cost (yuan, such as the average value of the past week);

[0043] ΔN represents the increment of the number of times the heat storage module is charged and discharged during the current period (times);

[0044] N base Indicates the increment of historical benchmark charge and discharge times (times, such as the average value of the past week);

[0045] ΔT store Indicates the temperature change of the heat storage module in the current period (℃);

[0046] T base Indicates the change in historical base temperature (°C, such as the average value of the past week).

[0047] In a preferred embodiment, the adaptive learning submodule is internally equipped with a data storage repository, a pattern recognition engine, a parameter self-update mechanism, and a policy iteration module. The data storage repository utilizes a hierarchical storage structure: the bottom layer contains raw operation logs containing historical forecast data, allocation strategies, actual energy consumption, and device status; the middle layer contains cleaned key indicators such as daily energy costs, thermal storage module charge and discharge times, and temperature fluctuations; and the top layer contains scenario-based knowledge graphs, including feature labels for typical operating scenarios such as winter peak and summer trough. The pattern recognition engine, based on unsupervised learning algorithms such as DBSCAN clustering, mines implicit correlation patterns from historical data. For example, it identifies patterns in thermal storage consumption under conditions of continuous rainy weather and low night temperatures, or the cost optimization potential of a strategy combining charging during off-peak electricity prices and discharging during peak electricity prices. The parameter self-update mechanism automatically adjusts key parameters of the prediction and allocation decision submodules, such as the learning rate of the long-short-term memory network and the price correction coefficient for dynamic allocation, using a policy gradient algorithm from reinforcement learning, with the objective function of minimizing comprehensive energy consumption costs, thus avoiding the lag inherent in manual parameter adjustment. The strategy iteration module triggers system-wide backtracking verification regularly, such as weekly, inputs historical operating data into the current optimal strategy, and calculates the deviation between theoretical energy consumption and actual energy consumption. If the deviation exceeds a threshold, such as 5%, the pattern recognition results are called to generate a new set of candidate strategies. The optimal strategy is selected through A / B testing to replace the old strategy, thereby achieving continuous evolution of the decision-making logic.

[0048] In a preferred embodiment, the heating output module is internally equipped with a heat exchanger, a circulating pump, a terminal heat dissipation device, and a flow control valve. The heat exchanger, consisting of a plate heat exchanger and a circulating water tank, transfers the heat energy released by the heat storage module to the dedicated circulating water for heating. The circulating pump is installed in the main heating pipeline, providing power to drive the hot water through the system. The terminal heat dissipation device can include a radiator, floor heating coil, or fan coil, depending on the building type (including residential, office, and public spaces), which releases the hot water heat into the room through radiation or convection. Flow control valves are distributed in each branch pipeline, and by adjusting the valve opening, the heat supply is balanced between different areas to ensure uniform indoor temperature.

[0049] In a preferred embodiment, the control system module is internally equipped with a central processing unit (CPU), a communication interface, an actuator drive unit, and a safety protection unit. The CPU utilizes an industrial-grade PLC controller, integrating a multi-objective optimization algorithm and a dynamic allocation strategy. It is responsible for receiving sensor data from each module and generating control instructions. The communication interface supports Modbus, CAN bus, and 5G wireless protocols, enabling real-time data exchange with the solar collector module, heat storage module, heating module, and energy monitoring module. The actuator drive unit connects to the tracking bracket motor, the circulating pump inverter, and the flow control valve servo mechanism, converting control instructions into mechanical actions. The safety protection unit integrates overtemperature protection, pressure alarm, and fault self-diagnosis functions. The overtemperature protection function cuts off circulation when the collector plate temperature exceeds 100°C, and the pressure alarm function activates pressure relief when the pipeline pressure exceeds 0.6MPa, ensuring safe system operation.

[0050] In a preferred embodiment, the energy monitoring and user interaction module is internally equipped with a data acquisition terminal, a real-time monitoring interface, a user command input module, and an abnormality alarm unit. The data acquisition terminal aggregates sensor data from each module, including heat collection temperature, heat storage capacity, and heating flow, and uploads this data to a cloud database on a minute-by-minute basis. The real-time monitoring interface displays the system's operating status in the form of a visual dashboard, including solar irradiation curves, heat storage capacity trends, and heating temperature distribution in each area. The user command input module supports touch screen, mobile app, and voice interaction, allowing users to set target temperatures, select operating modes (including economy mode, comfort mode, energy-saving mode), or manually adjust flow distribution. The abnormality alarm unit triggers audible and visual alarms through threshold detection, including heat storage temperature below 20°C and circulation pump failure shutdown, and simultaneously pushes notifications to the user's mobile phone, prompting prompts to promptly troubleshoot the problem.

[0051] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0052] 1. In this invention, the AI ​​prediction submodule of the intelligent thermal energy allocation module significantly improves the matching accuracy of energy supply and demand. The system can collect multiple types of data in real time, such as meteorological parameters, historical loads, and thermal storage status. Through feature extraction and multi-model prediction, it accurately predicts heating demand, solar energy input fluctuations, and thermal storage efficiency. It then continuously corrects the prediction results through error calibration. This allows the system to know in advance how much heat will be needed and how much solar energy can provide, thereby more rationally allocating energy flows in the heat collection, storage, and heating links, reducing heat waste or insufficient supply.

[0053] 2. In the present invention, the adaptive learning submodule effectively enhances the long-term operation capability of the system. The data sedimentation warehouse stores multi-layer data from raw logs to scenario knowledge. The pattern recognition engine can dig out hidden experiences such as the heat storage consumption rules during continuous rainy weather and low temperatures, and the cost optimization potential of charging at valley electricity prices. The parameter self-update mechanism automatically adjusts the key parameters of the prediction model and control strategy without frequent manual adjustment. The strategy iteration module verifies historical data every week. If it finds that the current strategy is less effective, a new strategy is generated and replaced through testing. These functions enable the system to achieve self-learning, which not only maintains operational stability, but also continuously optimizes energy efficiency as the external environment changes. It is more energy-saving, convenient, and quick to use in the long term. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a block diagram of the overall system of the present invention;

[0055] Figure 2 This is a system block diagram of the intelligent thermal energy allocation module in the present invention;. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0057] Example:

[0058] Reference Figure 1-2 A solar low calorific value heat storage heating system, the system includes: a solar heat collection module, a low calorific value heat storage module, an intelligent heat energy allocation module, a heating output module, a control system module and an energy monitoring and user interaction module;

[0059] The internal settings of the intelligent heat energy allocation module include: AI prediction submodule, dynamic allocation decision submodule, multi-objective optimization submodule and adaptive learning submodule;

[0060] The heat transfer medium circulation loop outlet of the solar heat collection module is connected to the inlet end of the heat exchange coil of the low calorific value heat storage module through a pipeline;

[0061] The outlet of the heat exchange coil of the low calorific value heat storage module is connected to the inlet of the heat exchange unit of the heating output module through a pipe.

[0062] The communication interface of the control system module is respectively connected to the temperature sensor data output terminal of the solar thermal collection module, the temperature sensor data output terminal of the low calorific value heat storage module, and the flow sensor data output terminal of the heating output module;

[0063] The control signal output end of the actuator drive unit of the control system module is connected to the motor control end of the solar collector module tracking bracket, the inverter control end of the heating output module circulating pump and the servo mechanism control end of the flow control valve.

[0064] The signal input terminal of the data acquisition terminal of the energy monitoring and user interaction module is connected in parallel with the sensor data output terminal of each module to summarize the operating data;

[0065] The signal output end of the user command input module of the energy monitoring and user interaction module is connected to the command receiving end of the control system module through the communication interface, and the user settings are transmitted to the control system for execution.

[0066] The prediction data output of the intelligent thermal energy allocation module is connected to the algorithm input of the control system module to provide a prediction basis for the generation of control strategies.

[0067] The solar thermal module is equipped with a solar collector array, a tracking bracket, a heat transfer medium circulation loop, and a temperature sensor. The solar collector array consists of multiple flat-plate or vacuum tube solar collectors connected in parallel, covered with a high-absorptivity coating, which converts solar radiation into heat. The tracking bracket adjusts the angle of the solar collector in real time via a dual-axis drive mechanism. The dual-axis drive mechanism includes a horizontal rotation axis and a pitch adjustment axis to ensure that the panel surface remains perpendicular to the incident sunlight. The heat transfer medium circulation loop uses a mixed solution of antifreeze and water as the heat carrier, connecting the solar collector and the heat storage module via pipes. Heat is extracted by natural circulation driven by temperature differences or forced circulation driven by pumps. Temperature sensors are evenly distributed on the surface of the solar collector and in the circulation pipes, monitoring the heat collection efficiency and medium temperature in real time, providing a basis for adjustment in the control system.

[0068] The low-calorific-value heat storage module is internally equipped with a heat storage material layer, a phase-change container, an insulating outer shell, and a heat exchange coil. The heat storage material layer is a composite of organic phase-change materials such as paraffin and fatty acids, and expanded graphite. It has a stable phase-change temperature of 40-60°C and a high latent heat density. The phase-change container is a sealed metal canister that encapsulates the heat storage material as a separate unit to prevent leakage and aging. The insulating outer shell, composed of a composite of aerogel felt and polyurethane foam, covers the outer surface of the phase-change container to reduce heat loss during the heat storage process. The heat exchange coil is spirally embedded in the phase-change material layer, with one end connected to the heat transfer medium circuit of the solar thermal collection module and the other end to the hot water circulation system of the heating output module. Heat is stored and released through the flow of the medium.

[0069] The AI prediction submodule is internally provided with a data acquisition layer, a feature extraction unit, a prediction model cluster, and an error calibration module. The data acquisition layer is responsible for real-time acquisition of multi-source heterogeneous data, including meteorological parameters such as solar irradiance, ambient temperature, and cloud cover provided by weather stations, historical heating load data recorded by building energy consumption monitoring systems, and state information such as temperature and capacity of the heat storage module. All data is synchronized to the module database at a 15-minute cycle. The feature extraction unit extracts periodic features such as daily / weekly / monthly load fluctuation patterns, correlation features such as the negative correlation between solar irradiance and heating demand, and anomaly point features such as load surges caused by extreme weather from the original data based on time series analysis techniques, forming structured feature vectors that can be input into the model. The prediction model cluster includes multiple parallel running submodels, among which the long short-term memory network is responsible for capturing the time dependence of heating demand, the random forest model is used to predict the nonlinear fluctuations of solar input, and the support vector regression is used to predict the heat storage efficiency in small sample scenarios. The fusion weights of each submodel are dynamically adjusted to ensure that the prediction results continuously approach the true value as the system runs.

[0070] The dynamic allocation decision submodule dynamically determines the allocation ratio of solar thermal energy between direct heating and heat storage charging based on multi-source data. The inputs of the module include the solar input prediction value P sol (unit: kilowatts), the future period heating demand prediction curve D(t) (unit: kilowatts), and the real-time collected heat storage module state (such as the current heat storage capacity C cur , the maximum heat storage capacity C max , the heat storage material temperature T store , and the real-time electricity price of the power grid ρ(t) (unit: yuan / kilowatt-hour).

[0071] The operation process of the module consists of three key links: first, basic demand matching, which determines the initial allocation ratio a base based on the real-time comparison of solar input and heating demand: if the solar energy is sufficient P sol ≥ D(t), the current heating demand is prioritized, and the remaining thermal energy is used for heat storage; if the solar energy is insufficient P sol <D(t)), all solar energy is used for heating, and the insufficient part is supplemented by the heat storage module. Second, economic correction, which adjusts the allocation strategy in combination with electricity price fluctuations: in the valley electricity price period ρ(t) is lower, the direct heating ratio is reduced, and the heat storage charging is increased to utilize low-priced energy; in the peak electricity price period ρ(t) is higher, the direct heating ratio is increased, and the dependence on high-priced grid energy is reduced. Finally, the heat storage state correction optimizes the charging behavior according to the characteristics of the heat storage material: if the heat storage temperature T storeClose to the material phase transition temperature T phase , the charging speed is reduced to avoid heat loss caused by overcharging; if the temperature is too low, T store <<T phase , then the charging ratio is increased to improve the heat storage efficiency. Finally, the ratio α after comprehensive basic matching, economy and heat storage state correction is used as the output to guide the execution of heat energy allocation;

[0072] in:

[0073] The comprehensive correction formula for the dynamic allocation ratio is:

[0074]

[0075] in:

[0076] α base Indicates the basic allocation ratio (determined by the matching relationship between solar energy and demand, range 0≤α base ≤1);

[0077] β represents the electricity price correction coefficient (empirical value, reflecting the impact of electricity price on the allocation ratio, 0<β<1);

[0078] ρ(t) represents the real-time electricity price in the current period (yuan / kWh);

[0079] ρth represents the electricity price threshold (valley price ceiling, yuan / kWh);

[0080] γ represents the temperature correction coefficient (empirical value, reflecting the influence of heat storage temperature on the distribution ratio, 10<γ<1);

[0081] Tphase represents the reference value of the phase change temperature of the heat storage material (°C, such as 60°C for paraffin);

[0082] Tstore represents the current temperature of the thermal storage module (°C).

[0083] Based on the initial allocation strategy output by the dynamic allocation decision submodule, the multi-objective optimization submodule uses user comfort as the bottom line constraint, while weighing the three key objectives of energy cost, equipment life, and system stability. Through quantitative analysis and a dynamic weighting mechanism, it generates a comprehensive and optimal heat allocation plan.

[0084] The input of the module includes: initial parameters of the dynamic allocation strategy (such as direct heating ratio α, thermal storage charging ratio 1-α), real-time collected energy cost data (such as grid electricity price ρ(t), solar energy equivalent cost ρ sol ), equipment status data (such as the number of times the heat storage module is charged and discharged N, temperature fluctuation ΔT store ), and the user-set comfort threshold (such as the indoor temperature allowable range [T set -δ,Tset +δ]).

[0085] The operation of the module is divided into three key links:

[0086] Target quantification: converting abstract targets into calculable indicators. Energy cost is quantified by multiplying the current total energy consumption by the electricity price (C cost ); the equipment life is reflected by the increment of the number of charge and discharge cycles of the heat storage module (ΔN) to reflect the material aging; the system stability is reflected by the fluctuation amplitude of the heat storage temperature (ΔT store )Measure operational risk.

[0087] Dynamic weight adjustment: Dynamically adjust the priority weight of each target w according to the system operation stage (such as heating peak / valley) or user preferences (such as "economic mode" and "comfort mode"). cost (cost weight), w wear (life weight), w stab (stability weight), the sum of the three is always 1. For example, when the user selects "Economy Mode", wcost increases; when heating demand is high at night in winter, wstab increases to ensure thermal storage stability.

[0088] Comprehensive optimization solution: Under the premise of satisfying the comfort constraint (indoor temperature does not exceed [Tset-δ, Tset+δ]), the allocation strategy that minimizes the comprehensive goal is found through iterative calculation to ensure that energy utilization achieves the optimal balance between cost, life and stability.

[0089] The quantitative expression of the comprehensive optimization objective is:

[0090]

[0091] in:

[0092] J represents the comprehensive optimization target value (the smaller the value, the better the solution);

[0093] w cost ,w wear ,w stab are the dynamic weight coefficients of energy cost, equipment life and system stability (0 <w cost ,w wear ,w stab <1, and the sum is 1);

[0094] C cost Indicates the total energy cost of the current period (yuan);

[0095] C base Indicates the historical benchmark energy cost (yuan, such as the average value of the past week);

[0096] ΔN represents the increment of the number of times the heat storage module is charged and discharged during the current period (times);

[0097] N base Indicates the increment of historical benchmark charge and discharge times (times, such as the average value of the past week);

[0098] ΔT store Indicates the temperature change of the heat storage module in the current period (℃);

[0099] T base Indicates the change in historical base temperature (°C, such as the average value of the past week).

[0100] The adaptive learning submodule comprises a data storage repository, a pattern recognition engine, a parameter self-update mechanism, and a policy iteration module. The data storage repository utilizes a hierarchical storage structure. The bottom layer contains raw operation logs containing historical forecast data, allocation strategies, actual energy consumption, and device status. The middle layer contains cleansed key indicators such as daily energy costs, thermal storage module charge and discharge times, and temperature fluctuations. The top layer contains a scenario-based knowledge graph, including feature labels for typical operating scenarios such as winter peak and summer trough. The pattern recognition engine, based on unsupervised learning algorithms such as DBSCAN clustering, mines implicit correlation patterns from historical data. For example, it identifies thermal storage consumption patterns under conditions of continuous rainy weather and low night temperatures, or the cost optimization potential of combining charging during off-peak electricity prices and discharging during peak electricity prices. The parameter self-update mechanism automatically adjusts key parameters of the prediction and allocation decision submodules, such as the learning rate of the long-short-term memory network and the price correction coefficient for dynamic allocation, using a policy gradient algorithm from reinforcement learning to minimize comprehensive energy costs, thus avoiding the lag associated with manual parameter adjustment. The strategy iteration module triggers system-wide backtracking verification regularly, such as weekly, inputs historical operating data into the current optimal strategy, and calculates the deviation between theoretical energy consumption and actual energy consumption. If the deviation exceeds a threshold, such as 5%, the pattern recognition results are called to generate a new set of candidate strategies. The optimal strategy is selected through A / B testing to replace the old strategy, thereby achieving continuous evolution of the decision-making logic.

[0101] The heating output module is equipped with a heat exchanger, a circulating pump, a terminal heat dissipation device, and a flow control valve. The heat exchanger, consisting of a plate heat exchanger and a circulating water tank, transfers the heat energy released by the thermal storage module to the dedicated circulating water for heating. The circulating pump is installed on the main heating pipeline, providing power to drive the hot water through the system. The terminal heat dissipation device can include radiators, floor heating coils, or fan coils, depending on the building type (including residential, office, and public spaces), which release the hot water heat into the room through radiation or convection. Flow control valves are distributed in each branch pipeline. By adjusting the valve opening, the heat supply is balanced between different areas to ensure uniform indoor temperature.

[0102] The control system module is internally equipped with a central processing unit (CPU), communication interface, actuator drive unit, and safety protection unit. The CPU utilizes an industrial-grade PLC controller, integrating a multi-objective optimization algorithm and dynamic allocation strategy. It is responsible for receiving sensor data from each module and generating control commands. The communication interface supports Modbus, CAN bus, and 5G wireless protocols, enabling real-time data exchange with the solar collector module, heat storage module, heating module, and energy monitoring module. The actuator drive unit connects to the tracking bracket motor, circulation pump inverter, and flow control valve servo mechanism, converting control commands into mechanical actions. The safety protection unit integrates overtemperature protection, pressure alarm, and fault self-diagnosis functions. The overtemperature protection function cuts off circulation when the collector plate temperature exceeds 100°C, and the pressure alarm function activates pressure relief when the pipeline pressure exceeds 0.6MPa, ensuring safe system operation.

[0103] The energy monitoring and user interaction module is internally equipped with a data acquisition terminal, a real-time monitoring interface, a user command input module, and an abnormality alarm unit. The data acquisition terminal aggregates sensor data from each module, including collector temperature, heat storage capacity, and heating flow, and uploads this data to a cloud database on a minute-by-minute basis. The real-time monitoring interface displays the system's operating status in the form of a visual dashboard, showing solar irradiation curves, heat storage capacity trends, and heating temperature distribution in each area. The user command input module supports touch screen, mobile app, and voice interaction, allowing users to set target temperatures, select operating modes (economy, comfort, and energy-saving), or manually adjust flow distribution. The abnormality alarm unit triggers audible and visual alarms through threshold detection, including detection of heat storage temperatures below 20°C and circulation pump failure. These alarms are then sent to the user's mobile phone, prompting prompts for prompt troubleshooting.

[0104] In this invention, the AI ​​prediction submodule of the intelligent thermal energy allocation module significantly improves the accuracy of energy supply and demand matching. The system collects multiple data types in real time, including meteorological parameters, historical loads, and thermal storage status. Through feature extraction and multi-model prediction, it accurately predicts heating demand, solar input fluctuations, and thermal storage efficiency. Error calibration is then used to continuously refine the prediction results. This allows the system to know in advance how much heat will be needed and how much solar energy can provide, allowing for more rational energy flow coordination across heat collection, storage, and heating, reducing heat waste or insufficient supply.

[0105] In the present invention, the adaptive learning submodule effectively enhances the long-term operation capability of the system. The data sedimentation warehouse stores multi-layer data from raw logs to scenario knowledge. The pattern recognition engine can dig out hidden experiences such as the heat storage consumption rules during continuous rainy weather and low temperatures, and the cost optimization potential of charging at valley electricity prices. The parameter self-update mechanism automatically adjusts the key parameters of the prediction model and control strategy without frequent manual adjustment. The strategy iteration module verifies historical data every week. If it finds that the current strategy is less effective, a new strategy is generated and replaced through testing. These functions enable the system to achieve self-learning, which not only maintains operational stability, but also continuously optimizes energy efficiency as the external environment changes. It is more energy-saving, convenient, and quick to use in the long term.

[0106] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also includes elements that are inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0107] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A solar low calorific value heat storage heating system, characterized by: The system includes: a solar heat collection module, a low calorific value heat storage module, an intelligent heat energy allocation module, a heating output module, a control system module and an energy monitoring and user interaction module; The intelligent heat energy allocation module is internally configured with: an AI prediction submodule, a dynamic allocation decision submodule, a multi-objective optimization submodule, and an adaptive learning submodule; The heat transfer medium circulation loop outlet of the solar heat collection module is connected to the inlet end of the heat exchange coil of the low calorific value heat storage module through a pipeline; The outlet end of the heat exchange coil of the low calorific value heat storage module is connected to the inlet end of the heat exchange unit of the heating output module through a pipeline. The communication interface of the control system module is respectively connected to the temperature sensor data output end of the solar thermal collection module, the temperature sensor data output end of the low calorific value heat storage module, and the flow sensor data output end of the heating output module; The control signal output end of the actuator drive unit of the control system module is respectively connected to the motor control end of the solar thermal collector module tracking bracket, the inverter control end of the heating output module circulating pump and the servo mechanism control end of the flow control valve. The data acquisition terminal signal input terminal of the energy monitoring and user interaction module is connected in parallel with the sensor data output terminal of each module to summarize the operation data; The signal output end of the user command input module of the energy monitoring and user interaction module is connected to the command receiving end of the control system module through the communication interface, and the user setting is transmitted to the control system for execution; The prediction data output terminal of the intelligent thermal energy allocation module is connected to the algorithm input terminal of the control system module to provide a prediction basis for the generation of the control strategy.

2. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The solar heat collection module is internally provided with a heat collection panel array, a tracking bracket, a heat transfer medium circulation loop and a temperature sensor.

3. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The low calorific value heat storage module is internally provided with a heat storage material layer, a phase change container, a heat insulation shell and a heat exchange coil.

4. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The AI ​​prediction submodule is internally provided with a data acquisition layer, a feature extraction unit, a prediction model cluster and an error calibration module.

5. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The dynamic allocation decision submodule dynamically determines the allocation ratio of solar thermal energy between direct heating and thermal storage charging based on multi-source data; the module input includes the solar input prediction value P provided by the AI ​​prediction submodule sol , the heating demand forecast curve D(t) for the future period, and the real-time collected status of the heat storage module; The operation process of the module is divided into three key links: First is the basic demand matching, determining the initial allocation ratio α according to the real-time comparison between solar energy input and heating demand. base If the solar energy is sufficient, P sol ≥ D(t), then the current heating demand is preferentially satisfied, and the remaining thermal energy is used for heat storage; if the solar energy is insufficient, P sol < D(t)), then all solar energy is used for heating, and the insufficient part is supplemented by the heat storage module; Second is the economic correction, adjusting the allocation strategy in combination with the electricity price fluctuation: during the low electricity price period, ρ(t) is low, reducing the direct heating ratio and increasing the heat storage charging to utilize low-cost energy. When ρ(t) is high during peak electricity price period, the proportion of direct heating is increased to reduce the dependence on high-priced grid energy; Finally, the heat storage state is corrected and the charging behavior is optimized according to the characteristics of the heat storage material: If the heat storage temperature T store Close to the material phase transition temperature T phase , then reduce the charging speed to avoid heat loss caused by overcharging; If the temperature is too low, T store < <T phase , the charging ratio is increased to improve the heat storage efficiency; finally, the ratio α after comprehensive basic matching, economy and heat storage state correction is used as the output to guide the execution of heat energy distribution; in: The comprehensive correction formula for the dynamic allocation ratio is: in: α base Indicates the basic allocation ratio (determined by the matching relationship between solar energy and demand, range 0≤α base ≤1); β represents the electricity price correction coefficient; ρ(t) represents the real-time electricity price during the current period; ρth represents the electricity price threshold; γ represents the temperature correction coefficient; Tphase represents the phase change temperature reference value of the heat storage material; Tstore represents the current temperature of the thermal storage module (°C).

6. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The multi-objective optimization submodule is based on the initial allocation strategy output by the dynamic allocation decision submodule, takes user comfort as the bottom line constraint, and weighs the three key goals of energy cost, equipment life and system stability. It generates a comprehensive optimal heat allocation plan through quantitative analysis and dynamic weight mechanism. The quantitative expression of the comprehensive optimization objective of the multi-objective optimization submodule is: in: J represents the comprehensive optimization target value (the smaller the value, the better the solution); w cost ,w wear ,w stab are the dynamic weight coefficients of energy cost, equipment life and system stability (0 <w cost ,w wear ,w stab <1, and the sum is 1); C cost Indicates the total energy cost of the current period; C base represents the historical benchmark energy cost; ΔN represents the increment of the number of times the heat storage module is charged and discharged during the current period; N base Indicates the increment of historical benchmark charge and discharge times; ΔT store Indicates the temperature change of the heat storage module in the current period; T base Indicates the historical baseline temperature change.

7. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The self-adaptive learning submodule is internally provided with a data precipitation warehouse, a pattern recognition engine, a parameter self-update mechanism and a strategy iteration module.

8. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The heating output module is internally provided with a heat exchange unit, a circulation pump, a terminal heat dissipation device and a flow regulating valve.

9. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The control system module is internally provided with a central processing unit, a communication interface, an actuator drive unit and a safety protection unit.

10. The solar low calorific value heat storage heating system according to claim 1, characterized in that: The energy monitoring and user interaction module is internally provided with a data acquisition terminal, a real-time monitoring interface, a user instruction input module and an abnormal alarm unit.

Images