Solar water heating system control method and device, terminal and medium

Abstract

The invention provides a solar water heating system control method and device, a terminal and a medium. The solar water heating system control device comprises a solar water heater, a heat storage water tank, a heating pipeline and a circulating pump; the method comprises the steps that solar energy temperature information, heat storage water tank temperature information, heating pipeline temperature information, current time information and solar radiation intensity are obtained; according to the solar energy temperature information, the heat storage water tank temperature information and the heating pipeline temperature information, temperature fluctuation rate information is determined; and under the three operation conditions, a time control model, a temperature control method based on a PID algorithm and a combined control model are adopted to control hot water flowing among the solar water heater, the heat storage water tank and the heating pipeline. According to the solar water heater, intelligent control over the water temperature and the water amount is achieved, solar energy resources are fully utilized, other energy sources are saved, and the use experience of a user is improved.

Description

Technical Field

[0001] This invention relates to the field of solar heating technology, specifically to a control method, device, terminal, and medium for a solar water heating system. Background Technology

[0002] Solar water heaters are widely used in residential areas in my country. However, the hot water temperature varies with the weather, specifically different levels of solar radiation, making it impossible for users to set the temperature according to their needs. For example, in winter, when solar radiation is weak, solar water heaters cannot heat all the water in the storage tank to a suitable temperature, resulting in low-temperature hot water that fails to meet users' requirements. In summer, when solar radiation is very strong, solar energy can heat the water in the storage tank to over 90°C. However, excessively high temperatures pose certain safety hazards when used. Furthermore, in spring and autumn, the hot water temperature fluctuates with the weather, forcing users to either use lukewarm water or add cold water to achieve excessively high temperatures. Manually adjusting the amount of cold and hot water added can easily lead to water temperatures that are too high or too low, affecting user experience and hindering the full utilization of solar energy resources, energy conservation, and intelligent control of the amount of water added. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention proposes a control method, device, terminal, and medium for a solar water heating system, aiming to achieve intelligent control of water temperature and volume, fully utilize solar energy resources, save other energy sources, and improve the user experience.

[0004] In a first aspect, embodiments of this application provide a solar water heating system control method. The solar water heating system control method is applied to a solar water heating system control device, which includes a solar water heater, a hot water storage tank, heating pipes, and a circulating pump installed on the heating pipes. The solar water heating system control method includes: Acquire solar energy temperature information, hot water storage tank temperature information, heating pipe temperature information, current time information, and solar radiation intensity; Based on the solar energy temperature information, the hot water storage tank temperature information, and the heating pipe temperature information, temperature fluctuation information is determined; the temperature fluctuation information includes solar energy temperature fluctuation, hot water storage temperature fluctuation, and pipe temperature fluctuation. When the current time information is not during a peak period, the current solar radiation intensity is greater than or equal to a first intensity threshold, and any one of the temperature fluctuation rates is less than or equal to a first fluctuation threshold, the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes is controlled according to the time control model; the time control model is a model trained based on a neural network model. When the current time information is not during peak hours, the current solar radiation intensity is less than the second intensity threshold, and any one of the temperature fluctuation rates is greater than the second fluctuation threshold, the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes is controlled according to the temperature control method based on the PID algorithm. When the current time information is during a peak period, the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes is controlled according to the joint control model.

[0005] Optionally, controlling the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes according to the time control model includes: Obtain basic meteorological information and location latitude and longitude information; The basic meteorological information, the location latitude and longitude information, the current time information, and the solar radiation intensity are input into the pre-trained time control model to obtain the solar radiation prediction value and prediction confidence. The time control model is a hybrid neural network model with spatiotemporal awareness, which is formed by adding a dual-branch feature extraction module and a dynamic weight adjustment layer on the basis of the traditional 3-layer BP network. The flow of hot water between solar water heaters, hot water storage tanks, and heating pipes is controlled based on the predicted solar radiation value and the prediction confidence level.

[0006] Optionally, controlling the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes according to the temperature control method based on the PID algorithm includes: Obtain heating pipe settings; The temperature deviation is obtained based on the heating pipe temperature information and the heating pipe set value; The circulation pump flow rate is determined based on the PID algorithm and real-time temperature deviation. The flow of hot water from the heating pipe to the hot water storage pipe is controlled according to the flow rate of the circulating pump.

[0007] Optionally, controlling the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes according to the joint control model includes: Based on the solar radiation prediction and the temperature fluctuation information, time control weights and temperature control weights are obtained, and the time control weights and temperature control weights are the weights of the joint control model. Based on the heating pipe temperature information, the time control weight, and the temperature control weight, the multimodal control coefficients are obtained; The running time of the circulating pump is obtained based on the multimodal control coefficients; The operation of the circulating pump is controlled based on the circulating pump flow rate and the circulating pump operating time.

[0008] Optionally, obtaining the multimodal control coefficients based on the heating pipe temperature information, the time control weight, and the temperature control weight includes: When the current temperature of the heating pipes is less than the first temperature threshold; ; When the current temperature of the heating pipes is less than or equal to the second temperature threshold and greater than or equal to the first temperature threshold; ; When the current temperature of the heating pipes is greater than the second temperature threshold; ; in, These are the multimodal control coefficients; Time weighting; This refers to solar radiation values. This is the predicted average radiation value for the day; Temperature weighting; Real-time water temperature for solar water heaters; Real-time water temperature of the heating pipes; This refers to temperature deviation.

[0009] Secondly, embodiments of this application provide a solar water heating system control device, including a solar water heater, a hot water storage tank, heating pipes, and a circulation pump installed on the heating pipes, and further including: The information acquisition module is used to acquire solar energy temperature information, hot water storage tank temperature information, heating pipe temperature information, current time information, and solar radiation intensity. The volatility determination module is used to determine temperature volatility information based on the solar energy temperature information, the hot water storage tank temperature information, and the heating pipe temperature information; the temperature volatility information includes solar energy temperature volatility, hot water storage temperature volatility, and pipe temperature volatility. The first operating module is used to control the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes according to a time control model when the current time information is not during a peak period, the current solar radiation intensity is greater than or equal to a first intensity threshold, and any one of the temperature fluctuation rates is less than or equal to a first fluctuation threshold; the time control model is a model trained based on a neural network model. The second operation module is used to control the flow of hot water between the solar water heater, the hot water storage tank and the heating pipes according to the temperature control method based on the PID algorithm when the current time information is not in a peak period, the current solar radiation intensity is less than the second intensity threshold, and any one of the temperature fluctuation rates in the temperature fluctuation rate information is greater than the second fluctuation threshold. The third operation module is used to control the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes according to the joint control model when the current time information is in a peak period.

[0010] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the solar water heating system control method as described in any one of the first aspects above.

[0011] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the solar water heating system control method as described in any one of the first aspects above.

[0012] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the solar water heating system control method described in any one of the first aspects.

[0013] In this embodiment, solar energy temperature information, hot water storage tank temperature information, heating pipe temperature information, current time information, and solar radiation intensity are acquired. Temperature fluctuation rate information is determined based on the solar energy temperature information, the hot water storage tank temperature information, and the heating pipe temperature information. Under three operating conditions, a time-based control model, a PID algorithm-based temperature control method, and a combined control model are used to control the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes, respectively. This achieves intelligent control of water temperature and volume, fully utilizes solar energy resources, saves other energy sources, and improves the user experience. Attached Figure Description

[0014] Figure 1 This is a schematic flowchart of the first embodiment of the control of a solar water heating system provided in this application; Figure 2 This is a schematic diagram of the structure of the solar water heating system control device provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation

[0015] Figure 1 The diagram illustrates a first embodiment of the solar water heating system control method provided in this application. This is intended as an example and not a limitation. The solar water heating system control method is applied to a solar water heating system control device, which includes a solar water heater, a hot water storage tank, heating pipes, and a circulating pump installed on the heating pipes. This method can be applied to the aforementioned solar water heating system control device. Figure 1 As shown, the method may include: S10, acquires solar energy temperature information, hot water storage tank temperature information, heating pipe temperature information, current time information, and solar radiation intensity; To achieve intelligent control of water temperature and volume, fully utilize solar energy resources, conserve other energy sources, and improve user experience, the solar water heating system control device acquires solar energy temperature information, hot water storage tank temperature information, heating pipe temperature information, current time information, and solar radiation intensity. Specifically, the solar energy temperature information is the water temperature of the solar water heater over a given time period, obtained from a temperature sensor installed on the solar water heater; the hot water storage tank temperature information is the water temperature of the hot water storage tank over a given time period, obtained from a temperature sensor installed on the heating pipes; and the heating pipe temperature information is the water temperature of the heating pipes over a given time period, obtained from a temperature sensor installed on the heating pipes. The current time information is the solar radiation intensity information acquired over a given time period.

[0016] S20, based on the solar energy temperature information, the hot water storage tank temperature information, and the heating pipe temperature information, determine the temperature fluctuation rate information, which includes the solar energy temperature fluctuation rate, the hot water storage tank temperature fluctuation rate, and the pipe temperature fluctuation rate. After acquiring solar energy temperature information, hot water storage tank temperature information, and heating pipe temperature information, the solar water heating system control device determines temperature fluctuation rate information based on these information. The temperature fluctuation rate information includes solar energy temperature fluctuation rate, hot water storage tank temperature fluctuation rate, and pipe temperature fluctuation rate.

[0017] As one implementation method, determining the temperature fluctuation rate information based on the solar energy temperature information, the hot water storage tank temperature information, and the heating pipe temperature information may include: A1. Based on the solar energy temperature information, obtain the solar energy starting temperature, the highest solar energy temperature, and the lowest solar energy temperature; After acquiring solar temperature information, the solar water heating system control device obtains the solar initial temperature, the solar maximum temperature, and the solar minimum temperature based on the solar temperature information. The solar initial temperature, the solar maximum temperature, and the solar minimum temperature are the initial temperature, the maximum temperature, and the minimum temperature of the solar water heater within a preset time range. A2, based on the temperature information of the hot water storage tank, obtain the initial temperature of heat storage, the highest temperature of heat storage, and the lowest temperature of heat storage; After acquiring the temperature information of the hot water storage tank, the solar water heating system control device obtains the initial temperature of heat storage, the highest temperature of heat storage, and the lowest temperature of heat storage based on the temperature information of the hot water storage tank. The initial temperature of heat storage, the highest temperature of heat storage, and the lowest temperature of solar water heating are the initial temperature, the highest temperature, and the lowest temperature of the hot water storage tank within a preset time range. A3, based on the heating pipe temperature information, obtain the pipe's initial temperature, maximum temperature, and minimum temperature; After acquiring the temperature information of the heating pipes, the solar water heating system control device obtains the initial temperature, maximum temperature and minimum temperature of the pipes based on the heating pipe temperature information. The initial temperature, maximum temperature and minimum temperature of the pipes are the initial temperature, maximum temperature and minimum temperature of the heating pipes within a preset time range. A4. Based on the solar energy starting temperature, the highest solar energy temperature, and the lowest solar energy temperature, the solar energy temperature fluctuation rate is obtained. After obtaining the solar initial temperature, the highest solar temperature, and the lowest solar temperature, the solar water heating system control device obtains the solar temperature fluctuation rate based on the solar initial temperature, the highest solar temperature, and the lowest solar temperature. A5. Based on the thermal storage starting temperature, the highest thermal storage temperature, and the lowest thermal storage temperature, the thermal storage temperature fluctuation rate is obtained. After obtaining the initial temperature, maximum temperature and minimum temperature of the thermal storage, the control device of the solar water heating system obtains the thermal storage temperature fluctuation rate based on the initial temperature, maximum temperature and minimum temperature of the thermal storage. A6. Based on the initial temperature, maximum temperature, and minimum temperature of the pipeline, the pipeline temperature fluctuation rate is obtained.

[0018] After obtaining the initial temperature, maximum temperature, and minimum temperature of the pipeline, the solar water heating system control device obtains the pipeline temperature fluctuation rate based on these temperatures.

[0019] After obtaining the solar energy temperature fluctuation rate, thermal storage temperature fluctuation rate, and pipe temperature fluctuation rate, the solar water heating system control device dynamically switches the trigger conditions for these three factors through "operating condition thresholds." For example, 1. When the solar radiation intensity is ≥500W / m²... 2And remain stable (fluctuation ≤10% within 1 hour), automatically switching to set time control; 2. When solar radiation intensity <300W / m 2 If the fluctuation exceeds 20% within 1 hour, it will automatically switch to temperature control; 3. When it is during the peak energy consumption period (such as 6:00-8:00, 18:00-21:00), regardless of the lighting conditions, joint control will be forcibly activated (prioritizing user experience).

[0020] S30, when the current time information is not during a peak period, the current solar radiation intensity is greater than or equal to a first intensity threshold, and any one of the temperature fluctuation rates in the temperature fluctuation rate information is less than or equal to a first fluctuation threshold, the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes is controlled according to the time control model; the time control model is a model trained based on a neural network model; After acquiring current time information, solar radiation intensity, and temperature fluctuation rate, the solar water heating system control device, based on a time control model, controls the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes when the current time is not during peak hours, the current solar radiation intensity is greater than or equal to a first intensity threshold, and any one of the temperature fluctuation rates is less than or equal to a first fluctuation threshold. The time control model is a model trained based on a neural network model. Peak hours are 6:00-8:00 and 18:00-21:00. The first intensity threshold is 500W / m². 2 The first fluctuation threshold is 10%.

[0021] As one implementation method, the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes is controlled according to a time control model; the time control model is a model trained based on a neural network model, specifically including: B1, to obtain basic meteorological information and location latitude and longitude information; After obtaining the solar energy temperature fluctuation rate, the thermal storage temperature fluctuation rate, and the pipeline temperature fluctuation rate, the solar water heating system control device acquires basic meteorological information and location latitude and longitude information; wherein, the basic meteorological information includes real-time air temperature, cloud cover rate, and wind speed, and the location latitude and longitude information is a latitude and longitude correction coefficient.

[0022] B2, input the basic meteorological information, the location latitude and longitude information, the current time information, and the solar radiation intensity into the pre-trained time control model to obtain the solar radiation prediction value and prediction confidence level; After obtaining the solar temperature fluctuation rate, the thermal storage temperature fluctuation rate, and the pipe temperature fluctuation rate, the solar water heating system control device inputs the basic meteorological information, the location latitude and longitude information, the current time information, and the solar radiation intensity into the pre-trained time control model to obtain the solar radiation prediction value and prediction confidence level. The time-controlled model is a hybrid neural network model with spatiotemporal awareness, formed by adding a dual-branch feature extraction module and a dynamic weight adjustment layer to the traditional 3-layer BP network. Current time information includes time encoding, and solar radiation intensity includes radiation values ​​from the previous 3 hours.

[0023] The time control model structure is as follows: Input layer: Expanded into 8-dimensional feature vectors: Includes basic meteorological features (real-time temperature) Cloud cover rate (C), wind speed (v), historical time series characteristics (solar radiation value in the previous 3 hours). , , Spatial features (latitude and longitude correction factor L), time coding (sine and cosine coding) ).

[0024] The feature extraction layer includes temporal branches and spatial branches. The temporal branch uses a 1D convolutional layer (3 convolutional kernels, kernel size 3) to extract the temporal correlation of radiation values, and the output feature is Ft; the spatial branch processes meteorological and spatial features through a fully connected layer (16 neurons), and the output feature is Fs.

[0025] Attention fusion layer: Calculates the dynamic weights of temporal and spatial features. ; ; in, This is a time-series feature vector, a 1×16 dimensional vector, extracted from historical solar radiation time-series features by a 1D convolutional layer, containing radiation values ​​from the previous 3 hours. , , The correlation information reflects the energy supply pattern in the "time dimension"; This is a spatial feature vector, a 1×16 dimensional vector, composed of meteorological and spatial features output from a fully connected layer, including real-time temperature. Cloud coverage C, wind speed v, and latitude and longitude correction coefficient L reflect the environmental influencing factors of "spatial dimension". The time-series feature weight matrix is ​​a 16×1 dimensional matrix with learnable parameters. It is optimized through model training and used to adjust the importance of each element in the time-series feature vector (such as assigning higher weights to recent radiation values) to ensure the adaptability of time-series features to control targets. The spatial feature weight matrix is ​​a 16×1 dimensional matrix with learnable parameters. Through training and optimization, it is used to strengthen the contribution of key factors in spatial features (such as the influence of cloud cover C on radiation having a higher weight than wind speed v), thereby enhancing the guiding value of spatial features. These are the dynamic weighting coefficients for temporal features, representing the proportion of temporal features in the fusion result, determined by... and The inner product is calculated using the exp function; the larger the value, the stronger the influence of time series characteristics on time control decisions. The dynamic weighting coefficients for spatial features represent the proportion of spatial features in the fusion result, and are related to... Complementary; the larger the value, the stronger the influence of spatial characteristics on time control decisions.

[0026] in , For learnable weight matrices, features are fused. .

[0027] Output layer: Employs a dual-output structure; main output: predicted solar radiation for the next hour. ; Auxiliary output: Prediction confidence (Between 0 and 1) is used to determine the reliability of subsequent control logic.

[0028] 2. Improved loss function and training strategy To improve the model's prediction accuracy under extreme weather conditions, a weighted mixture loss function is designed: in: (Mean squared error, basic fit loss) (Mean absolute error, reducing extreme value bias) (Confidence loss, where Ptrue is the confidence label corresponding to the actual error) Weighting coefficients: , , ; The training process employs phased learning rate scheduling: First 50 rounds: learning rate η=0.02, fast convergence Rounds 51-150: η=0.005, fine-tuning Rounds 151-300: η=0.001, freeze convolutional layer weights, optimize only the attention layer. 3. Optimization of dynamic time control formula Based on the prediction results and confidence level Design a dual-threshold adaptive control logic: (1) Cyclic start condition: and Among them, dynamic threshold Adaptive adjustment over time: (By utilizing the properties of the cosine function, the noon threshold is made higher than that of the morning and evening to match the actual solar radiation pattern.) (2) Calculation of cycle interval time Δt: Introducing a confidence correction term shortens the interval when prediction reliability is high, thereby improving energy efficiency: (minute) Constraints: (Avoid energy waste caused by excessively short cycles) (3) Exception handling mechanism: when When the time comes, automatically switch to historical mean correction mode: in This represents the real-time average radiation value for the day. This is the average radiation value for the same period over the past 30 days.

[0029] The dual-branch feature extraction module is applied in the "raw data preprocessing - feature separation and extraction" stage after the model input layer and before feature fusion. Its core function is to accurately extract features from two types of core data (temporal series and spatial environment types) of the solar-heating linkage system using a differentiated network structure, avoiding feature confusion caused by the traditional BP network's "single-channel processing of all data". The specific application method is as follows: 1. Application Scenario Definition When the system needs to process two types of heterogeneous inputs, namely "historical time series data" and "real-time spatial environment data", it triggers a dual-branch parallel extraction: Time-series data: solar radiation intensity sequence for the previous 12 hours (10 min / time, 72 data points in total), temperature change curve of hot water storage tank / heating pipe (5 min / time, 144 data points in total), historical start and stop records of circulating pump (time stamp + running time). Spatial environment data: real-time latitude and longitude correction values ​​(adapting to solar altitude angles in different regions), solar collector tilt angle / light-collecting area, outdoor wind speed / humidity / cloud coverage, and pipe insulation layer thickness (static parameters, fixed during model training and called during inference).

[0030] 2. The specific application steps are shown in Table 1. Table 1 3. Application Advantages To address the shortcomings of traditional BP networks in inputting temporal and spatial data—for example, traditional networks may mistakenly treat "high cloud layer (spatial data)" and "low radiation (temporal data)" as the same feature—the dual-branch approach, through its differentiated structure, improves the accuracy of temporal feature extraction by 40% and the signal-to-noise ratio of spatial features by 35%.

[0031] The application scenarios and implementation steps of the dynamic weight adjustment layer are as follows: This layer is applied to the "feature fusion" stage after the dual-branch feature extraction is completed and before the mixed features are input into the hidden layer of the traditional BP network. The core is to dynamically allocate features based on real-time operating conditions. and The weighting ratio is adjusted to avoid "feature importance mismatch" caused by fixed weights.

[0032] 1. Application Scenario Definition Dynamic weight adjustment is triggered when the system meets any of the following conditions: If the real-time solar radiation intensity deviates from the predicted value by more than 20% (e.g., if a clear day suddenly turns cloudy, the reliability of the temporal features decreases, and the weight of spatial features needs to be increased); Sudden changes in outdoor temperature >5℃ / h (the impact of the spatial environment intensifies, and the contribution of spatial characteristics needs to be strengthened); The system switches operating modes (e.g., from "hot water storage tank power supply" to "heating pipe power supply", and the characteristic demand changes from "time-series heat accumulation" to "spatial heat transfer efficiency").

[0033] 2. Specific application steps Weight calculation trigger: After the dual-branch outputs Ft and Fs, the "feature importance score" of both is calculated first. Time series feature scoring ( It is a 16×1 learnable weight matrix. (This is the bias term, optimized during training using cross-entropy loss). Spatial feature scoring ( Similarly, (for spatial feature bias).

[0034] Dynamic weight generation: Weight coefficients are calculated using an improved Softmax function. ; ; in (Time-series characteristic adjustment factor) (Spatial characteristic adjustment factor) is dynamically corrected based on real-time operating conditions: Sunny day (radiation > 500W / m) 2 ): =1.2, =0.8, Rise to 0.7-0.8; Cloudy day (radiation <200W / m) 2 ): =0.7, =1.3, It rose to 0.6-0.7.

[0035] Feature fusion and transfer: weighted fused features The first hidden layer (24 neurons) of a traditional 3-layer BP network is input, and the parameters are then optimized according to the backpropagation logic of the BP network. Finally, control commands such as the start time and runtime of the circulation pump are output.

[0036] 3. Application Examples Scene: Sunny noon (radiation 650W / m) 2 Suddenly encountered a brief period of dark clouds (cloud coverage suddenly increased to 70%). Dynamic adjustment process: Before the dark clouds: =1.8, =0.9, =0.75, =0.25 (based on time-series feature prediction of "radiation peak lasting 1 hour"); Behind the clouds: Real-time radiation was detected to have dropped to 300W / m². 2 , It dropped to 0.6. Rising to 1.4 It dropped to 0.9. Rise to 1.5 Adjusted to 0.3. Adjusted to 0.7 (focusing on spatial characteristics "cloud cover leads to reduced heat gain", and promptly shortening the running time of the circulating pump).

[0037] Through this layer of dynamic adaptation, the model time control error is reduced from ±12 min to ±4 min in the traditional BP network, and the response speed is improved by 60% when the operating conditions change abruptly.

[0038] An attention mechanism is introduced to achieve dynamic fusion of spatiotemporal features, addressing the insufficient processing capability of traditional BP networks for multi-dimensional features. A dual-output structure and hybrid loss function are designed to improve prediction robustness under extreme weather conditions. Dynamic threshold and confidence correction control logic makes time scheduling more closely match actual energy supply characteristics. Through these improvements, the model's prediction accuracy is improved by 18% (RMSE reduced to 45W / m). 2 (The following), the system's energy utilization efficiency is improved by 12%.

[0039] B3 controls the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes based on the predicted solar radiation value and the prediction confidence level.

[0040] After obtaining the predicted solar radiation value and prediction confidence level, the solar water heating system control device controls the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes based on the predicted solar radiation value and prediction confidence level.

[0041] The time setting is based on the hot and cold water exchange rules. 1.1 The core basis for setting the time (based on the output of the improved neural network) Setting a time requires meeting two core conditions: Condition 1: The neural network predicts the solar radiation value for the next hour. (Ensure the solar water heater has sufficient heat gain), and confidence level (To ensure the reliability of the prediction); Condition 2: Match user water / heating demand patterns (learn user water usage times in the past 30 days through neural networks, such as 6-8 am and 6-9 pm being peak water usage times, and 12-2 pm being a low heating demand period).

[0042] 1.2 Specific time control and interchange directions are shown in Table 2. Table 2 A specific example is as follows: A residential building in northern China, located at latitude 40°N, experiences an average daily temperature of -5 to 8°C during the heating season. Morning preparation phase (6:30-8:00): Neural network predicts the time period from 6:30 to 7:30. , The startup conditions are met; The first exchange is triggered at 7:00: solar water heater (water temperature 52℃) → hot water storage tank (water temperature 28℃). Cold water flows into the solar water heater from the bottom of the hot water storage tank, and hot water flows into the hot water storage tank from the top of the solar water heater. After 30 minutes, the water temperature in the hot water storage tank rises to 45℃, which meets the water needs for washing up in the morning.

[0043] Daytime heating period (9:00-11:00): Predicted timeframe: 9:00-10:00 , And the indoor temperature dropped to 18℃ (set threshold 20℃). At 9:00, the second exchange is triggered: solar water heater (water temperature 60℃) → heating pipe (water temperature 16℃). Hot water is sent from the outlet of the solar water heater to the inlet of the heating pipe through the circulation pump. The cold water (16℃) at the return end of the heating pipe flows back to the inlet of the solar water heater. After 1 hour, the indoor temperature rises to 20℃ and the exchange stops automatically.

[0044] Nighttime reheating phase (21:30-22:30): Predicted timeframe: 21:30-22:30 (No solar radiation) The indoor temperature dropped to 17℃; At 21:30, the third exchange was triggered: hot water storage tank (water temperature 50℃) → heating pipe (water temperature 15℃). Hot water flowed out from the middle of the hot water storage tank (avoiding cold water at the bottom) and was sent into the heating pipe by the circulation pump. The cold water in the heating pipe flowed back to the bottom of the hot water storage tank. After 40 minutes, the indoor temperature rose to 19℃, and the exchange stopped.

[0045] S40, when the current time information is not during a peak period, the current solar radiation intensity is less than the second intensity threshold, and any one of the temperature fluctuation rates in the temperature fluctuation rate information is greater than the second fluctuation threshold, the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes is controlled according to the temperature control method based on the PID algorithm. After acquiring current time information, solar radiation intensity, and temperature fluctuation rate information, the solar water heating system control device, when the current time is not during peak hours, the current solar radiation intensity is less than a second intensity threshold, or any one of the temperature fluctuation rates is greater than a second fluctuation threshold, controls the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes using a PID algorithm-based temperature control method. Peak hours are 6:00-8:00 and 18:00-21:00. The second intensity threshold is 300W / m². 2 The first fluctuation threshold is 20%.

[0046] As one implementation method, the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes is controlled according to a temperature control method based on a PID algorithm, including: C1, retrieves the heating pipe settings; After acquiring the current time information, solar radiation intensity, and temperature fluctuation rate, the solar water heating system control device obtains the set value of the heating pipes.

[0047] C2, based on the temperature information of the heating pipes and the set value of the heating pipes, the temperature deviation is obtained; After acquiring the set value of the heating pipe, the solar water heating system control device obtains the temperature deviation based on the heating pipe temperature information and the heating pipe set value.

[0048] C3 determines the circulating pump flow rate based on the PID algorithm and real-time temperature deviation; After obtaining the real-time temperature deviation, the solar water heating system control device determines the circulation pump flow rate based on the PID algorithm and the real-time temperature deviation.

[0049] The temperature control target is to keep the temperature of the heating pipes... Stabilize at the set value (Default 22℃), the control quantity is the circulating water pump flow rate Q (0-5m³ / h). 3 / h), core formula: in, This refers to temperature deviation.

[0050] When correcting fuzzy logic parameters, a two-dimensional fuzzy controller is designed, with the input being the deviation. and rate of change of deviation The output is the parameter correction amount. , , The fuzzy rules are as follows: when (Large deviation): Increase (Accelerate response), reduce (Avoid overshooting); when (medium deviation): medium Moderate ,Increase (Suppress oscillations); when (Small deviation): Reduce Increase (Eliminate steady-state error); Parameter correction formula: in, Initial parameters In anti-integral saturation and filtering processing: Points limit: When When the flow rate is at its maximum, the accumulation of points is paused.

[0051] Differential filtering: Performing a first-order low-pass filter on the differential term, with a cutoff frequency of... : in, This is the filtering time constant.

[0052] To make PID control more closely match the physical characteristics of the system, a temperature change rate prediction correction term is introduced, and the theoretical heating rate is calculated based on the heat transfer model: in The mass of water in the pipe is ρ = 1000 kg / m. 3 This is the density of water.

[0053] The deviation between theoretical and actual values As an auxiliary input to the fuzzy controller, Kd is dynamically adjusted: When the actual temperature rise is slower than the theoretical value, the differential gain is increased to enhance the regulation.

[0054] in, The theoretical heating rate of the heating pipes is the theoretical increase in water temperature per unit time calculated based on the heat transfer model. The value range is 0.2-1.5℃ / min; As a correction benchmark for the PID differential gain Kd, it is compared with the actual heating rate to determine whether the heat transfer meets the standard. The heat absorption efficiency of a solar water heater is the efficiency coefficient (including optical efficiency and heat loss correction) of converting solar radiation energy into water heat energy. The value range of the heat absorption efficiency of a solar water heater is 0.75-0.85. The heat absorption efficiency of a solar water heater reflects the solar heat collection capacity. The higher the efficiency, the higher the theoretical heating rate. The light-receiving area of ​​a solar water heater; the effective area of ​​the solar collector receiving solar radiation; the light-receiving area of ​​a solar water heater is 1.5-3.0 m². 2 The larger the solar water heater's light-receiving area, the more solar radiation it receives per unit time, directly affecting the heat input in the first term of the formula. Real-time solar radiation intensity; it is the solar radiation energy (including direct and diffuse radiation) received per unit area per unit time; the range of real-time solar radiation intensity is 100-1000 W / m². 2(During the day); Real-time dynamic input parameter of solar radiation intensity. R is high on sunny days and low on cloudy days, which directly determines the amount of heat contributed by solar energy. This refers to the real-time flow rate of the circulating water pump; it is the core control variable output by the PID algorithm, namely the circulating water flow rate between the hot water storage tank and the heating pipes; the real-time flow rate of the circulating water pump ranges from 1.5 to 5.0 m³ / h. 3 / h; The real-time flow rate of the circulating water pump regulates the heat transfer rate. The larger the Q value, the more heat the hot water storage tank transfers to the heating pipes. This is the specific heat capacity of water. This is the density of water. Real-time water temperature for solar water heaters; This refers to the real-time water temperature of the heating pipes, which ranges from 35 to 50 degrees Celsius. Total mass of circulating water in the heating pipes.

[0055] in, This represents the heat obtained by a solar water heater through radiative heat exchange. This represents the heat transferred from the hot water storage tank (indirectly from solar energy) to the heating pipes via convection heat exchange in the circulating water. (Denominator) This represents the total heat capacity of the heating pipe system. The larger the total heat capacity, the lower the theoretical heating rate for the same heat input. Using this formula, the PID algorithm can better reflect the actual heat transfer characteristics, avoiding control lag caused by relying solely on temperature deviations.

[0056] The optimized control flow of the PID algorithm consists of three steps: First, sensors collect parameters such as Ts, Tp, and R, and calculate the deviation e(t) and rate of change ec(t). Second, the fuzzy controller outputs parameter correction values ​​to update the control flow. , , After anti-integral saturation and filtering, the pump flow command Q(t) is calculated.

[0057] After the improvement, the steady-state error of temperature control accuracy was reduced from ±1.5℃ to ±0.5℃; the response time to reach the set value was shortened from 300s to 180s; and the anti-interference capability was improved, with the overshoot reduced from 8℃ to 3℃ when radiation intensity changed abruptly (such as when cloud cover occurred).

[0058] This application solves the nonlinear control problem of solar energy systems by integrating fuzzy logic and PID; it introduces a heat transfer theory correction term to achieve deep coupling between the algorithm and the physical model; and it improves the stability of the system under extreme conditions by anti-integral saturation and filtering design. Compared with traditional heat transfer algorithms, this method improves the control accuracy by 60% and reduces energy waste by 15%, and is more suitable for dynamic temperature regulation of multi-energy complementary systems.

[0059] C4, the flow of hot water from the heating pipe to the hot water storage pipe is controlled according to the flow rate of the circulating pump.

[0060] After obtaining the flow rate of the circulating pump, the solar water heating system control device controls the flow of hot water from the heating pipe to the hot water storage pipe based on the flow rate of the circulating pump.

[0061] The water flow circulation control logic based on temperature difference is as follows: Solar water heaters ( ), hot water storage tank ( ), heating pipes ( The temperature difference among the three factors is used as the core input, and the flow rate of the circulating water pump is dynamically adjusted through an improved PID algorithm. This enables precise control of water flow circulation, with the following specific rules: The circulation control of the solar water heater and its storage tank is as follows: Startup conditions: When (Set temperature difference threshold, default 5℃), and the deviation of PID calculation. ( Set the temperature for the hot water storage tank (default 50℃). When the temperature is ≥2℃, the cycle will be triggered.

[0062] Flow regulation logic: via PID formula .

[0063] Calculate the flow rate, where: if (Large temperature difference): Fuzzy controller increases (To 3.0), Reduce (Up to 0.05), increase up to 3-5m 3 / h, to accelerate hot water delivery; if (Medium temperature difference): Maintain , , Stable at 2-3 ;like (Small temperature difference): Reduce (Up to 2.0), Increase (to 0.15), Reduced to 1-2 To avoid invalid loops. Stopping condition: when... achieve ,or At that time, PID output The loop stops.

[0064] The circulation control of the hot water storage tank and heating pipes is as follows: Startup conditions: When (Default 8℃), and ( Set the temperature for the heating pipes (default 45℃). When the temperature reaches ≥3℃, start the circulation.

[0065] Flow regulation logic: Combining the heat transfer coupling mechanism, if the theoretical heating rate... With reality Deviation; if (Actual temperature rise is slower than theoretical), through Increase the differential gain, improve Up to 2.5-4 Enhance heat transfer; if (Actual temperature rise is too fast), reduce , Reduced to 1.5-2.5 ,avoid Overshoot.

[0066] Stop condition: when achieve ,or When the loop stops, the cycle ends.

[0067] Taking a typical December day in a northern residence during winter as an example: Solar water heater: 2m² light-receiving area 2 Ts monitoring interval is 10 min; Hot water storage tank: 200L capacity Time slot settings: 6-8 AM 55℃, 8-6 PM 48℃, 6-10 PM 52℃, 10 PM-6 AM 45℃; Heating pipes: Time slot settings: 6-8 AM 40℃, 8-6 PM 45℃, 6-10 PM 48℃, 10 PM-6 AM 42℃; PID initial parameters: .

[0068] Specific control process (1) 7:00-7:30 AM (peak water usage + temperature difference trigger) Sensor data: , , ; PID control: Fuzzy controller determines "large deviation + morning period". Revised to version 3.0 Corrected to 0.05, the calculated result is... ; Cyclic operation: Solar water heater → hot water storage tank cycle starts, after 30 minutes Rise to 55℃ When the PID output Q=0, the loop stops.

[0069] (2) 13:00-13:40 (Heating demand + time compensation) Sensor data: , ,but, Furthermore, the time period is "low demand for heating during midday"; PID control: Time compensation term makes The temperature was reduced to 0.45, and combined with ΔT˙ = 0.3℃ / min (the actual temperature rise is close to the theoretical value), the calculation yielded... ; Cycle operation: Hot water storage tank → heating pipe circulation starts, after 40 minutes Rise to 45℃ The loop stops.

[0070] (3) 20:00-20:50 (Double peak + anti-integral saturation) Sensor data: , ,but, The time period is characterized by "two peaks"; PID control: Integral term Accumulated temperature reaches 18°C·min (close to the nighttime limit of 20°C·min), triggering anti-integral saturation. (Not reached) ); Cyclic operation: Hot water storage tank → heating pipes + solar water heater → hot water storage tank dual circulation start, after 50 minutes Rising to 48℃ Maintain at 52℃, then stop the dual-cycle operation.

[0071] S50, when the current time information is during a peak period, the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes is controlled according to the joint control model.

[0072] After acquiring the current time information, solar radiation intensity, and temperature fluctuation rate, the solar water heating system control device controls the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes according to the joint control model when the current time information is during a peak period.

[0073] As one implementation method, controlling the hot water flow between the solar water heater, the hot water storage tank, and the heating pipes according to the joint control model specifically includes: D1, based on the predicted solar radiation value and the temperature fluctuation rate information, the time control weight and temperature control weight are obtained, and the time control weight and temperature control weight are the weights of the joint control model; After acquiring the current time information, solar radiation intensity, and temperature fluctuation rate, the solar water heating system control device obtains time control weights and temperature control weights based on the predicted solar radiation value and the temperature fluctuation rate information. The time control weights and temperature control weights are the weights of the joint control model. The dynamic weights of the improved joint control model are calculated as follows: Based on solar radiation predictions and temperature deviation Design a bivariate weighting function: ; ; Among them, when the light is sufficient ( And when the deviation is small, ≈0.8 (time-dominated); when light is weak ( And when the deviation is large, ≈0.3 (temperature-dominated); the weight changes smoothly in the transition range to avoid system oscillations caused by sudden changes.

[0074] D2, based on the heating pipe temperature information, the time control weight, and the temperature control weight, the multimodal control coefficients are obtained; After obtaining the heating pipe temperature information, the time control weight, and the temperature control weight, the solar water heating system control device obtains the multimodal control coefficients based on the heating pipe temperature information, the time control weight, and the temperature control weight. As one implementation method, the control scenarios are divided into three categories, and multimodal control coefficients are calculated using differentiation coefficients. Specifically, the multimodal control coefficients are obtained based on the heating pipe temperature information, the time control weight, and the temperature control weight, including: E1, when in heating mode ( When the current temperature of the heating pipes is lower than the first temperature threshold, solar energy is used first to enhance the radiation correlation of the time coefficient. ; E2, when in heat preservation mode ( When the current temperature of the heating pipe is less than or equal to the second temperature threshold and greater than or equal to the first temperature threshold, energy consumption and stability are balanced, and a hysteresis coefficient is introduced to suppress frequent actions. ; E3, when in heating mode ( (or the temperature of the hot water storage tank is ≥55℃). When the current temperature of the heating pipes is greater than the second temperature threshold, the auxiliary heat source restriction is activated and an energy consumption penalty is added. ; in, These are the multimodal control coefficients; Time weighting; This refers to solar radiation values. This is the predicted average radiation value for the day; Temperature weighting; Real-time water temperature for solar water heaters; Real-time water temperature of the heating pipes; This refers to temperature deviation.

[0075] in, To reflect temperature deviation The core parameters of direction, ; ; ; in , Set the temperature for the heating pipes (default 22℃). This is the real-time temperature of the heating pipes.

[0076] when : Indicates the real-time temperature of the heating pipes (Temperature is too low), heat needs to be added. At this time, the energy consumption penalty is 1-0.3×1=0.7, and the control coefficient K(t) is calculated at 70%. While ensuring the heat replenishment needs, excessive energy consumption should be avoided (e.g., limiting the full-load operation of the circulating pump).

[0077] when :express (Temperature met), no additional heating required. At this point, the penalty is 1 - 0.3 × 0 = 1, but due to the conditions for triggering the additional heating mode (… If the hot water storage tank temperature (≥55℃) is no longer met, the system will automatically switch to heat preservation mode to avoid... Abnormally high.

[0078] when :express (Temperature too high), supplemental heating needs to be suppressed. The penalty at this point is 1 - 0.3 × (-1) = 1.3. However, considering the core objective of the supplemental heating mode, "activate auxiliary heat source limitation," the system will further trigger... Upper limit constraint (not exceeding the maximum value under insulation mode) (Value), while reducing the running time of the circulating pump to prevent the temperature from continuing to rise.

[0079] In heating mode, it efficiently utilizes solar energy and reduces auxiliary energy consumption: through... in the formula , The term strongly correlates time weighting with solar radiation prediction—when Higher than (For example, at noon on a sunny day), the time-weighted contribution coefficient increases to 60%-70%, prioritizing the high-frequency operation of the circulation pump to quickly deliver the high-temperature hot water from the solar water heater to the heating pipes. This reduces the frequency of activation of auxiliary heat sources such as electric auxiliary heating, resulting in 25%-30% energy savings compared to the traditional fixed-coefficient mode. It also shortens the heating time and improves the user experience: (The formula...) The effect of enhanced temperature deviation on the control coefficient, when When there is a large temperature difference, the coefficient of temperature weight contribution can be increased to 0.8-1.2, and the circulation pump flow rate increases simultaneously, which shortens the time for the heating pipe water temperature to rise from 10℃ to 18℃ from 40 minutes in the traditional mode to 25 minutes, thus avoiding heating delays during low-temperature periods.

[0080] When in heat preservation mode, frequent start-stop cycles are suppressed, extending equipment lifespan and introducing [a certain feature]. Hysteresis coefficient – ​​when temperature deviation When the deviation is small, this coefficient drops to 0.3-0.5, which is the control coefficient. Synchronous reduction resulted in a decrease in circulating pump runtime from 10 minutes to 5-6 minutes, and an increase in start-stop interval from 15 minutes to 30 minutes. The average daily start-stop frequency of the circulating pump decreased from 40 times to 20 times, reducing equipment wear by 40%. Maintaining stable temperature and avoiding energy waste, while balancing the contributions of time and temperature weights, [the text abruptly ends here]. At that time, time weighting dominates (accounting for 60%-70%), and it operates according to a preset time period pattern; when At the same time, the temperature weighting ratio is increased to 50%-60%, and the running time is dynamically adjusted to reduce the fluctuation range of water temperature in heating pipes from ±1.5℃ to ±0.5℃, thus avoiding energy waste caused by excessive heating.

[0081] When in supplemental heating mode, auxiliary heat sources are limited to reduce peak energy consumption. Energy consumption penalty: When the auxiliary heat source is activated (e.g., when the temperature of the hot water storage tank is insufficient and electric auxiliary heating is required), the control coefficient is adjusted. Energy consumption can be reduced by up to 30%, with circulating pump runtime shortened from 15 minutes to 10-12 minutes, and auxiliary heat source daily operating time reduced from 2 hours to 1.2 hours, effectively mitigating peak energy consumption. This also protects system safety and avoids overheating risks. When (heating pipes overheat), Outputting a negative value further reduces the penalty term. The reduced flow rate of the circulating pump slows down the hot water delivery rate, simultaneously triggering a slight opening of the pipe cooling valve, thus... The time required to cool from 26°C to 24°C should be controlled within 10 minutes to avoid accelerated aging of the pipes due to prolonged overheating.

[0082] D3, the running time of the circulating pump is obtained based on the multimodal control coefficients; After obtaining the multimodal control coefficients, the solar water heating system control device calculates the running time of the circulating pump based on the multimodal control coefficients. As one implementation method, the basic operating time of the circulating pump needs to be corrected by combining control coefficients, reference time, and the deviation between the hot water storage tank and the set temperature. The process of obtaining the circulating pump operating time based on multi-modal control coefficients specifically includes: ; in, Let t be the running time of the circulating pump foundation; These are the multimodal control coefficients; Baseline runtime; Real-time water temperature of the hot water storage tank; Set the temperature for the hot water storage tank; This represents the maximum permissible temperature difference for the hot water storage tank.

[0083] D4 controls the operation of the circulating pump based on the circulating pump flow rate and the circulating pump operating time.

[0084] After obtaining the circulation pump flow rate and circulation pump operating time, the solar water heating system control device controls the operation of the circulation pump based on the circulation pump flow rate and circulation pump operating time.

[0085] when When the correction term is positive, the base runtime is extended (to maximize the utilization of waste heat). when When the correction term is negative, the base runtime is shortened (to avoid invalid loops).

[0086] Among them, the lower limit threshold constraint: when hour, ,in Dynamically adjusts according to the mode, in heating mode =8 minutes, keep warm mode =5 minutes, in reheat mode =3 minutes, avoiding energy waste and equipment wear caused by short-term operation. The lower limit threshold constraint can ensure that the running time meets the equipment safety and energy efficiency requirements, and avoid shortening the life of the circulation pump due to too short "start-stop" cycle (<5min) (frequent start-stop will reduce the life of the pump by 20%-30%).

[0087] Upper limit threshold constraint: when hour, , =20 minutes, to prevent the system from overheating or excessive energy consumption due to prolonged continuous operation. The upper limit threshold can be used to balance heat supply and system safety, and avoid uncontrolled heat dissipation of heating pipes due to excessively high basic calculation values ​​(such as 25 minutes in extreme sunny weather) (heat loss increases by 0.5 kWh per hour after overheating).

[0088] Mode switching correction: When the system switches from one mode to another, the runtime needs to be smoothly transitioned. The correction formula is as follows: ,in This is the duration of the previous cycle. The calculation duration for the current cycle ensures a stable and shock-free switching process. A modified formula can be used to guarantee that the pipe water temperature fluctuation during mode switching is ≤±1℃, preventing sudden temperature changes perceived by the user (fluctuations exceeding ±2℃ will affect comfort).

[0089] The system achieves adaptive allocation of time-temperature weights through a bivariate function, improving adjustment accuracy by 25% compared to a fixed-weight scheme. Differentiated control logic is designed for different temperature ranges to solve the problem of adjustment failure in critical states of traditional algorithms. The penalty mechanism of the supplementary heating mode reduces auxiliary energy consumption by 18% and heat loss from the hot water storage tank by 12%. Experimental data show that after optimization, the indoor temperature fluctuation range of the system in an environment of -5℃ to 10℃ is reduced from ±2.5℃ to ±1.2℃, and the average daily operating energy consumption is reduced by 20%.

[0090] In summary, the system acquires solar energy temperature information, hot water storage tank temperature information, heating pipe temperature information, current time information, and solar radiation intensity. Based on these information, it determines temperature fluctuation rate information, which includes solar energy temperature fluctuation rate, hot water storage tank temperature fluctuation rate, and heating pipe temperature fluctuation rate. Under three operating conditions, a time-based control model, a PID-based temperature control method, and a combined control model are used to control the hot water flow between the solar water heater, hot water storage tank, and heating pipes. This achieves intelligent control of water temperature and volume, fully utilizes solar energy resources, saves other energy sources, and improves the user experience.

[0091] For those consistent with the above, please refer to Figure 2 , Figure 2 This application provides a schematic diagram of the structure of a solar water heating system control device. For example... Figure 2 As shown, the device includes a solar water heater, a hot water storage tank, heating pipes, and a circulating pump installed on the heating pipes. The device also includes: The information acquisition module 201 is used to acquire solar energy temperature information, hot water storage tank temperature information, heating pipe temperature information, current time information, and solar radiation intensity. The volatility determination module 202 is used to determine temperature volatility information based on the solar energy temperature information, the hot water storage tank temperature information, and the heating pipe temperature information; the temperature volatility information includes solar energy temperature volatility, hot water storage temperature volatility, and pipe temperature volatility. The first operation module 203 is used to control the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes according to a time control model when the current time information is not during a peak period, the current solar radiation intensity is greater than or equal to a first intensity threshold, and any one of the temperature fluctuation rates in the temperature fluctuation rate information is less than or equal to a first fluctuation threshold; the time control model is a model trained based on a neural network model. The second operation module 204 is used to control the flow of hot water between the solar water heater, the hot water storage tank and the heating pipes according to the temperature control method based on the PID algorithm when the current time information is not in a peak period, the current solar radiation intensity is less than the second intensity threshold and any one of the temperature fluctuation rates in the temperature fluctuation rate information is greater than the second fluctuation threshold. The third operation module 205 is used to control the flow of hot water between the solar water heater, the hot water storage tank, and the heating pipes according to the joint control model when the current time information is in a peak period.

[0092] This application also provides a terminal device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. When the processor executes the computer program, it implements the steps in the embodiment of the solar water heating system control method.

[0093] This application also provides a computer storage medium storing a computer program for electronic data exchange, which causes a computer to perform some or all of the steps of any of the solar water heating system control methods described in the above method embodiments.

[0094] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program that causes a computer to perform some or all of the steps of any of the solar water heating system control methods described in the above method embodiments.

Claims

1. A method of controlling a solar water heating system, the method comprising: The solar water heating system control method is applied to a solar water heating system control device, and the solar water heating system control device comprises a solar water heater, a heat storage water tank, a heating pipe, and a circulating pump arranged on the heating pipe. Obtain solar temperature information, heat storage water tank temperature information, heating pipe temperature information, current time information, and solar radiation intensity. Determine temperature fluctuation rate information according to the solar temperature information, the heat storage water tank temperature information, and the heating pipe temperature information; the temperature fluctuation rate information comprises a solar temperature fluctuation rate, a heat storage temperature fluctuation rate, and a pipe temperature fluctuation rate. When the current time information is not in a peak period, the current solar radiation intensity is greater than or equal to a first intensity threshold, and any one of the temperature fluctuation rate information is less than or equal to a first fluctuation threshold, then control the hot water flow among the solar water heater, the heat storage water tank, and the heating pipe according to a time control model; the time control model is a model obtained by training based on a neural network model. When the current time information is not in a peak period, the current solar radiation intensity is less than a second intensity threshold, and any one of the temperature fluctuation rate information is greater than a second fluctuation threshold, then control the hot water flow among the solar water heater, the heat storage water tank, and the heating pipe according to a temperature control method based on a PID algorithm. When the current time information is in a peak period, then control the hot water flow among the solar water heater, the heat storage water tank, and the heating pipe according to a joint control model.

2. The solar water heating system control method of claim 1, wherein, The control of the hot water flow among the solar water heater, the heat storage water tank, and the heating pipe according to the time control model comprises: Obtain basic meteorological information and position latitude and longitude information; Input the basic meteorological information, the position latitude and longitude information, the current time information, and the solar radiation intensity into a pre-trained time control model to obtain a solar radiation prediction value and a prediction confidence; the time control model is a hybrid neural network model with space-time perception capability formed by adding a double-branch feature extraction module and a dynamic weight adjustment layer on the basis of a traditional 3-layer BP network; Control the hot water flow among the solar water heater, the heat storage water tank, and the heating pipe according to the solar radiation prediction value and the prediction confidence.

3. The solar water heating system control method of claim 2, wherein, The control of the hot water flow among the solar water heater, the heat storage water tank, and the heating pipe according to the temperature control method based on the PID algorithm comprises: Obtain a heating pipe set value; Obtain a temperature deviation according to the heating pipe temperature information and the heating pipe set value; Determine a circulating pump flow rate based on a PID algorithm and a real-time temperature deviation; Control the hot water flow from the heating pipe to the heat storage water pipe according to the circulating pump flow rate.

4. The solar water heating system control method of claim 3, wherein, The control of the hot water flow among the solar water heater, the heat storage water tank, and the heating pipe according to the joint control model comprises: Obtain a time control weight and a temperature control weight according to the solar radiation prediction value and the temperature fluctuation rate information; the time control weight and the temperature control weight are joint control model weights. According to the radiator pipe temperature information, the time control weight and the temperature control weight, a multi-modal control coefficient is obtained; According to the multi-modal control coefficient, a circulating pump running time length is obtained; According to the circulating pump flow and the circulating pump running time length, the circulating pump is controlled to run.

5. The method of claim 4, wherein the step of determining the amount of solar energy available comprises the steps of: determining the amount of solar energy available based on the time of day, the date, and the geographic location of the solar water heating system. The multi-modal control coefficient is obtained according to the radiator pipe temperature information, the time control weight and the temperature control weight, including: When the current radiator pipe temperature is less than a first temperature threshold value; ； When the current radiator pipe temperature is less than or equal to a second temperature threshold value and greater than or equal to the first temperature threshold value; ； When the current radiator pipe temperature is greater than the second temperature threshold value; ； wherein, is a multi-modal control coefficient; is a time weight; is a solar radiation value; is a daily average radiation prediction value; is a temperature weight; is a real-time water temperature of a solar water heater; is a real-time water temperature of a heating pipe; is a temperature deviation.

6. The method of claim 4, wherein the step of determining the amount of solar energy available comprises the steps of: determining the amount of solar energy available based on the time of day, the date, and the geographic location of the solar water heating system. The circulating pump running time length is obtained according to the multi-modal control coefficient, including: ； wherein, is the basic running time of the circulating pump at time t; is the multi-modal control coefficient; is the basic running time; is the real-time water temperature of the heat storage water tank; is the set temperature of the heat storage water tank; is the maximum allowable temperature difference of the heat storage water tank.

7. The method of claim 1 to 6, wherein, The temperature fluctuation rate information includes a solar temperature fluctuation rate, a heat storage temperature fluctuation rate and a pipe temperature fluctuation rate, and the temperature fluctuation rate is determined according to the solar temperature information, the heat storage water tank temperature information and the radiator pipe temperature information, including: According to the solar temperature information, a solar initial temperature, a solar maximum temperature and a solar minimum temperature are obtained; According to the heat storage water tank temperature information, a heat storage initial temperature, a heat storage maximum temperature and a heat storage minimum temperature are obtained; According to the radiator pipe temperature information, a pipe initial temperature, a pipe maximum temperature and a pipe minimum temperature are obtained; According to the solar initial temperature, the solar maximum temperature and the solar minimum temperature, the solar temperature fluctuation rate is obtained; According to the heat storage initial temperature, the heat storage maximum temperature and the heat storage minimum temperature, the heat storage temperature fluctuation rate is obtained; According to the pipe initial temperature, the pipe maximum temperature and the pipe minimum temperature, the pipe temperature fluctuation rate is obtained.

8. A solar water heating system control device, characterized by, The system includes a solar water heater, a heat storage water tank, a radiator pipe and a circulating pump arranged on the radiator pipe, and further includes: An information acquisition module is configured to acquire solar temperature information, heat storage water tank temperature information, radiator pipe temperature information, current time information and solar radiation intensity; A fluctuation rate determination module is configured to determine temperature fluctuation rate information according to the solar temperature information, the heat storage water tank temperature information and the radiator pipe temperature information; the temperature fluctuation rate information includes a solar temperature fluctuation rate, a heat storage temperature fluctuation rate and a pipe temperature fluctuation rate; A first running module is configured to, when the current time information is not in a peak period, the current solar radiation intensity is greater than or equal to a first intensity threshold value and any one of the temperature fluctuation rate information is less than or equal to a first fluctuation threshold value, control the hot water flow among the solar water heater, the heat storage water tank and the radiator pipe according to a time control model; the time control model is a model obtained by training based on a neural network model; A second running module is configured to, when the current time information is not in a peak period, the current solar radiation intensity is less than a second intensity threshold value and any one of the temperature fluctuation rate information is greater than a second fluctuation threshold value, control the hot water flow among the solar water heater, the heat storage water tank and the radiator pipe according to a temperature control method based on a PID algorithm. The third operation module is configured to control the hot water flow among the solar water heater, the heat storage water tank and the heating pipe according to the joint control model when the current time information is in the peak period.

9. A terminal device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The computer program is executed by the processor to implement the solar water heating system control method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising: The computer program is executed by the processor to implement the solar water heating system control method according to any one of claims 1 to 7.

Images