Electric heating coupling control method and system based on pool heat pump

Abstract

The application relates to the technical field of pool heat pump control, and provides a pool heat pump-based electric heating coupling control method and system.The method comprises the following steps: collecting multi-source monitoring data of a pool in real time; calculating a comprehensive temperature difference value and a heat loss rate based on the multi-source monitoring data; inputting the comprehensive temperature difference value and the heat loss rate of the pool into a coupling heating model based on a space-time attention mechanism to generate dynamic heat pump power control parameters; inputting the multi-source monitoring data into a water quality temperature linkage control model, taking the maximum water quality standard attainment rate and the minimum temperature control energy consumption as game objectives, and adopting a Nash equilibrium algorithm to solve an optimal heat pump circulation adjustment strategy; and performing electric heating control on the pool heat pump according to the optimal heat pump circulation adjustment strategy and the heat pump power control parameters, so as to improve the pool temperature control efficiency, guarantee the pool operation safety, and realize the minimum energy consumption on the premise of guaranteeing the water quality standard attainment.

Description

Technical Field

[0001] This application relates to the field of swimming pool heat pump control technology, and more specifically, to an electric heating coupling control method and system based on a swimming pool heat pump. Background Technology

[0002] Swimming pool heating often relies on electric heaters or gas boilers. These devices generate heat directly through the combustion of electricity or fuel, resulting in high energy consumption, high operating costs, and significant environmental pollution. For example, traditional electric heating requires the direct conversion of electrical energy into heat energy, leading to low energy efficiency and a significant increase in operating costs over the long term.

[0003] In related technologies, with the maturity of heat pump technology, swimming pool heat pumps absorb low-grade heat energy from the air through a reverse Carnot cycle, requiring only a small amount of electricity to drive the compressor to achieve efficient heating, greatly improving energy efficiency. However, this type of swimming pool heat pump system has problems such as a single control mode and poor equipment coordination. For example, the heat pump unit and the circulation pump are often controlled in series. When the heat pump stops, the circulation pump stops simultaneously, resulting in the inability to continuously filter and disinfect the pool water, affecting water quality stability. In addition, this type of swimming pool heat pump system relies on a single temperature sensor for control, making it difficult to cope with dynamic temperature control needs in complex environments, and prone to water temperature fluctuations. Summary of the Invention

[0004] In this context, the embodiments of this application aim to provide an electric heating coupling control method and system based on a swimming pool heat pump, which can minimize energy consumption while ensuring water quality meets standards, and solve the problem of water quality fluctuations and energy consumption imbalance caused by single parameter control in traditional equipment.

[0005] In a first aspect of this application, an electric heating coupling control method based on a swimming pool heat pump is provided, comprising: real-time acquisition of multi-source monitoring data of the swimming pool; the multi-source monitoring data including at least: pool inlet and outlet water temperature parameters, ambient temperature, ambient humidity, and water quality parameters; calculating a comprehensive temperature difference between the pool inlet water temperature and the ambient temperature based on the multi-source monitoring data, and calculating the pool's heat loss rate based on the pool outlet water temperature and the pool inlet water temperature; inputting the comprehensive temperature difference and heat loss rate of the pool into a coupling heating model based on a spatiotemporal attention mechanism to generate dynamic heat pump power control parameters; wherein, the dynamic heat pump power... The control parameters include at least: the compressor frequency for adjusting the heating power of the pool heat pump, and the electric heater start-up threshold; the multi-source monitoring data is input into the water quality and temperature linkage control model, with the goal of maximizing the water quality compliance rate and minimizing temperature control energy consumption, and the optimal heat pump cycle regulation strategy is solved using the Nash equilibrium algorithm; wherein, the water quality and temperature linkage control model uses an adaptive control rule base constructed with a dynamic Bayesian network for strategy optimization; based on the optimal heat pump cycle regulation strategy and the heat pump power control parameters, the pool heat pump is electrically heated to improve the pool temperature control efficiency and ensure the safety of pool operation.

[0006] In a second aspect of this application, an electric heating coupling control system based on a swimming pool heat pump is provided, comprising: a data acquisition module for real-time acquisition of multi-source monitoring data of the swimming pool; the multi-source monitoring data including at least: pool inlet and outlet water temperature parameters, ambient temperature, ambient humidity, and water quality parameters; a coupling heating prediction module for calculating, based on the multi-source monitoring data, a comprehensive temperature difference between the pool inlet water temperature and the ambient temperature, and calculating the pool heat loss rate based on the pool outlet water temperature and the pool inlet water temperature; inputting the comprehensive temperature difference and heat loss rate of the pool into a coupling heating model based on a spatiotemporal attention mechanism to generate dynamic heat pump power control parameters; wherein, the dynamic heat pump power... The rate control parameters include at least: the compressor frequency for adjusting the heating power of the pool heat pump, and the electric heater start-up threshold; a linkage prediction module for inputting the multi-source monitoring data into the water quality and temperature linkage control model, using the goal of maximizing the water quality compliance rate and minimizing temperature control energy consumption, and employing a Nash equilibrium algorithm to solve for the optimal heat pump cycle regulation strategy; wherein, the water quality and temperature linkage control model uses an adaptive control rule base constructed with a dynamic Bayesian network for strategy optimization; and a control module for electrically controlling the pool heat pump according to the optimal heat pump cycle regulation strategy and the heat pump power control parameters, in order to improve the pool temperature control efficiency and ensure the safety of pool operation.

[0007] In a third aspect of the embodiments of this application, a terminal device is provided, the terminal device comprising: at least one processor, a memory, and an input / output unit; wherein the memory is used to store a computer program, and the processor is used to invoke the computer program stored in the memory to execute the electro-heating coupling control method based on a pool heat pump as described in the first aspect.

[0008] In a fourth aspect of the embodiments of this application, a computer-readable storage medium is provided, which includes instructions that, when executed on a computer, cause the computer to perform the electro-heating coupling control method based on a pool heat pump as described in the first aspect.

[0009] In a fifth aspect of the embodiments of this application, a computer program product is provided, including a computer program that, when executed by a processor, implements the electro-heating coupling control method based on a pool heat pump as described in the first aspect.

[0010] According to an embodiment of this application, an electric heating coupling control method and system based on a swimming pool heat pump is proposed, which involves real-time acquisition of multi-source monitoring data of the swimming pool. The multi-source monitoring data includes at least: pool inlet and outlet water temperature parameters, ambient temperature, ambient humidity, and water quality parameters. Then, based on the multi-source monitoring data, the comprehensive temperature difference between the pool inlet water temperature and the ambient temperature is calculated, and the heat loss rate of the pool is calculated based on the pool outlet water temperature and the pool inlet water temperature. The comprehensive temperature difference and heat loss rate of the pool are then input into a coupling heating model based on a spatiotemporal attention mechanism to generate dynamic heat pump power control parameters. These dynamic heat pump power control parameters include at least: the compressor frequency for adjusting the heating power of the pool heat pump, and the electric heater start-up threshold. The multi-source monitoring data is input into a water quality and temperature linkage control model. With the goal of maximizing the water quality compliance rate and minimizing temperature control energy consumption, a Nash equilibrium algorithm is used to solve for the optimal heat pump cycle regulation strategy. The water quality and temperature linkage control model uses an adaptive control rule base constructed using a dynamic Bayesian network for strategy optimization. Finally, based on the optimal heat pump cycle adjustment strategy and the heat pump power control parameters, the pool heat pump is electrically heated to improve pool temperature control efficiency and ensure pool operational safety. This embodiment of the application can improve equipment synergy in the pool heat pump temperature control process, enhance pool temperature control efficiency, ensure the pool remains within the set temperature fluctuation range, and reduce pool heat pump energy consumption, thus meeting the dynamic temperature control needs in complex scenarios. Attached Figure Description

[0011] Figure 1 This application illustrates a flowchart of an electric heating coupling control method based on a swimming pool heat pump.

[0012] Figure 2This application shows a schematic diagram of an electric heating coupling control system based on a swimming pool heat pump.

[0013] Figure 3 The present invention provides a schematic diagram of the structure of a medium according to an embodiment of the present application. Detailed Implementation

[0014] The following is for reference. Figure 1 , Figure 1 This is a schematic flowchart illustrating an electric heating coupling control method based on a pool heat pump, provided as an embodiment of this application. It should be noted that the implementation of this application can be applied to any applicable underwater operation system usage and / or maintenance scenario.

[0015] To address at least one of the aforementioned technical problems, embodiments of this application provide an electric heating coupling control method and system based on a swimming pool heat pump.

[0016] Specifically, the implementation method of this application significantly improves the overall performance of the pool heating system through the deep integration of multi-dimensional data fusion and intelligent control strategies. In terms of energy efficiency optimization, the coupled heating model based on a spatiotemporal attention mechanism can accurately capture the spatiotemporal correlation between ambient temperature and humidity and pool water temperature, dynamically adjust the heat pump compressor frequency, and intelligently set the electric heater start-up threshold, ensuring the system always operates within a high-efficiency range and effectively reducing overall energy consumption. The water quality and temperature linkage control mechanism, through an adaptive rule base constructed using a dynamic Bayesian network, minimizes energy consumption while ensuring water quality meets standards, solving the problem of water quality fluctuations and energy consumption imbalances caused by single-parameter control in traditional equipment.

[0017] Furthermore, the intelligence level of the implementation method in this application is reflected in the synergistic application of multi-objective game optimization and digital twin technology. By solving the optimal solution of the game between water quality compliance rate and temperature control energy consumption through the Nash equilibrium algorithm, the system can autonomously select the optimal operating strategy. For example, when water quality is abnormal, it can prioritize ensuring the circulation filtration efficiency while dynamically compensating for the power shortfall of the heat pump. The simulation and prediction function of the digital twin model on the water quality evolution path enables the system to have forward-looking control capabilities, which can predict the impact of sudden environmental changes on water temperature in advance and actively adjust operating parameters to control water temperature fluctuations within a very small range.

[0018] In terms of safety and reliability, the implementation method of this application also constructs multiple safety defenses through an abnormal operating condition circuit breaker mechanism and a multi-level protection strategy. The compressor's operating status and water quality parameters are monitored in real time, and when an abnormal mode is detected, a graded response is automatically triggered, such as reducing the heat pump power and starting the backup circulation pump to avoid equipment overload damage. An adaptive learning mechanism continuously optimizes the control model parameters, enabling the system to adapt to different regional climate characteristics and pool usage scenarios, significantly extending equipment lifespan and reducing operation and maintenance costs.

[0019] Figure 1 The flowchart of an embodiment of this application provides an electric heating coupling control method based on a swimming pool heat pump, including:

[0020] Step S101: Collect multi-source monitoring data of the swimming pool in real time.

[0021] In this embodiment of the application, the multi-source monitoring data includes at least: pool inlet and outlet water temperature parameters, ambient temperature, ambient humidity, and water quality parameters.

[0022] In step S101, real-time acquisition of multi-source monitoring data from the swimming pool is fundamental to achieving precise control. For example, a sensor network deployed at different locations within the pool continuously acquires the pool's inlet and outlet water temperature parameters, accurately reflecting the heat exchange efficiency during water circulation and providing crucial information for assessing heat pump load and energy consumption. Ambient temperature and humidity data are captured in real-time by high-precision meteorological sensors, helping the system dynamically analyze the impact of external climate conditions on the pool's thermal balance. For instance, high temperature and high humidity environments may exacerbate heat loss, requiring advance adjustments to the heat pump's operating strategy to maintain stable water temperature.

[0023] For example, water quality parameter monitoring covers core indicators such as residual chlorine concentration, turbidity, and pH value, with data updates occurring every minute through multi-parameter water quality sensors. This data not only assesses disinfection effectiveness and water cleanliness but also reflects changes in pool usage frequency and contaminant load. For instance, abnormal fluctuations in residual chlorine concentration may indicate insufficient disinfectant dosage or increased organic pollution, requiring intervention in conjunction with the circulating filtration system. Continuous tracking of the inlet and outlet water temperature difference provides fundamental parameters for calculating heat loss rates. Combined with covariance analysis of ambient temperature and humidity, this allows for accurate prediction of the pool's overall heat load demand, providing scientific support for dynamically adjusting heat pump power and electric heating thresholds.

[0024] Step S102: Based on the multi-source monitoring data, calculate the comprehensive temperature difference between the pool inlet water temperature and the ambient temperature, and calculate the heat loss rate of the pool based on the pool outlet water temperature and the pool inlet water temperature.

[0025] In principle, step S102 achieves precise quantification of the pool's thermodynamic state through deep correlation analysis of multi-source monitoring data. Its core principle lies in constructing a dynamic model of the pool's heat exchange process, coupling environmental parameters with water parameters for analysis. Taking the comprehensive temperature difference calculation between the inlet water temperature and the ambient temperature as an example, the system not only directly compares the numerical differences between the two but also incorporates environmental humidity and wind speed parameters, quantifying the correction effect of evaporative heat dissipation on heat exchange efficiency through the thermodynamic model. For instance, when the ambient humidity is low, the evaporation rate of the pool surface accelerates under the same temperature difference, resulting in an actual heat loss intensity higher than the calculated value based solely on the temperature difference. In this case, the comprehensive temperature difference model dynamically corrects the temperature difference using the humidity coefficient, more accurately reflecting the heat pump load demand. This model overcomes the limitations of traditional static temperature difference calculations by incorporating dynamic environmental characteristics into the heat balance analysis system.

[0026] At the heat loss rate calculation level, the system establishes a multi-factor coupled model based on the principles of mass and energy conservation. The temperature difference between the pool outlet and inlet water reflects the overall heat loss trend of the water body, but the actual heat loss rate needs to be corrected by combining hydrodynamic parameters. For example, when the water flow rate increases, the improved heat conduction efficiency of the pipe inner wall will exacerbate heat loss. In this case, the system calculates the fluid boundary layer thickness using pipe thermal resistance parameters and Reynolds number to dynamically assess changes in heat conduction efficiency. Simultaneously, the environmental temperature and humidity covariance parameters are used to quantify the convective heat transfer coefficient of airflow. For example, in scenarios with sudden changes in wind speed, the system corrects the weight of the convective heat transfer term using the wind speed-humidity joint distribution function to avoid bias in heat loss rate assessment caused by local turbulence. This multi-parameter dynamic correction mechanism significantly improves the accuracy of heat loss rate calculation.

[0027] Taking the calculation of the combined temperature difference between the pool inlet water temperature and the ambient temperature as an example, in step S102, the real-time water temperature at the pool inlet (e.g., 28℃) and the ambient air temperature (e.g., 25℃) are first obtained. Combined with the pool surface wind speed and air humidity parameters, the combined temperature difference between the two is calculated using a thermodynamic model. This temperature difference not only reflects the intensity of direct heat exchange between the pool and the external environment, but also implies the dynamic changes in the pool's thermal boundary conditions. For example, when the ambient humidity is high, the evaporative heat dissipation efficiency of the pool surface will significantly increase under the same temperature difference, thereby affecting the load demand of the heat pump.

[0028] For example, in calculating the heat loss rate, a dynamic evaluation model for the heat loss rate is established based on the temperature difference between the pool outlet and inlet water (e.g., outlet temperature 26℃, inlet temperature 28℃), combined with the pool water flow rate and the pipe thermal conductivity coefficient. For instance, if the water flow rate is 10 cubic meters per hour and the temperature difference is 2℃, the heat loss per unit time is calculated using the specific heat capacity and mass flow rate of water. Simultaneously, the system incorporates the environmental temperature and humidity covariance parameter to correct for fluctuations in heat convection and conduction efficiency caused by airflow or humidity changes. For example, in a high-temperature and high-humidity environment, even with a small temperature difference, high humidity may enhance evaporative cooling, causing the actual heat loss rate to exceed the expected value calculated solely based on the temperature difference. These calculation results provide crucial information for heat pump power adjustment and electric heating threshold setting, ensuring the system maintains optimal energy efficiency and stable pool water temperature.

[0029] Step S103: Input the comprehensive temperature difference value and heat loss rate of the pool into the coupled heating model based on the spatiotemporal attention mechanism to generate dynamic heat pump power control parameters.

[0030] In this embodiment, the coupled heating model based on a spatiotemporal attention mechanism achieves dynamic modeling and precise control of the pool's thermodynamic state by fusing time series analysis and spatial feature extraction. Its core principle lies in constructing a multi-dimensional spatiotemporal perception network to jointly encode the heat exchange characteristics of different areas of the pool with dynamic changes over time. From a structural perspective, the bottom layer extracts the spatiotemporal features of local heat conduction in the pool using a convolutional neural network, such as the local temperature gradient changes at the pool's edge due to water evaporation. The middle layer utilizes a self-attention mechanism to establish cross-time step heat load correlations, capturing the fluctuations in heat demand caused by day-night cycles or sudden weather changes. The upper layer integrates environmental parameters (such as air temperature and humidity, wind speed) and pool structural parameters (such as volume and circulation pipe layout) through a spatiotemporal graph neural network to form a global thermal balance representation. This architecture overcomes the limitations of traditional single-dimensional modeling, enabling the system to simultaneously analyze the local heterogeneity of heat conduction and the global correlation of environmental disturbances.

[0031] Understandably, the coupled heating model based on spatiotemporal attention introduces a dynamic weight allocation mechanism to enhance feature interaction. For example, when dealing with areas with significant temperature differences between the east and west sides of the pool, the spatial attention module enhances the feature response weights at the corresponding locations, while the temporal attention module focuses on the heat pump operating status under similar temperature difference patterns in historical data. The dual-channel feature fusion network further decouples high-frequency water temperature fluctuation signals from low-frequency environmental trend signals, dynamically adjusting the contribution of each channel through a gating mechanism. The introduction of the federated learning framework enables multi-pool systems to share spatiotemporal feature learning results; for example, the pipe thermal resistance parameters learned from pools in cold regions can be transferred to other regions, improving the model's generalization ability. This design allows the system to retain regionally specific characteristics while inheriting cross-regional common patterns when dealing with complex thermal disturbances.

[0032] In this embodiment of the application, the dynamic heat pump power control parameters include at least: the compressor frequency for adjusting the heating power of the pool heat pump, and the electric heater start-up threshold.

[0033] Specifically, in step S103, the coupled heating model based on the spatiotemporal attention mechanism achieves precise control of the heat pump power by dynamically analyzing the multidimensional correlation between the pool's thermodynamic state and the external environment. The core principle of step S103 is to use a spatiotemporal attention network to model the heat exchange characteristics of different areas of the pool, such as identifying the enhanced local heat loss caused by water evaporation at the pool's edge. Simultaneously, it combines the fluctuation patterns of ambient temperature and humidity over time (such as diurnal temperature variations) to establish a nonlinear mapping relationship between heat load and equipment power. When an increase in the combined temperature difference between the inlet water temperature and the ambient temperature is detected, the model prioritizes increasing the heating power to compensate for heat loss by enhancing the adjustment weight of the heat pump compressor frequency; conversely, it reduces the operating intensity to avoid energy waste. The setting of the electric heater's start-up threshold is based on the heat loss rate distribution characteristics in historical data. For example, it automatically lowers the start-up threshold during continuous rainy weather to activate the auxiliary heating function in advance to cope with sudden surges in heat load.

[0034] For example, in a seaside swimming pool during the afternoon in summer, the inlet water temperature of 28°C and the ambient temperature of 32°C create an inverse temperature difference. The spatiotemporal attention model analyzes the abnormally high heat loss in the eastern area of ​​the pool due to accelerated evaporation caused by sea breezes. It dynamically increases the frequency of the heat pump compressor in that area to 55Hz, while simultaneously lowering the electric heater start-up threshold from the usual 2°C to 1.5°C. At night, after the ambient temperature drops to 24°C, the model, based on historical nighttime heat loss rate curves, gradually reduces the compressor frequency to 40Hz and resets the electric heater threshold to 2.5°C, ensuring the system always operates within its optimal energy efficiency range.

[0035] Thus, through adaptive fusion of spatiotemporal characteristics, the energy waste problem caused by static threshold settings in traditional constant temperature systems is effectively solved. The dynamic adjustment strategy improves the operating efficiency of the heat pump under complex weather conditions, while the precise start-stop strategy of the electric heater shortens the ineffective heating time. Practical applications show that when dealing with sudden heat load fluctuations (such as rainstorm cooling), this model can control the water temperature fluctuation within ±0.5℃, significantly outperforming the ±2℃ fluctuation performance of traditional PID control systems.

[0036] As an optional embodiment, in step S103, the comprehensive temperature difference value and heat loss rate of the pool are input into a coupled heating model based on a spatiotemporal attention mechanism to generate dynamic heat pump power control parameters. This includes: extracting time-series features from the comprehensive temperature difference value and heat loss rate of the pool to obtain a dynamic feature vector; the dynamic feature vector includes at least: water temperature gradient change rate and ambient temperature and humidity covariance; inputting the dynamic feature vector into a federated learning architecture to aggregate historical operating data from multiple pools, training the coupled heating model based on a spatiotemporal attention mechanism, and using the trained coupled heating model to generate the dynamic heat pump power control parameters to be set; in the federated learning architecture, dynamic learning weights are assigned to each pool to improve the diversity of historical pool data in the federated learning architecture.

[0037] Specifically, in step S103, the spatiotemporal attention-coupled heating model based on federated learning achieves cross-pool knowledge transfer through a distributed collaborative mechanism. Its core principle lies in constructing a decentralized feature learning network. The system extracts localized features from the temporal feature vectors of each pool (such as the rate of change of water temperature gradient and the covariance of ambient temperature and humidity), and simultaneously aligns and fuses the feature space through a federated learning architecture. For example, during training, the model dynamically assigns higher weights to pools in high-humidity areas to enhance the learning of evaporative heat dissipation features, while pools in colder areas focus on extracting features related to pipe heat conduction. This adaptive feature fusion strategy enables the model to overcome the limitations of single-pool data and capture common thermodynamic laws across regions.

[0038] For example, a cluster of swimming pools distributed across tropical and temperate regions utilizes a federated learning framework for collaborative model training. Each pool's local controller preserves the privacy of its original data, uploading only feature vectors and gradient update information. When a tropical pool detects abnormal evaporative cooling under high humidity, its learned feature weights are transmitted to the temperate pool nodes via an encrypted aggregation protocol, prompting the global model to adjust its focus on the environmental temperature and humidity covariance parameters. Simultaneously, the system assigns higher learning weights to newly opened pools with limited historical data, rapidly improving model adaptability through transfer learning, enabling them to achieve control accuracy similar to that of established pools from the initial operational phase.

[0039] The aforementioned federated learning framework significantly improves model generalization ability and data privacy protection. The federated learning mechanism enhances model training efficiency while ensuring the independence of data from each pool, and cross-regional feature transfer improves prediction accuracy under extreme weather conditions. Actual deployment data shows that pool groups using this technology experience reduced overall energy consumption and lower equipment failure rates, especially during seasonal transitions, maintaining water temperature fluctuations within set ranges, outperforming traditional centralized models.

[0040] In step S103 above, optionally, time-series feature extraction is performed on the comprehensive temperature difference value and heat loss rate of the pool to obtain a dynamic feature vector, including: performing multi-scale analysis on the dynamic feature vector, using the high-frequency component obtained by decomposition as the water temperature gradient change rate, and using the low-frequency component obtained by decomposition as the environmental temperature and humidity covariance, and inputting them into a gated recurrent unit for feature enhancement; using a dual-channel LSTM network to process the feature-enhanced water temperature gradient change rate and environmental temperature and humidity covariance, and outputting the first channel feature vector and the second channel feature vector.

[0041] Specifically, the dynamic feature enhancement mechanism constructed in step S103 using multi-scale feature decomposition and a dual-channel LSTM network is based on the core principle of decoupling and adaptively fusing the time-series features of the pool's thermodynamic state in multiple dimensions. First, methods such as Empirical Mode Decomposition (EMD) or wavelet transform are used to decompose the dynamic feature vector composed of the combined temperature difference and heat loss rate into high-frequency components (e.g., abrupt changes in water temperature gradient) and low-frequency components (e.g., periodic fluctuations in ambient temperature and humidity). The high-frequency components are enhanced using gated recurrent units (GRUs), which filter out key time windows for abrupt changes in water temperature, such as local heat exchange anomalies caused by changes in solar radiation at the pool edge. The low-frequency components are captured by another GRU network to capture long-term trends in environmental parameters, such as the evolution of the covariance of ambient temperature and humidity caused by day-night cycles. The dual-channel LSTM network processes the enhanced high-frequency and low-frequency features respectively. The first channel LSTM focuses on the short-term dynamic features of water temperature gradient changes, while the second channel LSTM models the long-range dependence of the environmental covariance. Finally, a spatiotemporal joint representation vector is formed through feature concatenation.

[0042] For example, during the rainy season, a mountain resort's swimming pool exhibited high-frequency oscillations in the combined temperature difference between the inlet water and the ambient temperature (e.g., fluctuations of ±0.3℃ every 10 minutes), while the ambient temperature and humidity covariance showed significant diurnal variations (e.g., increased daytime humidity leading to enhanced evaporative heat dissipation). A multi-scale decomposition module input the high-frequency oscillation components into a GRU network, identifying the phenomenon of intensified localized evaporation on the water surface caused by mountain wind disturbances. The low-frequency components, after GRU processing, captured the periodic attenuation of solar radiation intensity caused by cloud cover during the rainy season. A dual-channel LSTM learned the correlation between abrupt changes in water temperature gradient and the start-up / shutdown sequence of the pool circulation pump in the high-frequency channel, and the lag relationship between the ambient humidity covariance and the heat pump's heating efficiency in the low-frequency channel. Through feature fusion, the model dynamically adjusts the heat pump power allocation strategy, increasing the compressor frequency in advance during peak humidity periods at night in the rainy season, while simultaneously reducing the dependence on the electric heater.

[0043] Thus, through feature decoupling and adaptive fusion, the model's analytical capability for complex thermodynamic interactions is significantly improved. The dual-channel LSTM design shortens the prediction response time when dealing with multi-scale thermal disturbances, and the feature enhancement process effectively suppresses environmental noise interference, improving water temperature control accuracy. Actual deployment data shows that in regions with drastic seasonal climate changes, this approach reduces heat pump energy consumption while extending the lifespan of equipment components, verifying the effectiveness of multi-scale feature engineering and deep temporal network collaborative optimization.

[0044] In step S103 above, optionally, the dynamic heat pump power control parameters to be set are generated using the trained coupled heating model, including: calculating the feature interaction weights between the first channel feature vector and the second channel feature vector through a cross-channel attention mechanism; combining the feature interaction weights, the first channel feature vector, and the second channel feature vector to collaboratively predict the heat load state for the next two hours, thereby obtaining the heat load fluctuation range for the next two hours; generating a compressor frequency adjustment sequence using a model predictive control (MPC) algorithm based on the heat load fluctuation range for the next two hours; and generating the compressor frequency adjustment parameter in the dynamic heat pump power control parameters based on the compressor frequency adjustment sequence. The compressor frequency adjustment parameter is used to indicate the setting value of the compressor frequency to be updated every five minutes for the next two hours to ensure that the compressor operates within a preset high-efficiency range.

[0045] Specifically, the scheme in step S103 that uses a cross-channel attention mechanism in conjunction with model predictive control (MPC) to generate dynamic heat pump power control parameters is based on the core principle of constructing a multi-dimensional feature interaction and dynamic optimization framework. The cross-channel attention mechanism overcomes the limitations of traditional single-channel feature analysis by modeling the deep correlations between different feature channels. Taking the water temperature gradient change rate (first channel) and the environmental temperature and humidity covariance (second channel) as examples, the system first maps the two types of feature vectors to a high-dimensional space. It then dynamically generates feature interaction weights by calculating the cross-channel similarity matrix (such as cosine similarity or dot product attention). This weight reflects the coupling strength between the two types of features in heat load prediction. For example, when the environmental humidity covariance suddenly increases, the mechanism strengthens its attention weight to changes in the water temperature gradient, capturing the nonlinear impact of evaporative cooling on the heat load.

[0046] For example, in a coastal swimming pool under high temperature and humidity conditions during summer, the system detected short-term, sharp fluctuations in the rate of change of water temperature gradient (e.g., fluctuations of ±0.5℃ every 5 minutes), while the environmental temperature and humidity covariance exhibited significant diurnal periodic fluctuations. The cross-channel attention module, by calculating the cross-modal correlation between the two types of features, discovered a 0.8-second lag correlation between humidity covariance and abrupt changes in water temperature gradient. This weighting information was used by the collaborative prediction module, combined with historical heat load data, to generate a heat load fluctuation range for the next two hours (e.g., 28.5℃-30.2℃). Based on this range, the MPC algorithm, aiming for optimal compressor energy efficiency, optimized the frequency adjustment command every five minutes. For example, during the peak period of humidity covariance, the compressor frequency was increased to 90% of the rated value in advance to avoid heat accumulation caused by lag response.

[0047] It is understandable that the core of the cross-channel attention mechanism lies in establishing dynamic interaction relationships between different feature channels to capture the coupling effect of multi-source heterogeneous data. In step S103 of this application, the mechanism is applied to the water temperature gradient change rate (first channel feature vector) and the environmental temperature and humidity covariance (second channel feature vector). The specific implementation process is divided into the following three steps: First, feature mapping and similarity calculation. The two types of feature vectors are mapped to a high-dimensional space through linear transformation, and the cross-channel similarity matrix (such as dot product or cosine similarity) is calculated. For example, the water temperature gradient change rate reflects the local heat exchange intensity, while the environmental temperature and humidity covariance describes the evaporative heat dissipation trend. The two are quantified by the similarity matrix to quantify their correlation strength. Second, dynamic weight allocation. Feature interaction weights are generated based on similarity. If a sudden increase in environmental humidity leads to enhanced evaporative heat dissipation, the mechanism will significantly increase the weight of the second channel on the first channel, strengthening the lag correlation of "humidity-water temperature gradient" (such as 0.8-second lag). Third, feature collaborative fusion. The weights are fused with the original feature vectors to generate a joint representation vector. This process preserves the details of high-frequency water temperature changes while injecting global information on low-frequency environmental trends, providing multi-scale input for subsequent predictions.

[0048] The cross-channel attention mechanism in the above embodiments is manifested in the fact that it is the first time that cross-channel attention has been introduced into the heat pump control system, solving the problem that traditional single-channel models are unable to capture the nonlinear coupling of multi-source features (such as physical parameters and environmental parameters). Compared with traditional PID control that relies on fixed thresholds, it adapts to environmental disturbances (such as sudden changes in humidity) through dynamic weight allocation, thereby improving control accuracy. It combines the hierarchical processing of high-frequency and low-frequency features. The high-frequency channel (water temperature gradient) captures transient disturbances (such as water temperature oscillations caused by the start and stop of the circulating pump) through a short-term attention window, while the low-frequency channel (environmental covariance) predicts periodic trends (such as day-night temperature and humidity cycles) through long-range dependency modeling. Dual-channel decoupling avoids feature confusion and improves model robustness. Cross-channel attention is combined with model predictive control (MPC). The attention mechanism outputs the heat load fluctuation range for the next two hours (such as 28.5–30.2℃), and MPC continuously optimizes the compressor frequency adjustment sequence based on this range, ensuring that every 5-minute instruction aims at the optimal compressor energy efficiency (such as increasing the frequency to 90% of the rated value in advance during peak humidity periods), thereby achieving dynamic response and energy consumption balance.

[0049] Further optionally, after generating the compressor frequency adjustment sequence using the Model Predictive Control (MPC) algorithm based on the heat load fluctuation range for the next two hours in step S103, the method further includes:

[0050] Based on the enhanced environmental temperature and humidity covariance, a heating compensation coefficient is calculated using a nonlinear compensation function. This coefficient is then used to dynamically compensate for predicted heat load fluctuations, correcting outliers in the heat load fluctuation range for the next two hours and generating an electric heater start-up threshold that matches the heat load fluctuation range. Based on the compressor frequency adjustment sequence and the electric heater start-up threshold matching the heat load fluctuation range, a compressor switching command is generated in the dynamic heat pump power control parameters. This compressor switching command indicates the timing of compressor start-up or shutdown operations within the next two hours to ensure the pool heat pump remains within the set temperature range.

[0051] Specifically, in principle, the coordinated changes in ambient temperature and humidity are key factors causing fluctuations in the pool's heat load. The relationship between the two is not a simple linear one, but involves complex interactions. By enhancing the processing of the ambient temperature and humidity covariance, key features in the joint changes of temperature and humidity can be extracted, such as abnormal linkage patterns between temperature and humidity in different seasons, making the impact of these environmental factors on the heat load easier to quantify and capture. The role of nonlinear compensation functions is to accurately characterize this nonlinear relationship. For example, a sudden increase in humidity coupled with a slight decrease in temperature may lead to abnormal fluctuations in the heat load. Such functions can specifically transform this complex correlation into corresponding heating compensation coefficients, providing a basis for subsequent corrections. Dynamically compensating for heat load fluctuation ranges essentially uses calculated compensation coefficients to correct prediction deviations caused by sudden environmental changes (such as short-term strong winds accompanied by a sharp increase in humidity), eliminating outliers that do not conform to reality, and making the fluctuation range more closely match the actual heat load change trend. Matching the electric heater start-up threshold with the corrected heat load range is to clarify the intervention point of auxiliary heating equipment, ensuring that when the compressor alone cannot meet the heating demand, the auxiliary equipment can be activated in time to supplement heat. The core of generating a switching command by combining the compressor frequency adjustment sequence with this start-up threshold is to coordinate the working rhythm of the main heating equipment (compressor) and the auxiliary heating equipment, so as to avoid the single equipment from running at high load or inefficiently for a long time, and ultimately achieve stable control of the pool temperature.

[0052] Taking the operation of a large indoor swimming pool during the rainy season as an example, when it is predicted that there may be a sudden increase in outdoor humidity and a slight decrease in indoor temperature within the next two hours, the real-time collected temperature and humidity data will first be enhanced to highlight the linkage between the rate of increase in humidity and the magnitude of decrease in temperature, thus obtaining the enhanced temperature and humidity covariance to reflect the intensity of the joint changes in environmental factors at this time. Subsequently, through a nonlinear compensation function, considering that the evaporation rate of pool water is accelerated under the high humidity environment of the rainy season, resulting in more significant heat loss than under normal conditions, the corresponding heating compensation coefficient will be calculated. The originally predicted heat load fluctuation range may contain some outliers due to insufficient consideration of the special characteristics of this temperature and humidity combination. After correction by the compensation coefficient, the upper and lower limits of the range will be adjusted to more accurately reflect the actual heat required. Based on the corrected heat load fluctuation range, the system will determine the start-up threshold of the electric heater, clarifying at what level of heat load the electric heater needs to be activated for auxiliary heating. At the same time, combined with the compressor frequency adjustment sequence generated by the model predictive control algorithm, the compressor switching commands will be further generated to clarify when the compressor will start and when it will stop within the next two hours. For example, when the heat load is low, the compressor may run at a lower frequency for a period of time and then shut down. When the heat load rises to near the start-up threshold, the compressor will start in advance and run at the corresponding frequency, working in conjunction with any electric heaters that may be activated to maintain the pool temperature within the set range.

[0053] In the example above, in the step of calculating the heating compensation coefficient using a nonlinear compensation function based on the enhanced environmental temperature and humidity covariance, the heating compensation coefficient... Mapping covariance using a nonlinear function This can be expressed as the following formula: ;in, This represents the covariance of temperature and humidity in the pool environment. This represents the basic compensation weight, which is obtained by fitting historical data. A value of 1.2 indicates that an additional 20% thermal compensation is required by default. It is expressed as a Gaussian decay factor, which controls the sensitivity to extreme covariance values. When the value is greater than 0, noise interference is suppressed. and This is a non-linear adjustment parameter used to enhance the compensation effect of negative covariance. For example, in dry and cold conditions. If it is less than 0, it needs to be increased. . The covariance is mapped to an unbounded space by using the inverse hyperbolic tangent function, thus resolving the gradient vanishing problem in the saturation region. The parameters in the above formula... , , , It can be calibrated using thermodynamic differential equations and correlated with the material's specific heat capacity and convection coefficient. This formula transforms the environmental covariance into a quantifiable thermal compensation command, solving the underfitting problem of traditional linear compensation in nonlinear heat exchange scenarios.

[0054] This significantly improves the temperature control accuracy and operating efficiency of the pool heat pump system. Through feature enhancement and nonlinear compensation, the prediction deviation of the heat load fluctuation range is effectively corrected, and the elimination of outliers makes the fluctuation range closer to the actual situation, providing a reliable basis for subsequent equipment control. The electric heater start-up threshold matches the corrected heat load range, avoiding the blind start-up and shutdown of auxiliary heating equipment. It avoids premature start-up before the heat load reaches the required level, thus preventing energy waste, and also avoids delayed start-up when the heat load exceeds the compressor's capacity alone, leading to temperature fluctuations. The generation of compressor switching commands coordinates the working states of the compressor and electric heater, allowing them to work together in an orderly manner at different heat load stages, reducing the equipment's ineffective operating time and unnecessary energy consumption. At the same time, this coordinated control reduces frequent compressor start-ups and sudden frequency changes, reducing mechanical wear and tear on the equipment and helping to extend its service life. Ultimately, the entire system can stably maintain the pool temperature within the set range under complex and variable environmental conditions, ensuring the comfort of the pool while achieving multiple goals of energy saving and equipment protection.

[0055] Step S104: The multi-source monitoring data is input into the water quality and temperature linkage control model. With the goal of maximizing the water quality compliance rate and minimizing temperature control energy consumption, the optimal heat pump cycle regulation strategy is solved using a Nash equilibrium algorithm. In this embodiment, the water quality and temperature linkage control model uses an adaptive control rule base constructed with a dynamic Bayesian network for strategy optimization.

[0056] As an optional embodiment, in step S104, the multi-source monitoring data is input into the water quality and temperature linkage control model. With the goal of maximizing the water quality compliance rate and minimizing temperature control energy consumption, the optimal heat pump cycle regulation strategy is solved using the Nash equilibrium algorithm, including:

[0057] Temporal features are extracted from the pool inlet / outlet water temperature parameters and ambient temperature in the multi-source monitoring data to obtain the temperature control energy consumption variation characteristics; temporal features are extracted from the water quality parameters in the multi-source monitoring data to obtain the water quality parameter deviation characteristics; a game theory model is constructed, setting a water quality control party and an energy consumption control party, as well as corresponding water quality control payoff functions and energy consumption control payoff functions; wherein, the objective of the water quality control party is to maximize the water quality compliance rate, and the adjustable variables are the circulation pump power and the start / stop status of the electric heater; the objective of the energy consumption control party is to minimize the temperature control energy consumption, and the adjustable variables are... The regulating variables are the heat pump compressor frequency and the proportion of circulating pump runtime. The temperature control energy consumption variation characteristics and water quality parameter deviation characteristics are projected onto the strategy space for discretization to obtain a finite strategy set. Based on the water quality control payoff function and the energy consumption control payoff function, a payoff matrix for multi-party game is constructed, with each matrix cell storing the payoff value corresponding to both parties at discrete feature points. The mixed strategies of the water quality control party and the energy consumption control party are input into the payoff matrix, and the optimal heat pump cycle regulation strategy is obtained by solving the mixed strategy Nash equilibrium using the Lagrange multiplier method.

[0058] Specifically, in step S104, the principle behind time-series feature extraction lies in capturing key patterns related to temperature control energy consumption and water quality from dynamically changing multi-source monitoring data, providing a quantitative basis for subsequent game theory decisions. Time-series feature extraction refers to processing time-series data to extract features that reflect the patterns of data change. These features can be used for subsequent model analysis or decision-making. For the extraction of temperature control energy consumption change features, the core is to analyze the dynamic correlation between the pool inlet / outlet water temperature and the ambient temperature, capturing the patterns of their changes over time. Specifically, a sliding window approach is used to process continuous water temperature data. Here, the sliding window is a data processing technique that continuously calculates the data within a fixed-length time window to capture local change features. By observing the average rate of change of water temperature within a specific time window, short-term fluctuation characteristics are grasped. At the same time, combined with the lag correlation analysis of ambient temperature, such as the delayed effect of the water temperature drop trend after a sudden change in ambient humidity, short-term fluctuations are combined with long-term trends to form a composite feature that comprehensively reflects energy consumption changes. The extraction of water quality parameter deviation characteristics focuses on abnormal changes in water quality parameters over time. By comparing real-time water quality data with historical normal ranges, the degree of parameter deviation from the baseline is identified, and smoothing is performed using an exponentially weighted moving average. Here, the exponentially weighted moving average is a time series smoothing method that assigns higher weights to recent data, effectively smoothing noise and highlighting data trend changes. This weakens the impact of random fluctuations and clearly presents the abnormal trends of water quality parameters, such as the sustained impact of a sudden drop in residual chlorine content or a sharp fluctuation in pH.

[0059] The specific calculation process for the above steps can be as follows: During time-series feature extraction, for the pool inlet and outlet water temperatures and ambient temperatures, an appropriate sliding window duration is first set. Within each window, the mean and rate of change of water temperature are calculated to reflect short-term water temperature fluctuations. Simultaneously, ambient temperature data is lagged by a certain time interval to analyze its correlation with water temperature changes. For example, the decrease in water temperature 10 minutes and 20 minutes after a change in humidity is calculated to quantify the lagged impact of environmental factors. Finally, this is integrated into a temperature control energy consumption change index that combines short-term and long-term characteristics. For water quality parameters, the normal range of each parameter is first determined based on historical data. Then, the difference between the current parameter and the baseline of the normal range is calculated in real time. An exponentially weighted moving average is used to smooth the differences over multiple consecutive time points, weakening the impact of random fluctuations and highlighting the trend of continuous deviation, resulting in a deviation characteristic that accurately reflects the degree of water quality anomalies.

[0060] Understandably, the principle behind constructing a game theory model is to treat water quality control and energy consumption control as two entities with conflicting objectives. By setting their respective payoff functions, the interests of both parties are quantified, thereby finding the optimal strategy to balance them. A game theory model is a mathematical model built upon game theory, used to analyze how multiple participants choose strategies to achieve their goals in an environment of mutual influence. The payoff function is the tool in the game theory model for quantifying the interests of participants, mathematically linking strategies with payoffs and reflecting the degree to which different strategies achieve the objective. The water quality control party aims to maximize the water quality compliance rate. Its adjustable circulation pump power and the start / stop status of the electric heater directly affect the stability of water quality parameters. For example, increasing the circulation pump power can accelerate water treatment, and proper start / stop of the electric heater can avoid the impact of abnormal water temperature on water quality. The corresponding payoff function will focus on the percentage of time the water quality meets the standard, while also considering the losses caused by frequent equipment operation, and using penalty terms to suppress unreasonable adjustments. The energy control side aims to minimize temperature control energy consumption. This can be achieved by adjusting the heat pump compressor frequency and the circulation pump's operating time. For example, reducing the compressor frequency can decrease energy consumption, but may affect the temperature control effect. Its payoff function is directly related to the energy consumption level, encouraging low-energy operation while meeting basic temperature control requirements. This game is a two-player non-zero-sum game, meaning the total payoff for both parties is not zero. An increase in one party's payoff does not necessarily come at the expense of a decrease in the other party's payoff; there is a possibility of achieving a win-win situation through strategic coordination.

[0061] In specific calculations, the water quality control benefit function is constructed based on the percentage of time that water quality parameters (such as residual chlorine and pH value) remain within the acceptable range. The higher the percentage, the higher the benefit. Simultaneously, when the circulating pump frequently adjusts its power or the electric heater frequently starts and stops, a corresponding penalty value is introduced to reduce the total benefit, thereby guiding a stable operating rhythm. The energy consumption control benefit function focuses on the actual amount of energy consumed; the lower the energy consumption, the higher the benefit. It also considers the operating efficiency of the heat pump to avoid excessively reducing energy consumption and resulting in substandard temperature control. A reasonable weighting is used to balance the relationship between energy consumption and temperature control.

[0062] Projecting features onto the policy space for discretization transforms continuously changing features into a finite number of actionable policy options, reducing the complexity of game solving. The policy space is the set of all possible policies; discretization converts continuously changing policies into a finite number of discrete options to reduce computational complexity. Since temperature control energy consumption changes and water quality parameter deviations are continuously changing, directly using them in a game model would result in an infinite number of policies, making it difficult to solve. By categorizing features according to certain rules—for example, dividing temperature control energy consumption changes into low, medium, and high levels, and water quality parameter deviations into normal, warning, and abnormal levels—a finite set of features can be formed. Each combination corresponds to a specific operational policy. For instance, when energy consumption is at a medium level and water quality deviation is at a warning level, the circulation pump uses medium power, and the compressor maintains a specific frequency, thus forming a finite policy set. A finite policy set is a collection of a finite number of discrete policies, each with a specific operational instruction, facilitating rapid model traversal and solution.

[0063] For example, in the above steps, the critical values ​​for feature classification can be determined based on historical data and actual operating experience. For instance, when the rate of change in water temperature is within a certain range, it can be classified as a medium-level energy consumption change; when the degree to which water quality parameters deviate from the baseline reaches a certain value, it can be classified as a warning level. Then, each feature combination is associated with the corresponding equipment adjustment method to ensure that each strategy has a clear operational direction, transforming the originally continuous feature space into a finite, enumerable strategy option, laying the foundation for the subsequent construction of the revenue matrix.

[0064] The principle behind constructing a payoff matrix for a multi-party game is to map each strategy combination in the discretized strategy set to the specific payoffs of the water quality control party and the energy consumption control party, forming a matrix that intuitively reflects the interests of both parties. The payoff matrix is ​​a table in the game model used to store the payoffs of each party under different strategy combinations. Rows and columns represent the strategies of different participants, and cell values ​​represent the payoffs under the corresponding strategies, clearly showing the correspondence between strategies and payoffs. Each cell in the matrix represents the payoff value obtained by the water quality control party and the energy consumption control party under a specific strategy. These payoff values ​​are calculated based on a pre-defined payoff function, clearly showing the impact of different strategies on the goals of both parties. For example, a certain strategy might increase the water quality compliance rate (high payoff for the water quality control party) but increase energy consumption (low payoff for the energy consumption party), thus providing a basis for finding the optimal strategy that balances the two.

[0065] During the calculation process, for each strategy combination in the strategy set, the water quality control party's payoff is first calculated based on parameters such as the circulating pump power and electric heater status under that strategy. This payoff is the payoff corresponding to the water quality compliance rate minus the penalty value for frequent equipment operation. Simultaneously, the energy consumption control party's payoff is calculated based on parameters such as compressor frequency and circulating pump runtime. This payoff value is derived from the energy consumption level. These payoff values ​​are then filled into the corresponding cells in the matrix, ensuring that the matrix fully reflects the payoffs of both parties under all possible strategies, becoming the core basis for solving the game.

[0066] The Lagrange multiplier method is used to solve mixed-strategy Nash equilibrium problems. The principle is to find an equilibrium state by coordinating the mixed strategies (i.e., the probability of each pure strategy being adopted) of the water quality control party and the energy consumption control party. In this state, neither party can improve its own payoff by changing its strategy alone, thus achieving a balance between maximizing the water quality compliance rate and minimizing temperature control energy consumption. Nash equilibrium is a stable state in game theory, where, given the strategies of other participants, no single participant can gain a higher payoff by changing its strategy alone. Mixed strategies refer to situations where participants do not choose a single strategy but rather select different pure strategies with a certain probability distribution; mixed-strategy Nash equilibrium is the stable state under this probability distribution. The Lagrange multiplier method is a mathematical method for solving constrained optimization problems. By introducing multipliers, it integrates the constraints into the objective function, transforming it into an unconstrained optimization problem. Here, it is used to handle constraints such as physical limitations of equipment operation, ensuring the feasibility of the optimal strategy.

[0067] Specifically, the probability distributions for different strategies adopted by the water quality control party and the energy consumption control party are first defined. Then, the expected returns for both parties under these probability distributions are calculated based on the payoff matrix. The constraints are incorporated into the objective function using the Lagrange multiplier method, and the probability distributions of both parties are iteratively adjusted until the expected returns of both parties reach a stable state, that is, changing the probability distribution of either party will not increase its own expected return. The corresponding strategy combination at this point is the optimal heat pump cycle regulation strategy, which includes specific adjustment commands such as cycle pump power, compressor frequency, and electric heater start / stop.

[0068] Through the above steps, firstly, by extracting time-series features, the dynamic changes in temperature control energy consumption and water quality can be accurately captured, providing high-quality input for the game theory model and ensuring that subsequent decisions are based on an accurate understanding of the system state. Secondly, the construction of the game theory model and the solution of the Nash equilibrium effectively coordinate the two conflicting objectives of water quality compliance and energy consumption control, avoiding the trade-offs caused by optimizing a single objective. For example, it avoids sacrificing water quality to reduce energy consumption, or excessively consuming energy to pursue absolute water quality compliance, thus achieving a dynamic balance between the two. In addition, the construction of a finite strategy set reduces the solution complexity, enabling the game theory model to run quickly in the actual system, meeting the needs of real-time control, and ensuring that the adjustment strategy can respond promptly to changes in the system state. The use of the payoff matrix makes the payoff relationship between the two parties clearly visible, providing an intuitive basis for solving the Nash equilibrium and improving the reliability of the solution results. Ultimately, the optimal heat pump cycle regulation strategy obtained through this process can minimize temperature control energy consumption, reduce ineffective equipment operation, and extend equipment life while ensuring that the pool water quality meets the standards. This allows the entire heat pump system to operate in a highly efficient and stable state, taking into account the needs of performance, economy, and equipment maintenance.

[0069] As an optional embodiment, in step S104, after inputting the hybrid strategies of the water quality control party and the energy consumption control party into the payoff matrix, and obtaining the optimal heat pump cycle regulation strategy by performing Nash equilibrium solution of the hybrid strategy using the Lagrange multiplier method, the method further includes: inputting the optimal heat pump cycle regulation strategy into a Bayesian network, updating the conditional probability table of each node executing the corresponding heat pump cycle regulation strategy based on real-time data, and dynamically optimizing the optimal heat pump cycle regulation strategy based on the conditional probability table. The Bayesian network includes at least: environmental temperature and humidity nodes, population density nodes, water quality parameter nodes, and equipment status nodes.

[0070] Specifically, in terms of implementation principle, the core of this step is to leverage the probabilistic reasoning capabilities of Bayesian networks to allow the optimal heat pump cycle regulation strategy to dynamically adapt to changes in system state based on real-time data. A Bayesian network is a probabilistic model that represents the dependencies between variables using a directed graph and quantifies these dependencies using conditional probability tables. Each node represents a random variable, the edges between nodes represent causal relationships between variables, and the conditional probability table stores the conditional probabilities of a node taking different values ​​from its parent node. This Bayesian network includes nodes related to ambient temperature and humidity, population density, water quality parameters, and equipment status. These nodes collectively constitute the key factors affecting the effectiveness of the heat pump regulation strategy. Changes in ambient temperature and humidity affect the heat load, increases or decreases in population density may alter the rate of water pollution, fluctuations in equipment status affect the efficiency of the regulation strategy, and water quality parameters directly reflect the strategy's objective.

[0071] When the optimal heat pump cycle regulation strategy is input into the Bayesian network, the network first establishes probabilistic relationships between each node and the strategy's execution effect based on initial data. Examples include "the probability of water quality parameters meeting standards when using a certain strategy under high population density" and "the probability of energy consumption control effectiveness of a certain strategy under low ambient temperature." With continuous input of real-time data, the network continuously updates the conditional probability tables of each node. For instance, if a sudden increase in population density is detected, the conditional probability from the "population density node" to the "water quality parameter node" is adjusted immediately to reflect the change in the intensity of the population's impact on water quality. Similarly, if a slight abnormality occurs in the equipment status, the probability relationship between the "equipment status node" and the "strategy execution effect node" is updated to reflect the impact of decreased equipment performance on the strategy. This updating essentially corrects the dependency strength between variables through probabilistic reasoning, allowing the network to more accurately capture the impact of various factors on the strategy under current operating conditions. Based on the updated conditional probability tables, the optimal strategy is dynamically optimized, ensuring that it can still balance the goals of water quality compliance and energy consumption control under new conditions.

[0072] In the specific calculation process, the optimal heat pump cycle regulation strategy first needs to be transformed into a "strategy execution variable" that can be recognized by the Bayesian network, and then associated with each node in the network. For example, the circulation pump power regulation in the strategy is related to the "equipment status node," and the start / stop of the electric heater is related to the "ambient temperature and humidity node." Initially, based on historical operating data, the conditional probability table of each node is assigned values ​​to determine the probability distribution of each node's state under different strategies. For example, when the strategy of "compressor low-frequency operation + circulation pump short-term operation" is adopted, the initial probability of the "water quality parameter node" being in the "compliant" state is determined. As the system operates, real-time collected data such as ambient temperature and humidity, people count, water quality test values, and equipment operating parameters are input into the Bayesian network as observation evidence to update the node probabilities. For example, when real-time data shows a sudden increase in ambient humidity and a rise in people density, the network will first update the "ambient temperature and humidity node" to the "high humidity" state and the "people density node" to the "high density" state, and then adjust the probabilities of the associated nodes through a probability propagation algorithm. For example, the probability of a "water quality parameter node" failing to meet standards increases due to high humidity and high density, which in turn affects the probability distribution of "strategy execution effect nodes". In this case, the network will re-evaluate the execution effect probability of the current optimal strategy based on the updated conditional probability table. If it finds that the probability of "water quality compliance rate decreasing" or "energy consumption abnormally increasing" exceeds the threshold under the new conditions, the strategy will be optimized accordingly. For example, the operating time of the circulating pump may be extended while maintaining the compressor frequency, or the start-stop frequency of the electric heater may be appropriately reduced while ensuring water quality, so that the adjusted strategy is better matched to the real-time operating conditions.

[0073] From a technical perspective, this process significantly improves the adaptability and robustness of the optimal heat pump cycle regulation strategy. The introduction of Bayesian networks frees the strategy from dependence on fixed models, enabling it to dynamically correct itself based on real-time data and effectively cope with uncertainties in the system. For example, sudden peak traffic can cause rapid changes in water quality parameters, or sudden changes in ambient temperature and humidity can affect the heat load balance. In such cases, the strategy optimized by Bayesian networks can quickly adjust to offset these disturbances, preventing water quality from exceeding standards or energy consumption from surging. Simultaneously, the dynamic updating of the conditional probability table allows the system to continuously learn the correlation patterns between various factors. As operating time increases, the probability model becomes increasingly closer to actual operating conditions, and the accuracy of strategy optimization also improves. For instance, after long-term operation, the system can more accurately predict "the degree of impact of traffic density on water quality under a certain combination of ambient temperature and humidity," thereby adjusting the circulation pump power or compressor frequency in advance to achieve proactive control "preventing problems before they occur."

[0074] In addition, this dynamic optimization can reduce the invalid operation of the equipment. When the equipment status is slightly abnormal, the strategy will appropriately reduce its dependence on it based on the updated probability table, avoid failure caused by equipment overload, indirectly extend the service life of the equipment, and ultimately enable the heat pump system to always maintain a balance between water quality compliance and energy consumption control in complex and ever-changing real-world scenarios, achieving more efficient and stable operation.

[0075] As an optional embodiment, in step S104, after projecting the temperature control energy consumption change characteristics and water quality parameter deviation characteristics onto the strategy space for discretization to obtain a finite strategy set, the method further includes: constructing a pool twin based on historical data, inputting the finite strategy set into the pool twin, simulating the water quality energy consumption evolution path of different strategy combinations in the finite strategy set; using the NSGA-II algorithm to solve the Pareto front to screen for non-dominated solutions; monitoring the heat pump compressor current, electric heater temperature, and circulating pump pressure in real time to obtain multi-dimensional monitoring parameters, and selecting the optimal solution from the Pareto front based on the multi-dimensional monitoring parameters obtained in real time to generate the corresponding water quality energy consumption optimization strategy; and combining the water quality energy consumption optimization strategy to optimize the water quality energy consumption adjustment strategy in the optimal heat pump circulation adjustment strategy to further improve water quality and reduce energy consumption.

[0076] In terms of implementation principle, this step combines digital twins with multi-objective optimization algorithms to construct a closed-loop optimization mechanism of "virtual simulation - solution space filtering - real-time adaptation". The swimming pool twin, built based on historical data, is a digital mapping of the real swimming pool system. It integrates physical models such as the pool's heat exchange patterns, water quality evolution characteristics, and equipment operating characteristics, enabling it to reproduce the operating state of the real system in a virtual environment. Inputting a finite set of strategies into the twin essentially involves virtually testing the long-term impact of different strategy combinations on water quality and energy consumption. This simulates the change path of water quality parameters (such as residual chlorine and pH value) over time for each strategy, as well as the dynamic curve of energy consumption (such as the total energy consumption of compressors, electric heaters, and circulation pumps). This allows for a comprehensive evaluation of the potential effects of the strategies without interfering with the real system.

[0077] The NSGA-II algorithm was introduced to resolve the conflict between achieving water quality standards and reducing energy consumption. When improving one objective might worsen the other, the algorithm uses multi-objective optimization to select a set of non-dominated solutions (i.e., the Pareto front). Each solution in this set cannot improve the other without compromising one objective. For example, one solution might maintain a high water quality compliance rate while having low energy consumption, while another might have extremely low energy consumption at a slightly lower compliance rate. These solutions collectively constitute the alternatives that balance the two objectives. Real-time monitoring of multi-dimensional parameters (heat pump compressor current reflecting load status, electric heater temperature reflecting heating efficiency, and circulating pump pressure reflecting water flow resistance) is used to capture real-time deviations in the actual system's operating conditions. Since the twin simulation may differ from reality, these parameters are used to determine the current system's operating status (e.g., whether equipment is inefficient or whether the load has changed abruptly). The algorithm then selects the optimal solution from the Pareto front that best matches the real-time state, ensuring that the generated optimization strategy adapts to fluctuations in the real environment. Finally, this optimization strategy is combined with the original heat pump cycle regulation strategy to specifically optimize the water quality and energy consumption regulation portion, achieving further synergy between the two.

[0078] In the specific calculation process, a twin model must first be constructed based on the historical operating data of the swimming pool (including water temperature changes in different seasons, water quality parameter fluctuations, equipment energy consumption records, etc.). By calibrating the model parameters (such as heat loss coefficient, water quality reaction rate, equipment efficiency curve), the output of the virtual system is made consistent with the real data. Subsequently, the finite set of discretized strategies (such as combinations of "high frequency compressor + long-time circulation pump" and "low frequency compressor + intermittent electric heater") are input into the twin one by one, the simulation duration is set (such as the next 24 hours), and the model is run to obtain the water quality evolution path (such as the time from residual chlorine to deviation from compliance, pH fluctuation range) and energy consumption evolution path (such as peak energy consumption per unit time, total cumulative energy consumption) corresponding to each strategy.

[0079] Next, the water quality compliance rate (e.g., the percentage of time spent meeting standards) and total energy consumption corresponding to these paths are used as optimization objectives and input into the NSGA-II algorithm. The algorithm generates candidate solutions through population initialization, selects the better-performing solutions based on non-dominated sorting and crowding distance, and iteratively optimizes through selection, crossover, and mutation operations until it finally converges to obtain the Pareto front, i.e., the set of non-dominated solutions. Then, parameters such as heat pump compressor current, electric heater temperature, and circulating pump pressure are collected in real time, and the current system state is determined by analyzing these parameters. For example, an abnormally high compressor current may indicate excessive load, and a low circulating pump pressure may indicate changes in pipeline resistance. Based on these state characteristics, the solution best suited to the current state is selected from the Pareto front: if the equipment load is high, a solution with a more gradual energy consumption distribution is preferred; if the risk of water quality fluctuations is high, a solution with better water quality stability is prioritized. The selected solution is transformed into a specific water quality and energy consumption optimization strategy (such as adjusting the power of the circulating pump to balance water flow and energy consumption, and optimizing the start-up and shutdown timing of the electric heater to reduce water quality fluctuations). Then, this strategy is used to correct the water quality and energy consumption adjustment parts (such as the running time of the circulating pump and the frequency range of the compressor) in the original optimal heat pump cycle regulation strategy.

[0080] From a technical perspective, this series of steps significantly improves the applicability and optimization accuracy of the strategy. The virtual simulation of the pool twin avoids the risks of testing the strategy in a real system, reducing water quality exceeding standards or equipment damage caused by inappropriate strategies. Simultaneously, extensive virtual testing covers more operating conditions, making the solution space more comprehensive for subsequent selection. The Pareto front solved by the NSGA-II algorithm overcomes the limitations of single-objective optimization, providing multiple solutions that balance water quality and energy consumption, avoiding the situation where pursuing one objective excessively sacrifices another.

[0081] The introduction of real-time monitoring parameters solves the discrepancy between virtual simulation and the real system, allowing the selection of the optimal solution to dynamically adapt to the actual operating conditions. For example, when equipment suddenly becomes inefficient, the strategy can quickly adjust to maintain water quality and avoid a surge in energy consumption, thus improving the system's robustness. The final optimized heat pump cycle regulation strategy achieves a more refined balance between water quality compliance and energy consumption control. It reduces treatment costs caused by water quality fluctuations and lowers unnecessary energy consumption. Furthermore, because the strategy is more closely aligned with the actual operating conditions of the equipment, it indirectly reduces extreme equipment operation and extends the equipment's lifespan, resulting in a comprehensive improvement in the system's economy, stability, and durability.

[0082] Step S105: Based on the optimal heat pump cycle adjustment strategy and the heat pump power control parameters, the pool heat pump is electrically heated to improve the pool temperature control efficiency and ensure the safety of pool operation.

[0083] In terms of implementation principle, the core of step S105 is to construct a precise control logic for the electric heating system through the synergistic effect of the optimal heat pump cycle regulation strategy and the heat pump power control parameters, balancing temperature regulation efficiency and operational safety. The optimal heat pump cycle regulation strategy provides a macroscopic synergistic framework for electric heating control, clarifying the coordination rhythm between the electric heater and equipment such as the heat pump compressor and circulation pump. For example, the strategy defines the start-up threshold of electric heating (such as when the heat pump alone cannot maintain the set temperature) and the percentage of its operating time, ensuring that electric heating only intervenes when necessary, avoiding conflicts with the main function of the heat pump. The heat pump power control parameters provide a microscopic basis for adjustment. Parameters such as compressor frequency and power range directly determine the output intensity of electric heating: when the compressor is running at high frequency and there is still a temperature gap, electric heating needs to quickly compensate for the heat with higher power; when the compressor is in a low-frequency energy-saving state, electric heating assists in maintaining temperature stability with low power.

[0084] Meanwhile, a safety assurance mechanism runs throughout the entire control process. By real-time correlation of data from devices such as temperature sensors and current monitors, a dual protection logic is constructed: On the one hand, when the pool water temperature approaches the upper limit of the set range, the electric heating power is automatically reduced or shut off in advance to prevent the water temperature from exceeding the standard. On the other hand, when fault signals such as abnormal electric heater current or a sudden rise in heating element temperature are detected, the electric heating circuit is immediately cut off to avoid equipment damage or safety hazards. This collaborative control based on a strategy framework, parameter guidance, and safety monitoring essentially ensures that the electric heating system serves as an effective supplement to heat pump temperature control without becoming an energy consumption burden or a safety risk.

[0085] In the specific calculation process, the system first converts the electric heating-related clauses in the optimal heat pump cycle regulation strategy into executable control rules. For example, the statement in the strategy, "When the real-time water temperature is 2°C lower than the lower limit of the set range and the compressor frequency has reached the upper limit, start electric heating," will be parsed into specific trigger conditions. By comparing the difference between the real-time collected pool water temperature and the set range, and combining this with whether the compressor's current operating frequency has reached the upper limit threshold in the power control parameters, it is determined whether electric heating needs to be started.

[0086] Once the startup conditions are met, the system determines the initial output intensity of the electric heater based on the dynamic power range in the heat pump power control parameters. If the parameters indicate that the current heat pump system is under high load (e.g., the compressor frequency is close to the rated value), the electric heater will start at medium power to avoid overloading the circuit. If the heat pump is under low load, the electric heater can quickly replenish heat at higher power, shortening the temperature rise time. During operation, the system continuously adjusts through closed-loop feedback: for every 0.5℃ reduction in the deviation between the real-time water temperature and the set value, the electric heater power is reduced proportionally until the water temperature enters the set range and then gradually shuts down. Simultaneously, the safety monitoring module collects the electric heater's operating data (such as the surface temperature of the heating element and the loop current) at fixed intervals. If the data exceeds the preset safety range, an adjustment command is immediately triggered. For example, the power is reduced to within the safety threshold, or in extreme cases, it is directly shut down and an alarm is issued, ensuring that the control process always operates within the safety boundaries.

[0087] From a technical perspective, this electric heating control method significantly improves the overall performance of pool temperature control. In terms of efficiency, through the synergy of optimal strategies and power parameters, the timing and output intensity of electric heating are precisely controlled, avoiding the problems of "blind start-up" or "excessive power" in traditional control: when the heat pump is sufficient to maintain the temperature, the electric heating remains dormant, reducing ineffective energy consumption. When a temperature gap occurs, the electric heating can quickly respond and match the required power, significantly shortening the time for the water temperature to return to the set range from the deviation and significantly reducing temperature fluctuations.

[0088] In terms of safety, the dual protection mechanism effectively mitigates potential risks: the possibility of water temperature overshoot is controlled to an extremely low level, ensuring that the pool water temperature is always within a range that is comfortable for humans and safe for the equipment. The probability of electric heater failure due to overload or overheating is significantly reduced, extending the stability and service life of the equipment. In addition, this synergistic control indirectly optimizes the energy consumption structure of the entire heat pump system. Electric heating, as a "supplementary heat source" rather than a "primary heat source," has its energy consumption share reasonably reduced. While ensuring temperature control, it further reduces the total energy consumption of the system, achieving multiple goals of efficiency, safety, and energy saving.

[0089] Further optionally, after performing electric heating control on the pool heat pump according to the heat pump circulation regulation strategy and the heat pump power control parameters in step S105, the method further includes: real-time monitoring of the heat pump compressor current, electric heater temperature, and circulation pump pressure to obtain multi-dimensional equipment monitoring parameters; if any equipment parameter in the multi-dimensional monitoring equipment parameters exceeds a threshold, an alarm is triggered and the system switches to the backup pool electric heating equipment; if the ambient temperature in the multi-source monitoring data is lower than a set threshold and the heat pump stops, the circulation pump is automatically started and the pool electric heating equipment is turned on to prevent the pipeline from freezing.

[0090] The core of this step is to build a dual-protection mechanism based on multi-dimensional monitoring, addressing both equipment malfunction risks and preventing system failures caused by environmental factors. Real-time monitoring of equipment parameters such as heat pump compressor current, electric heater temperature, and circulating pump pressure is crucial because these parameters directly reflect the equipment's operating status. Abnormal compressor current may indicate overload or mechanical failure; a sudden rise in electric heater temperature may suggest aging heating elements or poor heat dissipation; and abnormal circulating pump pressure may indicate pipe blockage or pump damage. Comparing these parameters with preset thresholds allows for rapid identification of abnormal equipment conditions. If any parameter exceeds the limit, an alarm is triggered, and the system switches to backup equipment. Essentially, this redundancy design prevents system paralysis due to a single equipment failure, ensuring uninterrupted heating. The control logic for ambient temperature stems from the potential threat of low temperatures to the piping system. When the ambient temperature is below the set threshold, if the heat pump stops and the circulating pump is not running, the water in the pipes may freeze and expand due to the low temperature, causing pipe rupture or equipment damage. Automatically starting the circulating pump keeps the water flowing, reducing the risk of freezing. Simultaneously, activating the pool's electric heating equipment indirectly raises the pipe temperature by heating the water, forming a dual anti-freeze protection. This logical design considers both the operational safety of the equipment itself and the impact of environmental factors on the system, thus constructing a comprehensive security protection system.

[0091] From a technical perspective, this mechanism significantly improves the reliability and safety of the pool heat pump system. Regarding equipment protection, real-time monitoring and backup switching functions drastically shorten fault response time, preventing heating interruptions due to single device malfunctions, ensuring stable pool water temperature within the set range, and reducing the impact of temperature fluctuations on the swimming experience. Simultaneously, timely alarms alert maintenance personnel to troubleshoot faults, reducing the risk of equipment failure escalation and extending the overall lifespan of the equipment. In terms of environmental adaptability, the low-temperature anti-freeze logic effectively solves the problem of pipe freezing in cold regions or during winter operation, reducing maintenance costs and downtime caused by freezing damage, allowing the system to maintain basic functionality even in extreme environments. Furthermore, this automated protection mechanism reduces the need for manual intervention, lowering the workload of maintenance personnel, and is particularly suitable for large pools or unattended scenarios, making the entire heat pump system operate more stably and efficiently under complex conditions, balancing both functionality and safety requirements.

[0092] In this embodiment, the comprehensive performance of the pool heating system is significantly improved through deep integration of multi-dimensional data fusion and intelligent control strategies. At the energy efficiency optimization level, a coupled heating model based on a spatiotemporal attention mechanism accurately captures the spatiotemporal correlation between ambient temperature and humidity and pool water temperature, dynamically adjusts the heat pump compressor frequency, and intelligently sets the electric heater start-up threshold, ensuring the system operates continuously within a high-efficiency range and reducing overall energy consumption. The water quality and temperature linkage control mechanism relies on a dynamic Bayesian network to construct an adaptive rule base. Through multi-objective game optimization, it achieves a dynamic balance between water quality compliance rate and energy consumption, solving the problem of water quality fluctuations and energy consumption imbalances caused by single-parameter control in traditional equipment. Furthermore, the synergistic application of multi-objective game theory and digital twin technology further enhances temperature control accuracy and energy consumption optimization. The Nash equilibrium algorithm is used to solve the optimal solution for maximizing water quality compliance rate and minimizing temperature control energy consumption. Combined with the simulation and prediction function of the digital twin model for water quality evolution paths, the impact of sudden environmental changes on water temperature can be predicted in advance, and operating parameters can be proactively adjusted to control water temperature fluctuations within ±0.3℃. In terms of safety and reliability, an abnormal operating condition circuit breaker mechanism and a multi-level protection strategy are constructed. The compressor operating status and water quality parameters are monitored in real time, and graded responses (such as reducing power and starting the backup pump) are triggered to avoid equipment overload. The adaptive learning mechanism continuously optimizes the model parameters, significantly extending the equipment life and reducing operation and maintenance costs.

[0093] After introducing the methods of exemplary embodiments of this application, the following references are made. Figure 2This application describes an exemplary embodiment of an electric heating coupling control system based on a swimming pool heat pump. The device includes: a data acquisition module for real-time acquisition of multi-source monitoring data of the swimming pool; the multi-source monitoring data includes at least: pool inlet and outlet water temperature parameters, ambient temperature, ambient humidity, and water quality parameters; a coupling heating prediction module for calculating the comprehensive temperature difference between the pool inlet water temperature and the ambient temperature based on the multi-source monitoring data, and calculating the pool's heat loss rate based on the pool outlet water temperature and the pool inlet water temperature; inputting the comprehensive temperature difference and heat loss rate of the pool into a coupling heating model based on a spatiotemporal attention mechanism to generate dynamic heat pump power control parameters; wherein, the dynamic heat pump... The power control parameters include at least: the compressor frequency for adjusting the heating power of the pool heat pump, and the electric heater start-up threshold; a linkage prediction module for inputting the multi-source monitoring data into the water quality and temperature linkage control model, using the Nash equilibrium algorithm to solve for the optimal heat pump cycle regulation strategy with the game objective of maximizing the water quality compliance rate and minimizing temperature control energy consumption; wherein, the water quality and temperature linkage control model uses an adaptive control rule base constructed with a dynamic Bayesian network for strategy optimization; and a control module for electrically controlling the pool heat pump according to the optimal heat pump cycle regulation strategy and the heat pump power control parameters to improve the pool temperature control efficiency and ensure the safety of pool operation. The above system can implement the steps described in the above method implementation, and the specific implementation methods of each step will not be repeated here.

[0094] After introducing the methods and systems of the exemplary embodiments of this application, a terminal device of the exemplary embodiments of this application will be described next. The terminal device can implement the steps described in the above method embodiments, and the specific implementation of each step will not be repeated here.

[0095] After introducing the methods, systems, and terminal devices of exemplary embodiments of this application, the following references will be made. Figure 3 The computer-readable storage medium of exemplary embodiments of this application will be described, please refer to... Figure 3 The computer-readable storage medium shown is an optical disc 30, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it implements the steps described in the above method embodiments. The specific implementation methods of each step will not be repeated here.

[0096] It should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for electric heating coupling control based on a swimming pool heat pump, characterized in that, The method comprises: Real-time acquisition of multi-source monitoring data of the swimming pool; the multi-source monitoring data at least comprises: swimming pool inlet and outlet water temperature parameters, ambient temperature, ambient humidity, water quality parameters; Based on the multi-source monitoring data, the comprehensive temperature difference value between the swimming pool inlet water temperature and the ambient temperature is calculated, and the heat loss rate of the swimming pool is calculated based on the swimming pool outlet water temperature and the swimming pool inlet water temperature; The comprehensive temperature difference value and the heat loss rate of the swimming pool are input into a coupling heating model based on a space-time attention mechanism to generate a dynamic heat pump power control parameter; wherein the dynamic heat pump power control parameter at least comprises: a compressor frequency for adjusting the heating power of the swimming pool heat pump, and an electric heater starting threshold; The multi-source monitoring data is input into a water quality temperature linkage control model to maximize the water quality compliance rate and minimize the temperature control energy consumption, and a Nash equilibrium algorithm is used to solve the optimal heat pump circulation adjustment strategy, which comprises: performing time sequence feature extraction on the swimming pool inlet and outlet water temperature parameters, the ambient temperature in the multi-source monitoring data to obtain temperature control energy consumption change characteristics; performing time sequence feature extraction on the water quality parameters in the multi-source monitoring data to obtain water quality parameter deviation characteristics; constructing a game model, setting a water quality control party and an energy consumption control party, and corresponding water quality control benefit function and energy consumption control benefit function; wherein the target of the water quality control party is to maximize the water quality compliance rate, and the adjustable variable is the circulating pump power and the electric heater start-stop state; the target of the energy consumption control party is to minimize the temperature control energy consumption, and the adjustable variable is the heat pump compressor frequency and the circulating pump operation time length proportion; the temperature control energy consumption change characteristics and the water quality parameter deviation characteristics are projected to the strategy space for discretization processing to obtain a limited strategy set; based on the water quality control benefit function and the energy consumption control benefit function, a benefit matrix of multi-party game is constructed, and each matrix cell stores the corresponding benefit value of the two parties in the discrete characteristic points; the mixed strategy of the water quality control party and the energy consumption control party is input into the benefit matrix, and the mixed strategy Nash equilibrium solution is obtained by Lagrange multiplier method to obtain the optimal heat pump circulation adjustment strategy; wherein the adaptive control rule library constructed by the dynamic Bayesian network in the water quality temperature linkage control model is used for strategy optimization; According to the optimal heat pump circulation adjustment strategy and the heat pump power control parameter, the electric heating control of the swimming pool heat pump is performed to improve the swimming pool temperature control efficiency and ensure the swimming pool operation safety.

2. The pool heat pump based electric heating coupled control method of claim 1, wherein, The dynamic heat pump power control parameter is generated by inputting the comprehensive temperature difference value and the heat loss rate of the swimming pool into a coupling heating model based on a space-time attention mechanism, comprising: Performing time sequence feature extraction on the comprehensive temperature difference value and the heat loss rate of the swimming pool to obtain a dynamic feature vector; the dynamic feature vector at least comprises: water temperature gradient change rate, ambient temperature and humidity covariance; input the dynamic feature vector into a federated learning architecture, aggregate historical operation data of multiple swimming pools, train a coupled heating model based on a spatiotemporal attention mechanism, and generate a dynamic heat pump power control parameter to be set by using the trained coupled heating model; in the federated learning architecture, a dynamic learning weight is allocated to each swimming pool to improve the diversity of historical swimming pool data in the federated learning architecture.

3. The pool heat pump based electric heating coupled control method of claim 2, wherein, The time sequence feature extraction is performed on the comprehensive temperature difference value and the heat loss rate of the swimming pool to obtain a dynamic feature vector, including: The dynamic feature vector is subjected to multi-scale analysis, and a high-frequency component obtained by decomposition is taken as a water temperature gradient change rate, and a low-frequency component obtained by decomposition is taken as an environmental temperature and humidity covariance, which are respectively input into a gated recurrent unit for feature enhancement; A double-channel LSTM network is used to process the water temperature gradient change rate and the environmental temperature and humidity covariance after feature enhancement, respectively, to output a first channel feature vector and a second channel feature vector; The dynamic heat pump power control parameter to be set is generated by using the trained coupled heating model, including: The feature interaction weight between the first channel feature vector and the second channel feature vector is calculated through a cross-channel attention mechanism; The heat load state in the next two hours is collaboratively predicted by combining the feature interaction weight, the first channel feature vector, and the second channel feature vector to obtain a heat load fluctuation interval in the next two hours; A compressor frequency adjustment sequence is generated by using a model predictive control (MPC) algorithm according to the heat load fluctuation interval in the next two hours; A compressor frequency adjustment parameter in the dynamic heat pump power control parameter is generated based on the compressor frequency adjustment sequence; The compressor frequency adjustment parameter is used to indicate that the setting value of the compressor frequency is updated every five minutes in the next two hours to ensure that the compressor operates in a preset efficient interval.

4. The pool heat pump based electric heating coupling control method according to claim 3, wherein, After the model predictive control (MPC) algorithm is used to generate the compressor frequency adjustment sequence according to the heat load fluctuation interval in the next two hours, the following steps are further included: A heating compensation coefficient is calculated by a nonlinear compensation function based on the environmental temperature and humidity covariance after feature enhancement; The heat load fluctuation interval obtained by prediction is dynamically compensated by using the heating compensation coefficient to correct abnormal values in the heat load fluctuation interval in the next two hours, and an electric heater starting threshold value matched with the heat load fluctuation interval is generated; A compressor on-off instruction in the dynamic heat pump power control parameter is generated based on the compressor frequency adjustment sequence and the electric heater starting threshold value matched with the heat load fluctuation interval, The compressor on-off instruction is used to indicate the starting or stopping operation timing of the compressor in the next two hours to ensure that the swimming pool heat pump remains in a set temperature interval.

5. The pool heat pump based electric heating coupled control method of claim 1, wherein, After the mixed strategy of the water quality control party and the energy consumption control party is input into a payoff matrix, and a mixed strategy Nash equilibrium solution is obtained by using the Lagrange multiplier method, the following steps are further included: inputting the optimal heat pump cycle regulation strategy into a Bayesian network, updating a conditional probability table of each node executing a corresponding heat pump cycle regulation strategy based on real-time data, and dynamically optimizing the optimal heat pump cycle regulation strategy based on the conditional probability table; The Bayesian network at least includes: an environment temperature and humidity node, a human flow density node, a water quality parameter node, and a device state node.

6. The pool heat pump based electric heating coupling control method according to claim 1, wherein, After projecting the temperature control energy consumption change characteristics and the water quality parameter deviation degree characteristics into the strategy space for discretization processing to obtain the limited strategy set, the method further includes: Based on historical data, a swimming pool twin is constructed, the limited strategy set is input into the swimming pool twin, and the water quality energy consumption evolution path of different strategy combinations in the limited strategy set is simulated; The NSGA-II algorithm is used to solve the Pareto front to screen the non-inferior solution set; Real-time monitoring of heat pump compressor current, electric heater temperature, and circulating pump pressure is performed to obtain multi-dimensional monitoring parameters, and an optimal solution is selected from the Pareto front according to the multi-dimensional monitoring parameters obtained through real-time monitoring, and a corresponding water quality energy consumption optimization strategy is generated; In combination with the water quality energy consumption optimization strategy, the water quality energy consumption regulation strategy in the optimal heat pump cycle regulation strategy is optimized to further improve water quality and reduce energy consumption.

7. The pool heat pump based electric heating coupling control method according to claim 1, wherein, After the electric heating control of the swimming pool heat pump is performed according to the heat pump cycle regulation strategy and the heat pump power control parameter, the method further includes: Real-time monitoring of heat pump compressor current, electric heater temperature, and circulating pump pressure is performed to obtain multi-dimensional monitoring parameters; If any of the multi-dimensional monitoring device parameters exceeds a threshold value, an alarm is triggered and a backup swimming pool electric heating device is switched to; If the environment temperature in the multi-source monitoring data is less than a set threshold value and the heat pump is stopped, the circulating pump is automatically started and the swimming pool electric heating device is turned on to prevent pipeline freezing.

8. A pool heat pump based electric heating coupled control system, characterized in that, The system performs the electric heating coupling control method based on a swimming pool heat pump according to any one of claims 1 to 7, and the system includes: A collection module is configured to collect multi-source monitoring data of a swimming pool in real time, and the multi-source monitoring data at least includes: swimming pool inlet and outlet water temperature parameters, environment temperature, environment humidity, and water quality parameters. A coupling heating prediction module is configured to calculate a comprehensive temperature difference value between the swimming pool inlet water temperature and the environment temperature based on the multi-source monitoring data, and calculate a heat loss rate of the swimming pool based on the swimming pool outlet water temperature and the swimming pool inlet water temperature; input the comprehensive temperature difference value and the heat loss rate of the swimming pool into a coupling heating model based on a space-time attention mechanism to generate a dynamic heat pump power control parameter; wherein the dynamic heat pump power control parameter at least includes: a compressor frequency for adjusting the heating power of the swimming pool heat pump, and an electric heater starting threshold value. A linkage prediction module is configured to input the multi-source monitoring data into a water quality and temperature linkage control model to maximize the water quality compliance rate and minimize the temperature control energy consumption as a game target, and use a Nash equilibrium algorithm to solve an optimal heat pump cycle regulation strategy; wherein the water quality and temperature linkage control model uses an adaptive control rule library constructed by a dynamic Bayesian network for strategy optimization. The linkage prediction module is specifically used for: performing time sequence feature extraction on the pool inlet and outlet water temperature parameters and the environmental temperature in the multi-source monitoring data to obtain a temperature control energy consumption change feature; performing time sequence feature extraction on the water quality parameters in the multi-source monitoring data to obtain a water quality parameter deviation degree feature; constructing a game model, setting a water quality control party and an energy consumption control party, and corresponding water quality control benefit function and energy consumption control benefit function; wherein, the target of the water quality control party is to maximize the water quality compliance rate, and the adjustable variable is the circulating pump power and the electric heater start-stop state; the target of the energy consumption control party is to minimize the temperature control energy consumption, and the adjustable variable is the heat pump compressor frequency and the circulating pump operation time length proportion; projecting the temperature control energy consumption change feature and the water quality parameter deviation degree feature to the strategy space for discretization processing to obtain a limited strategy set; constructing a benefit matrix of multi-party game based on the water quality control benefit function and the energy consumption control benefit function, and each matrix cell stores the corresponding benefit value of the two parties in the discrete feature point; inputting the mixed strategy of the water quality control party and the energy consumption control party into the benefit matrix, and executing the mixed strategy Nash equilibrium solving through the Lagrange multiplier method to obtain the optimal heat pump circulation adjustment strategy; The control module is used for performing electric heating control on the pool heat pump according to the optimal heat pump circulation adjustment strategy and the heat pump power control parameter, so as to improve the pool temperature control efficiency and guarantee the pool operation safety.

9. A terminal device, comprising: The terminal device comprises at least one processor, a memory and an input-output unit; wherein the memory is used for storing a computer program, and the processor is used for calling the computer program stored in the memory to execute the pool heat pump-based electric heating coupling control method in any one of claims 1 to 7.

Images