Intelligent management system for integrated energy station

Abstract

The present application relates to the technical field of energy management, and more particularly to a kind of wisdom management system of integrated energy station, the system first obtains photovoltaic output fluctuation and heat pump load full temperature field monitoring data;Second, for different meteorological conditions, the power mutation caused by cloud cover is analyzed, and the instability degree of photovoltaic output is quantified;Further, by evaluating the synchronization of energy supply and demand and the consistency of power change, the matching degree of supply and demand and the coupling correlation of power response are obtained respectively;Finally, the energy efficiency optimization potential is determined by comprehensively considering the above three indexes, and adaptive collaborative control optimization strategy is generated accordingly.The present application realizes the deep coupling analysis and accurate matching of source-load characteristics, effectively improves the stability of system, renewable energy consumption rate and overall energy efficiency level under complex weather conditions.

Description

Technical Field

[0001] This invention relates to the field of energy management technology, and in particular to a smart management system for an integrated energy station. Background Technology

[0002] With the deepening implementation of the dual-carbon strategy, multi-energy complementary integrated energy systems are being widely used as an important vehicle for improving energy utilization efficiency and promoting the consumption of renewable energy. Among them, the photovoltaic-energy storage-heat pump integrated system has become a typical integrated energy solution in scenarios such as industrial parks and buildings due to its clean power generation, flexible energy storage, and efficient heating and cooling capabilities. However, in actual operation, photovoltaic power generation is significantly affected by weather conditions, exhibiting strong randomness and intermittency, which can easily lead to system power fluctuations and affect power supply stability; as the main energy-consuming equipment, the heat pump unit's heat load demand changes dynamically with factors such as ambient temperature and user behavior, resulting in complex operating characteristics.

[0003] Existing control strategies mostly employ independent regulation or simple linkage modes, lacking in-depth coupling analysis of the photovoltaic output fluctuation characteristics and heat pump load response characteristics. This makes it difficult to achieve precise matching and global energy efficiency optimization between the source, load, and storage systems. Especially under typical meteorological conditions such as rapid cloud movement and sudden temperature changes, the superposition of sudden changes in photovoltaic output and heat load fluctuations can easily lead to problems such as frequent charging and discharging of energy storage, decreased energy utilization, and even degradation of equipment operating efficiency. Therefore, there is an urgent need to construct an intelligent management method that can comprehensively sense the system's operating status, quantitatively analyze the coupling relationships of multiple energy flows, and generate adaptive collaborative control strategies accordingly, in order to improve the safety, stability, and overall energy efficiency of multi-energy complementary systems.

[0004] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this invention is to provide a smart management system for integrated energy stations, which aims to solve the technical problem that existing control strategies lack in-depth coupling analysis of photovoltaic power output fluctuations and heat pump load response characteristics, making it difficult to achieve accurate matching of source-load-storage and global energy efficiency optimization under complex weather conditions.

[0006] To achieve the above objectives, the present invention provides a smart management system for an integrated energy station, the system comprising:

[0007] The data sensing module is used to acquire photovoltaic output fluctuation monitoring data and heat pump unit load full temperature field monitoring data during the operation of the integrated energy station;

[0008] The fluctuation analysis module is used to analyze the power output fluctuation monitoring data caused by changes in cloud cover under different meteorological environmental conditions, and to obtain the degree of instability of the photovoltaic power output of the integrated energy station.

[0009] The coupling analysis module is used to analyze the energy supply and demand synchronization of the photovoltaic power output fluctuation monitoring data and the heat pump unit load full temperature field monitoring data to obtain the supply and demand matching overlap, and to analyze the power change consistency of the photovoltaic power output fluctuation monitoring data and the heat pump unit load full temperature field monitoring data to obtain the power response coupling correlation.

[0010] The strategy generation module is used to determine the energy efficiency optimization potential of the integrated energy station based on the degree of instability of the photovoltaic output, the degree of overlap of supply and demand matching, and the coupling correlation of the power response, and to determine the collaborative control optimization strategy of the integrated energy station based on the energy efficiency optimization potential.

[0011] Optionally, acquiring photovoltaic output fluctuation monitoring data and heat pump unit load full temperature field monitoring data during the operation of the integrated energy station includes:

[0012] Real-time power output fluctuation monitoring data of photovoltaics during operation is obtained by power sensors pre-installed in the photovoltaic array combiner box and inverter;

[0013] By pre-deploying a temperature sensing network in different temperature zones of the evaporator and condenser of the heat pump unit, real-time monitoring data of the entire temperature field of the unit load is obtained.

[0014] Optionally, the analysis of the photovoltaic power output fluctuation monitoring data under different meteorological environmental conditions, focusing on the sudden changes in power output caused by cloud cover variations, yields the degree of instability of the photovoltaic power output of the integrated energy station, including:

[0015] Based on the meteorological environment level, the photovoltaic power output fluctuation monitoring data is divided into multiple power output monitoring sub-data.

[0016] By analyzing the power mutations corresponding to the inherent mode function components in the multiple power output monitoring sub-data, the degree of instability of the photovoltaic power output of the integrated energy station is obtained.

[0017] Optionally, the analysis of power abrupt changes corresponding to the inherent mode function components in the plurality of power output monitoring sub-data to obtain the degree of instability of the photovoltaic power output of the integrated energy station includes:

[0018] For the multiple output monitoring sub-data, extract the first instability degree corresponding to the high-frequency intrinsic mode function component;

[0019] For the multiple output monitoring sub-data, extract the second instability degree corresponding to the low-frequency intrinsic mode function component;

[0020] By combining the first degree of instability and the second degree of instability, the degree of instability of the photovoltaic output of the integrated energy station is obtained.

[0021] Optionally, extracting the first instability degree corresponding to the high-frequency intrinsic mode function component from the plurality of output monitoring sub-data includes:

[0022] Based on the multiple power output monitoring sub-data, the number of high-frequency intrinsic mode function components and the entropy corresponding to the high-frequency intrinsic mode function components in each of the power output monitoring sub-data are obtained;

[0023] Based on the entropy corresponding to the high-frequency intrinsic mode function components in each of the power output monitoring sub-data, determine the trend of entropy variation with the meteorological environment level;

[0024] Based on the trend of entropy change with the meteorological environment level and the number of high-frequency intrinsic mode function components, the first degree of instability corresponding to the high-frequency intrinsic mode function components is determined.

[0025] Optionally, extracting the second instability degree corresponding to the low-frequency intrinsic mode function component from the plurality of output monitoring sub-data includes:

[0026] Based on the plurality of power output monitoring sub-data, the number of low-frequency intrinsic mode function components and the energy corresponding to the low-frequency intrinsic mode function components in each of the power output monitoring sub-data are obtained;

[0027] Based on the energy corresponding to the low-frequency intrinsic mode function component in each of the power output monitoring sub-data, determine the trend of energy variation with the meteorological environment level;

[0028] Based on the changing trends of the entropy with the meteorological environment level, the changing trends of the energy with the meteorological environment level, and the number of low-frequency intrinsic mode function components, the second degree of instability corresponding to the low-frequency intrinsic mode function components is determined.

[0029] Optionally, the analysis of the energy supply and demand synchronicity of the photovoltaic output fluctuation monitoring data and the heat pump unit load full temperature field monitoring data to obtain the supply and demand matching overlap includes:

[0030] According to the preset time series step size, the photovoltaic power output fluctuation monitoring data is divided into multiple first time units, and the heat pump unit load full temperature field monitoring data is divided into multiple second time units.

[0031] For each of the first time units, the power abrupt change point of the photovoltaic output is determined by the kurtosis peak value;

[0032] For each of the second time units, differential monitoring is used to determine the response point corresponding to the maximum rate of change of heat load in the full temperature field monitoring data;

[0033] The energy supply and demand synchronicity between the power mutation point and the response point is analyzed to obtain the supply and demand matching overlap degree.

[0034] Optionally, the analysis of the consistency of power change between the photovoltaic power output fluctuation monitoring data and the heat pump unit load full temperature field monitoring data to obtain the power response coupling correlation includes:

[0035] The first output amplitude of the photovoltaic power output fluctuation monitoring data for each first time unit at different meteorological environmental levels is obtained, and the second energy consumption amplitude of the heat pump unit load full temperature field monitoring data for each second time unit at different meteorological environmental levels is obtained.

[0036] Consistency analysis is performed on the changes in the first output amplitude and the second energy consumption amplitude to obtain the power response coupling correlation.

[0037] Optionally, determining the collaborative control optimization strategy for the integrated energy station based on the energy efficiency optimization potential includes:

[0038] The energy efficiency optimization potential is compared with a preset energy efficiency optimization threshold;

[0039] When the energy efficiency optimization potential is greater than or equal to the preset energy efficiency optimization threshold, the energy storage system's charge and discharge regulation and control are triggered, and the multi-energy complementary collaborative operation plan is initiated.

[0040] Optionally, the triggering of the energy storage system's charge and discharge regulation and control and the activation of the multi-energy complementary coordinated operation plan includes:

[0041] When the photovoltaic output exceeds the load demand of the heat pump unit and reaches the preset charging threshold, the energy storage system is controlled to perform charging operation and convert the excess electrical energy into heat energy and store it in the heat storage device.

[0042] When the photovoltaic output is less than the load demand of the heat pump unit and reaches the preset discharge threshold, the energy storage system is controlled to discharge, and the auxiliary heating equipment is coordinated to maintain the stable inlet water temperature of the heat pump unit, so as to realize the full absorption of photovoltaic power generation and the low energy consumption operation of the heat pump unit.

[0043] In the intelligent management system of an integrated energy station proposed in this invention, a full-process intelligent management mechanism integrating data perception, fluctuation analysis, coupling analysis, and strategy generation is constructed to achieve refined modeling and multi-dimensional quantitative evaluation of the fluctuation of photovoltaic output and the dynamic characteristics of heat pump load. By analyzing the abrupt changes in cloud cover characteristics of photovoltaic output under different meteorological levels, the instability risks in system operation are accurately identified. Combined with the dual analysis of energy supply and demand matching overlap and power response coupling correlation, the dynamic adaptation relationship between source and load is comprehensively characterized, thereby scientifically evaluating the energy efficiency optimization potential of the system and generating an adaptive collaborative control strategy accordingly. This method effectively improves the local consumption rate of photovoltaic power generation and the operating efficiency of the energy storage system, enhances the stability and response capability of the system under complex meteorological conditions, realizes deep synergy and global energy efficiency optimization of photovoltaic, energy storage, and heat pump multi-energy units, and significantly reduces the operating cost and carbon emission level of the integrated energy system. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the structure of the first embodiment of the intelligent management system for the integrated energy station of the present invention;

[0045] Figure 2 This is a flowchart illustrating the specific steps involved in obtaining the supply-demand matching overlap in the intelligent management system of the integrated energy station of this invention.

[0046] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0048] In one embodiment, such as Figure 1 As shown, a smart management system for an integrated energy station is provided, the system comprising:

[0049] Data sensing module 10 is used to acquire photovoltaic output fluctuation monitoring data and heat pump unit load full temperature field monitoring data during the operation of the integrated energy station;

[0050] The intelligent management system of the integrated energy station can be a multi-energy complementary system management architecture integrating data sensing, fluctuation analysis, coupling analysis, and strategy generation functions. It can be used to achieve coordinated operation and global energy efficiency optimization of photovoltaic, energy storage, and heat pump multi-energy units. The data sensing module can be a sensing and data aggregation unit used to collect dynamic data on photovoltaic output and heat pump load. In this embodiment, the data sensing module can collect power and temperature field data in real time through a sensor network deployed on the photovoltaic array and heat pump unit. Furthermore, the data sensing module can output photovoltaic output fluctuation monitoring data and heat pump unit load full temperature field monitoring data to the fluctuation analysis module and the coupling analysis module. The photovoltaic output fluctuation monitoring data can be measured sequence data reflecting the change of photovoltaic power generation over time. For example, the photovoltaic output fluctuation monitoring data can be recorded at a fixed sampling frequency using a photovoltaic inverter or a dedicated power sensor. The photovoltaic output fluctuation monitoring data can be used to characterize the output characteristics of the photovoltaic system affected by meteorological conditions, serving as input for fluctuation analysis and coupling analysis. The heat pump unit load full temperature field monitoring data can be a multi-point temperature distribution data set covering the heat pump evaporator, condenser, and ambient temperature. In one exemplary embodiment, full-temperature field monitoring data of the heat pump unit load can be synchronously collected at key nodes of the heat pump system through a distributed temperature sensor network. This full-temperature field monitoring data can be used to characterize the spatiotemporal dynamics of the heat pump's heat load demand, supporting load response modeling.

[0051] Acquiring monitoring data on photovoltaic power output fluctuations and full-temperature field monitoring data of heat pump unit load during the operation of an integrated energy station can be achieved by synchronously collecting photovoltaic power sequences and multi-point temperature field data of the heat pump through a sensor network. Furthermore, this operation can be implemented using a timestamp-aligned distributed data acquisition protocol, with local preprocessing via edge computing nodes before uploading, thereby establishing a high-dimensional dynamic data foundation at both the source and load ends.

[0052] The fluctuation analysis module 20 is used to analyze the power output fluctuation monitoring data caused by changes in cloud cover under different meteorological environmental conditions, and to obtain the degree of instability of photovoltaic power output of the integrated energy station.

[0053] The fluctuation analysis module can be a unit for extracting abrupt change features and quantifying risks under meteorological constraints on photovoltaic power output fluctuation data. In a specific embodiment, the fluctuation analysis module can identify power output abrupt change patterns caused by cloud cover changes under the meteorological environment level operating condition label. The fluctuation analysis module can be used to output the degree of instability of photovoltaic power output, avoiding misjudging gradual changes in natural weather as abnormal disturbances. The fluctuation analysis module can receive photovoltaic power output fluctuation monitoring data from the data sensing module and output the degree of instability to the strategy generation module. The meteorological environment level operating condition can be an operating environment classification standard based on meteorological parameters such as cloud cover, irradiance, and temperature. For example, the meteorological environment level operating condition can include, but is not limited to, clear and stable operating condition, cloudy and fluctuating operating condition, and sudden severe operating condition. The meteorological environment level operating condition can be used to provide contextual conditions for fluctuation analysis, distinguishing between normal gradual changes and abnormal abrupt changes. Cloud cover changes can be a physical phenomenon in which the movement of clouds in the atmosphere causes a rapid decay or recovery of solar irradiance. Cloud cover changes can be used as a major external disturbance source causing voltage and power abrupt changes. A sudden power output surge can be a phenomenon where photovoltaic (PV) output experiences a significant, step-like drop or rebound within a short period. Power output surges can be used as a key indicator to characterize system instability risk. The degree of PV output instability can be used as an assessment indicator to quantify the severity and unpredictability of PV power fluctuations. In an exemplary embodiment, the degree of PV output instability can be calculated based on the statistical characteristics of the amplitude, frequency, and duration of power surges under specific weather conditions. The degree of PV output instability can be used to identify potential instability risks during system operation.

[0054] For photovoltaic (PV) power output fluctuation monitoring data, under different meteorological environmental conditions, the sudden changes in power output caused by cloud cover variations are analyzed to determine the degree of instability of PV power output at the integrated energy station. This can be achieved by extracting power fluctuation characteristics and quantifying instability under meteorological level labels. Furthermore, this operation can be implemented by detecting abrupt change points based on wavelet transform and combining it with meteorological level threshold filtering, reconstructing normal fluctuation patterns using an LSTM-AE model, and calculating residual instability. This allows for accurate identification of abnormal fluctuations caused by rapid cloud movement and elimination of interference from gradual changes in normal weather.

[0055] The coupling analysis module 30 is used to analyze the energy supply and demand synchronization of photovoltaic power output fluctuation monitoring data and heat pump unit load full temperature field monitoring data to obtain the supply and demand matching overlap, and to analyze the power change consistency of photovoltaic power output fluctuation monitoring data and heat pump unit load full temperature field monitoring data to obtain the power response coupling correlation.

[0056] The coupling analysis module can be an analysis unit that analyzes the dynamic adaptation relationship between photovoltaics and heat pumps from two dimensions: energy and power. In a specific embodiment, the coupling analysis module can calculate the supply-demand matching overlap degree corresponding to energy supply-demand synchronicity, and the power response coupling correlation corresponding to power change consistency. The coupling analysis module can be used to achieve multi-dimensional decoupling and quantitative characterization of the source-load dynamic relationship. The coupling analysis module can receive two types of monitoring data from the data sensing module and output two coupling indicators to the strategy generation module. Energy supply-demand synchronicity can be the overlap matching characteristic of photovoltaic power supply and actual heat pump energy demand in the time dimension. Energy supply-demand synchronicity can be used as the basic dimension for calculating the supply-demand matching overlap degree. The supply-demand matching overlap degree can be a quantitative indicator that measures the degree of overlap between the photovoltaic power supply curve and the heat pump load curve in terms of energy time series. For example, the supply-demand matching overlap degree can be calculated by the ratio of the intersection and union of the integral areas of the two curves within a sliding window. The supply-demand matching overlap degree can be used to reflect the adaptation level of the source-load at the energy level. Power variation consistency can be seen as the synergy between the trend of photovoltaic power output changes and the direction of heat pump power response on a short-term time scale. Power variation consistency can be used as a fundamental dimension for calculating power response coupling correlation. Power response coupling correlation can be an indicator characterizing the strength of the dynamic response correlation between photovoltaic power fluctuations and heat pump power adjustments. In an exemplary embodiment, power response coupling correlation can be calculated using methods such as cross-correlation coefficients, Granger causality tests, or dynamic time warping distances. Power response coupling correlation can be used to characterize the interaction capability between the source and load at the power regulation level.

[0057] Analyzing the energy supply and demand synchronicity of photovoltaic (PV) output fluctuation monitoring data and heat pump unit load full-temperature field monitoring data yields the supply-demand matching overlap degree. This can be achieved by calculating the overlap degree between the PV power supply curve and the heat pump equivalent heat load curve on the time axis. Furthermore, this operation can be implemented by using a sliding window Jaccard index to calculate the energy overlap ratio and by dynamically programming to align the two curves and calculate the overlap area ratio, thus quantifying the source-load matching level at the energy level. Analyzing the consistency of power change between PV output fluctuation monitoring data and heat pump unit load full-temperature field monitoring data yields the power response coupling correlation. This can be achieved by analyzing the correlation between the PV power change rate and the heat pump power response direction on a short-time scale. Furthermore, this operation can be implemented by calculating the delayed Pearson correlation coefficient and using transfer entropy to assess the information flow direction and coupling strength, thus revealing the dynamic interaction capability of the source-load at the power regulation level.

[0058] The strategy generation module 40 is used to determine the energy efficiency optimization potential of the integrated energy station based on the degree of instability of photovoltaic output, the degree of overlap of supply and demand matching and the coupling correlation of power response, and to determine the collaborative control optimization strategy of the integrated energy station based on the energy efficiency optimization potential.

[0059] The strategy generation module can be a decision-making unit that integrates multi-dimensional evaluation indicators to generate adaptive collaborative control commands. In a specific embodiment, the strategy generation module can construct an energy efficiency optimization potential model based on the degree of instability, the degree of supply-demand matching overlap, and the coupling correlation of power response, and map it to specific control actions. The strategy generation module can be used to output a collaborative control optimization strategy to drive the charging and discharging of energy storage and the start-stop regulation of heat pumps. The strategy generation module can receive the outputs of the fluctuation judgment module and the coupling analysis module to generate the final control strategy. The energy efficiency optimization potential can be a spatial evaluation value that can further improve energy utilization efficiency based on the current system state. For example, the energy efficiency optimization potential can be comprehensively evaluated using a multi-dimensional weighted approach that integrates the degree of instability, the degree of supply-demand matching overlap, and the coupling correlation of power response. The energy efficiency optimization potential can be used as a decision-making basis for generating collaborative control strategies. The collaborative control optimization strategy can be a set of joint operation scheduling commands formulated for photovoltaic, energy storage, and heat pumps. The collaborative control optimization strategy can be used to achieve deep collaboration of multi-energy units and global energy efficiency optimization.

[0060] Based on the instability of photovoltaic output, the overlap of supply and demand matching, and the coupling correlation of power response, the energy efficiency optimization potential of integrated energy stations can be determined. This can be achieved by integrating three types of indicators to construct a multi-dimensional evaluation function, outputting the current system's optimizable space. Furthermore, this operation can be implemented by using a fuzzy comprehensive evaluation method with weighted synthesis and constructing a lightweight neural network regression model to map the indicators to potential values, thereby scientifically determining whether the system has regulatory value and optimization direction. Determining the collaborative control optimization strategy of integrated energy stations based on energy efficiency optimization potential can be achieved by mapping the energy efficiency optimization potential to specific energy storage charging and discharging commands and heat pump operation level adjustment schemes. Further, this operation can be achieved by using a lookup table method to match a preset strategy library and online solving of a mixed-integer linear programming model to generate the optimal scheduling sequence, thereby generating adaptive, closed-loop multi-energy collaborative control commands.

[0061] Taking the response of an integrated energy station in an industrial park to severe afternoon convective weather as an example, the intelligent management system of the integrated energy station in this embodiment can be described as follows: In a summer afternoon, when an integrated energy station in a certain park encounters a rapidly moving cumulonimbus cloud, the data sensing module simultaneously collects data showing that the photovoltaic output drops by 60% within 30 seconds, while the heat pump's heating load decreases due to the sudden drop in outdoor temperature. The fluctuation judgment module identifies this sudden change as a high-instability event under cloudy and fluctuating conditions; the coupling analysis module finds that the supply and demand matching overlap is low, but the power response coupling correlation is high, indicating that the heat pump can quickly reduce its power to match the photovoltaic drop. Based on this, the strategy generation module determines that there is significant potential for energy efficiency optimization and immediately issues instructions: the energy storage system temporarily suspends charging, and the heat pump switches to a low-frequency operation mode, thereby avoiding frequent energy storage switching and improving the local photovoltaic absorption rate, maintaining system stability.

[0062] In one embodiment, acquiring photovoltaic output fluctuation monitoring data and heat pump unit load full temperature field monitoring data during the operation of the integrated energy station includes:

[0063] Real-time power output fluctuation monitoring data of photovoltaics during operation is obtained by power sensors pre-installed in the photovoltaic array combiner box and inverter;

[0064] By pre-deploying a temperature sensing network in different temperature zones of the evaporator and condenser of the heat pump unit, real-time monitoring data of the entire temperature field of the unit load is obtained.

[0065] The photovoltaic array combiner box can be an electrical connection device that collects the current from multiple photovoltaic strings and outputs it to the inverter. It serves as one of the installation locations for a power sensor to acquire string-level or combiner-level photovoltaic output data. In this embodiment, the photovoltaic array combiner box and the inverter together constitute a key electrical node in the photovoltaic system, providing a physical deployment basis for the power sensor. For example, the photovoltaic array combiner box can adopt a standard DC combiner unit structure. The inverter can be a power electronic device that converts photovoltaic DC power to AC power and can be used as one of the installation locations for a power sensor to acquire system-level real-time photovoltaic output data. Furthermore, the inverter can form an electrical coupling interface with the power sensor to achieve high-sampling-rate power signal extraction. The power sensor can be a sensing device used to measure the real-time power value in the electrical circuit. It can be used to directly capture instantaneous fluctuations in photovoltaic output caused by meteorological disturbances such as cloud cover, avoiding estimation errors and lag. In an exemplary embodiment, the power sensor can calculate the instantaneous power by synchronously sampling voltage and current based on the Hall effect or sampling resistor principle. In one specific embodiment, the power sensor may include, but is not limited to, a high-bandwidth digital power sensor, a power frequency analog power transmitter, and a power sampling module built into a smart meter. The real-time output fluctuation monitoring data of photovoltaics can be the instantaneous photovoltaic power sequence directly measured by the power sensor at the combiner box or inverter, which can be used as a high-fidelity input for the fluctuation judgment module to truly reflect the characteristics of sudden output changes under meteorological disturbances.

[0066] By pre-installing power sensors in the photovoltaic array combiner box and inverter, real-time power output fluctuation monitoring data of the photovoltaic system can be obtained during operation. This can be achieved by deploying power sensors at key electrical nodes of the photovoltaic system to collect instantaneous power values ​​in real time. Furthermore, this operation can be realized by deploying string-level power sensors at the combiner box outlet to identify local shading and deploying high-sampling-rate power sensors on the AC side of the inverter to capture system-level abrupt changes. This allows for direct acquisition of the actual power output fluctuations affected by cloud shading, avoiding errors and delays caused by irradiance inversion or model extrapolation.

[0067] The evaporator of a heat pump unit can be a heat exchange component that absorbs ambient heat in a heat pump system. It can be used as a temperature sensing network deployment area to reflect the heat exchange status on the low-temperature side. The condenser of a heat pump unit can be a heat exchange component that releases heat to the heat-consuming end in a heat pump system. It can be used as a temperature sensing network deployment area to reflect the heat release status on the high-temperature side. Different temperature zones can be local areas with significant temperature differences formed on the surfaces or in the flow channels of the evaporator and condenser due to uneven heat exchange. They can be used to form the spatial basis for full-temperature field sensing, supporting fine-grained characterization of load dynamics. The temperature sensing network can be a multi-point temperature acquisition system distributed on key heat exchange components of the heat pump. It can be used to break through the traditional coarse-grained monitoring that only relies on inlet and outlet water temperatures, and achieve full-temperature field sensing of the heat pump load status. In a specific embodiment, the temperature sensing network can connect multiple temperature sensors through wired or wireless means to achieve spatial coverage and synchronous temporal acquisition. For example, the temperature sensing network can include, but is not limited to, a gridded uniform distribution network, a key node focused coverage network, and an adaptive density adjustable network. Real-time monitoring data of the entire temperature field of the unit load can be a collection of multiple temperatures synchronously collected by a temperature sensing network in different temperature zones of the evaporator and condenser. This data can be used as a fine-grained input for the coupled analysis module to accurately characterize the spatiotemporal distribution characteristics of the heat pump's heat load.

[0068] By pre-deploying a temperature sensor network across different temperature zones of the evaporator and condenser of the heat pump unit, real-time monitoring data of the entire temperature field of the unit load can be obtained. This can be achieved by placing multiple temperature sensors on the surface of the heat pump heat exchanger or at key locations in the flow channels to simultaneously collect temperature distribution data. Furthermore, this operation can be realized by deploying a gridded sensor array at the intersection of the evaporator fins and pipes, and by setting gradient temperature monitoring points at the inlet, outlet, and middle of the condenser. This allows for perception of the heat load status from a local to a global perspective, overcoming the limitations of traditional single-point temperature monitoring.

[0069] For example, in a scenario where a commercial complex's energy station is dealing with sudden changes in morning crevasse radiation, the intelligent management system of this embodiment can handle situations where sunlight shines through rapidly moving fragmented cumulus clouds, creating intense crevasse radiation that causes a 40% surge in photovoltaic output within 10 seconds. The high-bandwidth power sensor deployed at the inverter immediately captures this sudden change, while traditional methods based on radiometers fail to respond promptly due to the integral smoothing effect. Simultaneously, the heat pump condenser, due to the building's heating demand not yet being fully activated, shows a lag in temperature rise at the outlet temperature zone as indicated by its temperature sensing network. Based on this, the system determines there is a risk of short-term energy oversupply, and the strategy generation module immediately instructs the energy storage to switch to charging mode and temporarily suspends heat pump frequency reduction, effectively improving local photovoltaic consumption and preventing curtailment.

[0070] In one embodiment, for photovoltaic power output fluctuation monitoring data, under different meteorological environmental conditions, the sudden changes in power output caused by changes in cloud cover are analyzed to obtain the degree of instability of photovoltaic power output of the integrated energy station, including:

[0071] Based on the meteorological environment level, the photovoltaic power output fluctuation monitoring data is divided into multiple power output monitoring sub-data.

[0072] By analyzing the power abrupt changes corresponding to the inherent mode function components in multiple power output monitoring sub-data, the degree of instability of photovoltaic power output of the integrated energy station is obtained.

[0073] The power output monitoring sub-data can be a set of subsequences obtained by dividing the original photovoltaic power output fluctuation monitoring data according to meteorological environmental level operating conditions. This can be used to focus subsequent fluctuation analysis on power behavior under specific meteorological disturbance conditions, improving the accuracy of abrupt change identification. In this embodiment, the power output monitoring sub-data can be segmented into segments corresponding to the operating conditions based on real-time or forecast meteorological environmental level labels. For example, the power output monitoring sub-data can be dynamically sliced ​​based on the dominant meteorological level within a sliding time window, or the data can be segmented at meteorological level switching points using an event-triggered mechanism. The intrinsic mode function components can be physically meaningful eigenmodes extracted from non-stationary signals using empirical mode decomposition (EMD) methods. These can be used to separate high-frequency transient components reflecting abrupt cloud obstruction characteristics, filtering out slow-changing trends and noise interference. In an exemplary embodiment, the intrinsic mode function components can be obtained by decomposing the power output monitoring sub-data using adaptive time-frequency analysis methods such as Empirical Mode Decomposition (EMD), Ensemble Empirical Mode Decomposition (EEMD), or Complementary Ensemble Empirical Mode Decomposition (CEEMDAN). Furthermore, the intrinsic mode function components may include, but are not limited to, high-frequency transient components, mid-frequency fluctuation components, and low-frequency trend components.

[0074] Based on meteorological environmental levels, the photovoltaic (PV) output fluctuation monitoring data is divided into multiple output monitoring sub-data. This can be achieved by using meteorological environmental level operating conditions as classification labels to segment and aggregate continuous PV output time series. Furthermore, this operation can be implemented by dynamically slicing data based on the dominant meteorological level within a sliding time window and using an event-triggered mechanism to segment data at meteorological level switching points. This allows for strong correlation alignment between fluctuation data and external meteorological disturbance conditions, avoiding feature confusion caused by mixed operating conditions. The power mutations corresponding to the intrinsic mode function components in the multiple output monitoring sub-data are analyzed to obtain the degree of instability of the integrated energy station's PV output. This can be achieved by analyzing the intrinsic mode function components obtained after mode decomposition of each sub-data, identifying power mutation characteristics caused by cloud cover, and quantifying their instability. In one specific embodiment, this operation can be achieved by selecting high-frequency IMF components to calculate their energy entropy and mutation point density as instability indicators, jointly constructing mutation feature vectors from multiple IMF components, and inputting them into a lightweight classifier to output the instability level. This allows for the accurate extraction of instantaneous fluctuation components driven by meteorological factors, the elimination of equipment noise and natural gradual interference, and the improvement of the physical interpretability and accuracy of instability assessment.

[0075] Taking the operation of a park energy station under cloudy and rapidly moving weather conditions in spring as an example, the intelligent management system of the integrated energy station in this embodiment can be as follows: On a spring afternoon, the meteorological monitoring system determines that the current operating condition is cloudy and fluctuating. The photovoltaic output data collected by the data sensing module is divided into output monitoring sub-data under this operating condition. The fluctuation judgment module performs EEMD decomposition on the sub-data to obtain several intrinsic mode function components, among which the first two high-frequency components exhibit short-term large oscillations. The system identifies the step drop in these components as a power mutation caused by cloud cover, and calculates the degree of high instability based on its frequency and amplitude. This result is transmitted to the strategy generation module, triggering the energy storage system to temporarily suspend discharge and the heat pump to preheat and store energy in advance, thereby mitigating potential subsequent energy supply gaps and avoiding inappropriate control actions due to misjudgment as equipment failure.

[0076] In one embodiment, the power abrupt changes corresponding to the inherent mode function components in multiple power output monitoring sub-data are analyzed to obtain the degree of instability of the photovoltaic power output of the integrated energy station, including:

[0077] For multiple output monitoring sub-data, extract the first instability degree corresponding to the high-frequency intrinsic mode function component;

[0078] For multiple output monitoring sub-data, extract the second instability degree corresponding to the low-frequency intrinsic mode function component;

[0079] By combining the first and second instabilities, the instability of the photovoltaic output of the integrated energy station can be obtained.

[0080] The high-frequency intrinsic mode function (IMF) components can be high-frequency intrinsic modes that characterize the short-term, severe fluctuations in photovoltaic power output after empirical mode decomposition (EMD). These IMF components can be used to characterize instantaneous power disturbances caused by rapid cloud shading or sudden changes in local shading. In this embodiment, the IMF components can be obtained by performing EMD-type decomposition on the power output monitoring sub-data and filtering out the top few IMF components that satisfy high-frequency oscillation characteristics. Furthermore, the IMF components can include, but are not limited to, cloud gap flicker components, edge shading abrupt change components, and equipment start-up / shutdown interference components. The first instability level can be a short-term sudden disturbance risk indicator quantified based on the IMF components, which can be used to reflect the likelihood of the system facing power abrupt change events requiring rapid response. In an exemplary embodiment, the first instability level can be evaluated by calculating features such as the energy entropy, abrupt change point density, or peak change rate of the high-frequency IMF components. For example, extracting the first instability level corresponding to the high-frequency IMF components from multiple power output monitoring sub-data can be achieved by feature extraction and quantification evaluation of the high-frequency IMF components after mode decomposition of each power output monitoring sub-data. Furthermore, this operation can be achieved by calculating the L1 norm mutation intensity of the high-frequency IMF component and normalizing it into an instability score, and by counting the number of steps exceeding the threshold per unit time as the first degree of instability, thereby accurately identifying the risk of sudden power disturbances caused by rapid cloud movement.

[0081] Low-frequency intrinsic mode function (IMF) components can be low-frequency intrinsic modes that reflect the long-term trend and slow changes in photovoltaic (PV) output after empirical mode decomposition. They can be used to characterize persistent power shifts caused by changes in solar altitude angle, slow movement of large-scale cloud systems, etc. In a specific embodiment, the IMF components can be obtained by selecting several low-frequency components from the IMF sequence, typically corresponding to diurnal irradiance variations or gradual meteorological processes. For example, the IMF components can include, but are not limited to, diurnal trend components, multi-hourly slowly varying components, and weather system evolution components. The second instability level can be a risk indicator of long-term deviation from the steady-state trend obtained based on the quantification of the IMF components. It can be used to reflect the risk level of persistent insufficient or excessive power supply capacity of the system over a long time scale. Furthermore, the second instability level can be obtained by analyzing the trend slope, cumulative deviation from the benchmark value, or stationarity test results of the low-frequency IMF components. Extracting the second instability level corresponding to the low-frequency IMF components from multiple power output monitoring sub-data can be achieved by performing trend analysis and steady-state deviation assessment on the low-frequency IMF components. In one exemplary embodiment, this operation can be achieved by using the relative deviation of the low-frequency component mean within a sliding window from the historical benchmark as a second degree of instability and mapping the trend instability score to the p-value of the ADF stationarity test, thereby identifying slow but persistent shifts in power supply capacity and avoiding misjudging them as emergencies.

[0082] Combining the first and second instability levels to obtain the overall instability level of the photovoltaic output of the energy station can be achieved by fusing the two types of instability indicators according to their physical meaning or control requirements. Furthermore, this operation can be achieved by constructing a rule engine that uses a weighted linear combination (e.g., high-frequency weight 0.7, low-frequency weight 0.3) to output a four-quadrant instability level based on the high-low combination of the two, thereby generating a comprehensive instability assessment result that combines time scale differentiation and disturbance type identification capabilities.

[0083] For example, in a scenario where a commercial complex's energy station operates under mixed cloud cover in an autumn afternoon, the intelligent management system of this embodiment could be: A commercial complex's energy station experiences alternating cumulus and stratus cloud cover in an autumn afternoon. The fluctuation analysis module divides the photovoltaic output data into output monitoring sub-data according to the current multi-cloud fluctuation condition and performs CEEMDAN decomposition. The high-frequency IMF component shows multiple short-term sharp drops, calculating the first instability level as high; the low-frequency IMF component shows a slow downward trend, with the second instability level as medium. Based on this, the strategy generation module determines that: energy storage should be immediately called in for short-term replenishment to cope with the sudden drop, while simultaneously arranging heat pumps to store heat in advance during periods of low electricity prices to cope with subsequent continuous power shortages. This dual-dimensional assessment avoids misjudging a slow decline as an emergency event and over-dispatching energy storage, and also prevents ignoring sudden drops that could lead to power outages, achieving precise coordination.

[0084] In one embodiment, for multiple output monitoring sub-data, the first instability level corresponding to the high-frequency intrinsic mode function component is extracted, including:

[0085] Based on multiple power output monitoring sub-data, the number of high-frequency intrinsic mode function components and the entropy corresponding to the high-frequency intrinsic mode function components in each power output monitoring sub-data are obtained;

[0086] The number of high-frequency intrinsic mode function (IMF) components can be the number of high-frequency IMF components contained in a single power output monitoring sub-data after mode decomposition, and can be used to reflect the frequency density of abrupt events in photovoltaic power output under a specific meteorological level. In this embodiment, the number of high-frequency IMF components can be statistically determined based on a preset frequency threshold or IMF order rule to count the number of IMFs that satisfy high-frequency characteristics. For example, the number of high-frequency IMF components can be used in conjunction with the entropy corresponding to the high-frequency IMF components in each power output monitoring sub-data to characterize the spatiotemporal characteristics of abrupt events. The entropy corresponding to the high-frequency IMF components in each power output monitoring sub-data can be an information entropy index used to quantify the signal complexity and uncertainty of high-frequency IMF components, and can be used to characterize the severity and randomness of power abrupt changes under the meteorological conditions. Furthermore, the entropy corresponding to the high-frequency IMF components in each power output monitoring sub-data can be used to calculate the sample entropy, approximate entropy, or Shannon entropy for the high-frequency IMF components in each power output monitoring sub-data. In one exemplary embodiment, the entropy corresponding to the high-frequency intrinsic mode function component in each output monitoring sub-data can together with the number of high-frequency intrinsic mode function components to form a two-dimensional input for instability assessment.

[0087] Based on multiple power output monitoring sub-data sets, the number of high-frequency intrinsic mode function (IMF) components and the entropy corresponding to each IMF component in each sub-data set can be obtained. This can be achieved by performing mode decomposition on each sub-data set, identifying high-frequency IMF components and counting their numbers, while simultaneously calculating their information entropy. Furthermore, this operation can be implemented by using CEEMDAN decomposition, taking the first 2-3 IMFs as high-frequency components and counting them, and calculating the sample entropy of their joint signal; or by calculating the entropy of each high-frequency IMF separately and then taking the average. This allows for the simultaneous acquisition of quantitative features in two dimensions: mutation frequency and mutation intensity.

[0088] Based on the entropy corresponding to the high-frequency intrinsic mode function components in each output monitoring sub-data, determine the trend of entropy variation with meteorological environment level;

[0089] The entropy variation trend with meteorological environment level can be seen as a systematic evolution law of high-frequency IMF entropy values ​​under different meteorological environment level conditions, which can be used to reveal the mapping relationship between the intensity of external meteorological disturbances and the complexity of photovoltaic fluctuations. In a specific embodiment, the entropy variation trend with meteorological environment level can be fitted to a monotonic or nonlinear trend curve according to the level order of the average entropy value corresponding to each meteorological level. For example, the entropy variation trend with meteorological environment level can include, but is not limited to, a monotonically increasing trend, a nonlinear peak trend, and a plateau saturation trend. Based on the entropy corresponding to the high-frequency intrinsic mode function components in each output monitoring sub-data, the entropy variation trend with meteorological environment level can be determined. This can be achieved by classifying entropy values ​​according to meteorological environment level and analyzing their variation patterns in the level sequence. Furthermore, this operation can be achieved by using piecewise linear regression to fit the entropy-level relationship and using nonparametric kernel smoothing to estimate the trend curve, thereby establishing a correlation model between meteorological disturbance level and fluctuation complexity.

[0090] Based on the trend of entropy change with meteorological environment level and the number of high-frequency intrinsic mode function components, the first degree of instability corresponding to the high-frequency intrinsic mode function components is determined.

[0091] Based on the trend of entropy variation with meteorological environmental levels and the number of high-frequency intrinsic mode function (IMF) components, the first degree of instability corresponding to the high-frequency IMF components can be determined. This can be achieved by fusing the slope of the entropy trend and the number of high-frequency IMFs to construct an assessment model for the first degree of instability. Furthermore, this operation can be implemented by weighting the trend slope and the number of components to obtain the instability score; or by constructing a two-dimensional lookup table to map the trend type and quantity range to the instability level, thereby enabling adaptive quantification of meteorological scenarios with sudden disturbance risks.

[0092] For example, in the scenario of an industrial park energy station under a severe convective weather warning in the summer afternoon, the intelligent management system of the integrated energy station in this embodiment can be based on the weather system forecast that a sudden and severe operating condition is about to occur. Historical data shows that at this level, the high-frequency IMF entropy value increases by 40% compared to the previous level of cloudy and fluctuating conditions, and the number of high-frequency IMFs increases from an average of 1.8 to 3.2. The current real-time output monitoring sub-data, after decomposition, yields three high-frequency IMFs, whose joint sample entropy is 1.35, consistent with the trend of a sharp increase in entropy. Based on this, the system determines that the first level of instability is high-risk, and the strategy generation module immediately initiates preventive control: energy storage switches to a fast response mode, and the heat pump stores cold in advance to cope with the expected photovoltaic drop. This decision avoids the lag problem of waiting for the actual power drop before responding, and effectively suppresses the bus voltage fluctuation.

[0093] In one embodiment, for multiple output monitoring sub-data, a second instability level corresponding to the low-frequency intrinsic mode function component is extracted, including:

[0094] Based on multiple power output monitoring sub-data, the number of low-frequency natural mode function components and the energy corresponding to the low-frequency natural mode function components in each power output monitoring sub-data are obtained;

[0095] Based on the energy corresponding to the low-frequency intrinsic mode function components in each power output monitoring sub-data, determine the trend of energy variation with meteorological environment level;

[0096] Based on the trends of entropy and energy with meteorological environment level, and the number of low-frequency intrinsic mode function components, the second degree of instability corresponding to the low-frequency intrinsic mode function components is determined.

[0097] The number of low-frequency intrinsic mode function (IMF) components can be the number of low-frequency IMF components included in a single power output monitoring sub-data after mode decomposition. This number can reflect the structural complexity or persistence of slow fluctuations in photovoltaic power output under specific weather conditions. In this embodiment, the number of low-frequency IMF components can be calculated based on a preset frequency threshold or IMF order rules, counting the number of IMFs at the last few orders that satisfy low-frequency characteristics. For example, the number of low-frequency IMF components can include, but is not limited to, taking the last 2-3 orders of IMFs as low-frequency components and counting them. The energy corresponding to the low-frequency IMF components in each power output monitoring sub-data can be the square integral of the low-frequency IMF component signal in the time domain, characterizing the comprehensive intensity of its power deviation from the steady state. This can be used to quantify the coupling effect of long-term power shift amplitude and duration caused by diurnal irradiance variations or continuous cloudy weather. Furthermore, the energy corresponding to the low-frequency IMF components in each power output monitoring sub-data can be calculated by treating the energy of the low-frequency IMF component in each power output monitoring sub-data. This energy is a cumulative measure of the square of the signal amplitude over time.

[0098] Based on multiple power output monitoring sub-data, the number of low-frequency intrinsic mode function (IMF) components and the corresponding energy of each IMF component in each power output monitoring sub-data can be obtained. This can be achieved by performing mode decomposition on each power output monitoring sub-data, identifying low-frequency IMF components and counting their numbers, while simultaneously calculating their energy. In an exemplary embodiment, this operation can be implemented by taking the last 2-3 order IMFs as low-frequency components and counting them, then calculating the total energy of their joint signal; or by calculating the energy of each order low-frequency IMF separately and then summing the results. This allows for the simultaneous acquisition of the structural complexity and energy intensity characteristics of the slowly varying disturbance.

[0099] The trend of energy variation with meteorological environmental level can be seen as a systematic evolution law of low-frequency intrinsic mode function (IMF) energy under different meteorological environmental level conditions, and can be used to reveal the mapping relationship between meteorological gradual disturbances and long-term photovoltaic power supply capacity shifts. In a specific embodiment, the trend of energy variation with meteorological environmental level can be obtained by fitting the average low-frequency energy corresponding to each meteorological level into a trend curve in order of level. For example, the trend of energy variation with meteorological environmental level can include, but is not limited to, a monotonically decreasing trend, a plateau stable trend, and a step-like decreasing trend. Based on the energy corresponding to the low-frequency intrinsic mode function components in each power output monitoring sub-data, the trend of energy variation with meteorological environmental level can be determined. This can be achieved by classifying low-frequency energy values ​​according to meteorological environmental level and analyzing their variation patterns in the level sequence. Furthermore, this operation can be achieved by using moving average smoothing to fit a linear trend, or by using piecewise constant regression to identify energy jump points, thereby establishing a correlation model between the intensity of meteorological gradual changes and long-term power shift energy. Based on the trends of entropy and energy with varying meteorological environmental levels, and the number of low-frequency intrinsic mode function (IMF) components, the degree of second instability corresponding to the low-frequency IMF components is determined. This can be achieved by integrating the high-frequency entropy trend, the low-frequency energy trend, and the number of low-frequency IMF components to construct a second instability assessment model. In one specific embodiment, this operation can be implemented by standardizing the three indicators and inputting them into a lightweight multilayer perceptron to output an instability score; or by designing a rule engine, such as determining high second instability when the energy trend drops sharply and the number is ≥2. This allows for multidimensional coupling quantification of slowly changing unstable states, establishing the correlation logic between high-frequency abrupt changes and low-frequency drift.

[0100] For example, in a scenario where a hospital's integrated energy station operates under continuous cloudy weather conditions during winter, the intelligent management system of the integrated energy station in this embodiment could be as follows: The meteorological system determines the operating condition to be stable under cloudy conditions for three consecutive days. Historical data shows that the low-frequency IMF energy is 60% lower than on sunny days under this condition, and the number of low-frequency IMFs remains stable at 2. After decomposing the current output monitoring sub-data, two low-frequency IMFs are obtained, and their total energy is 35% of the historical sunny day baseline, which is consistent with the energy decline trend. At the same time, the previous high-frequency entropy trend shows that there is no sudden disturbance at present. Based on this, the system determines that the second instability level is medium to high, but the first instability level is low. The strategy generation module adjusts accordingly: the energy storage baseline charging power is reduced by 30%, and the heat pump is switched to the nighttime off-peak electricity heat storage mode to compensate for insufficient daytime photovoltaic power. This decision avoids insufficient daytime heating due to ignoring the continuous low frequency energy and also prevents over-discharge due to misjudging it as an emergency.

[0101] In one embodiment, the energy supply and demand synchronization of photovoltaic power output fluctuation monitoring data and heat pump unit load full temperature field monitoring data is analyzed to obtain the supply and demand matching overlap degree, including:

[0102] According to the preset time series step size, the photovoltaic power output fluctuation monitoring data is divided into multiple first time units, and the heat pump unit load full temperature field monitoring data is divided into multiple second time units.

[0103] For each first time unit, the power abrupt change point of photovoltaic output is determined by the kurtosis peak value;

[0104] For each second time unit, differential monitoring is used to determine the response point corresponding to the maximum rate of change of heat load in the full temperature field monitoring data;

[0105] Analyze the energy supply and demand synchronicity at power mutation points and response points to obtain the degree of supply and demand matching overlap.

[0106] The preset time series step size can be a time granularity benchmark parameter used to divide the photovoltaic and heat pump monitoring data, ensuring that source-load data analysis is performed at a unified time scale that adapts to dynamic characteristics. In this embodiment, the preset time series step size can include, but is not limited to, second-level step sizes, minute-level step sizes, and sliding overlapping window step sizes. The first time unit can be a continuous time period divided from the photovoltaic power output fluctuation monitoring data based on the preset time series step size, which can be used as the basic analysis interval for identifying photovoltaic power abrupt change points. The second time unit can be a continuous time period divided from the heat pump unit load full temperature field monitoring data based on the preset time series step size, which can be used as the basic analysis interval for locating heat load response points. According to the preset time series step size, the photovoltaic power output fluctuation monitoring data is divided into multiple first time units, which can be achieved by segmenting the photovoltaic power sequence at fixed or adaptive time intervals. Furthermore, this operation can be implemented by using non-overlapping equal-length segments or overlapping sliding window segments, thereby providing a structured analysis window for abrupt change point detection.

[0107] Dividing the full temperature field monitoring data of the heat pump unit load into multiple second time units can be achieved by synchronously segmenting the heat pump temperature field data according to the same time series step size as the photovoltaic data. For example, this operation can be implemented through event-triggered dynamic segmentation or static segmentation based on a fixed sampling period, thereby ensuring temporal alignment of the source-load analysis. Kurtosis peaks can be extreme values ​​of fourth-order statistics characterizing the sharpness of the tail of a time series distribution, and can be used to detect instantaneous, drastic fluctuations in photovoltaic output that deviate from a Gaussian distribution. In an exemplary embodiment, kurtosis peaks can include, but are not limited to, positive kurtosis anomaly peaks, negative kurtosis flat valleys, and mixed skewness abrupt change points. The power abrupt change point of photovoltaic output can be the moment within the first time unit when a significant non-stationary change in photovoltaic power occurs, identified by the kurtosis peak. Furthermore, the power abrupt change point of photovoltaic output can be obtained by calculating the kurtosis value of the photovoltaic power sequence within a sliding window and identifying local maxima exceeding a threshold, which can be used to accurately locate the actual power supply interruption or recovery time caused by disturbances such as rapid cloud cover. For each first time unit, the power abrupt change point of photovoltaic output is determined by the kurtosis peak value. This can be achieved by calculating the kurtosis of the power sequence within each first time unit and identifying the time corresponding to significant peaks. In a specific embodiment, this operation can be achieved by using a sliding window to calculate the local kurtosis in real time and setting a dynamic threshold, or by combining wavelet packet decomposition with kurtosis calculation in sub-bands to improve noise resistance. This allows for high-sensitivity capture of non-stationary power abrupt changes, which is superior to traditional mean or variance methods.

[0108] Differential monitoring can be an analytical method for calculating the rate of change of values ​​between adjacent sampling points in the full-temperature field monitoring data of a heat pump. It can be used to capture the dynamic response intensity and temporal characteristics of heat load changes with temperature field variations. The response point corresponding to the maximum rate of change of heat load can be the moment of most drastic change in heat pump heat load identified by differential monitoring within the second time unit. Furthermore, the response point corresponding to the maximum rate of change of heat load can be obtained by performing first- or second-order differencing on the full-temperature field data and taking the time point corresponding to the maximum absolute value. This can be used to reflect the actual starting point of the heat pump's response to sudden changes in ambient temperature or adjustments in user behavior. For each second time unit, determining the response point corresponding to the maximum rate of change of heat load in the full-temperature field monitoring data through differential monitoring can be achieved by performing differential operations on the full-temperature field data of the heat pump and extracting the extreme points of the rate of change. For example, this operation can be achieved by using first-order forward differencing to calculate the instantaneous rate of change, or by using central differencing to improve accuracy and combining it with smoothing filtering to suppress noise, thereby accurately locating the critical moment when the heat pump actually responds to external disturbances.

[0109] Energy supply and demand synchronicity can be measured by the degree of alignment between photovoltaic power abrupt change points and thermal load response points in the time dimension, as well as their energy matching relationship. This can serve as a direct basis for calculating the degree of overlap in supply and demand matching, reflecting the transient coordination capability of the source-load system. Analyzing the energy supply and demand synchronicity between power abrupt change points and response points to obtain the degree of overlap in supply and demand matching can be achieved by aligning the photovoltaic power abrupt change points and thermal load response points in time and assessing the degree of overlap between them in terms of energy supply and demand. Furthermore, this operation can be achieved by calculating the time difference between the two points and mapping it to a synchronization score, or by constructing a joint event window to statistically analyze the effective overlap ratio. This allows for a refined quantification of the transient matching level of the source-load system, avoiding distortion from average power matching.

[0110] Taking the integrated energy system of a commercial building in response to afternoon cumulus cloud passage as an example, the intelligent management system of the integrated energy station in this embodiment can be described as follows: A commercial complex encounters a rapidly moving cumulus cloud at 2 PM, causing a sudden 50% drop in photovoltaic output within 20 seconds. The system divides the data into 30-second preset time series steps. Within the first time unit, kurtosis peaks identify 14:02:15 as the power mutation point. Simultaneously, differential monitoring of the heat pump's full temperature field data reveals 14:02:28 as the response point with the largest heat load change rate. The coupling analysis module calculates a mere 13-second offset between the two points, determining a high degree of synchronization between energy supply and demand, with a high degree of overlap in supply and demand matching. Based on this, the strategy generation module maintains the current operating state of the heat pump, and the energy storage system does not initiate compensation, thereby avoiding ineffective charging and discharging and achieving efficient local consumption during photovoltaic fluctuations.

[0111] In one embodiment, the consistency of power change between photovoltaic power output fluctuation monitoring data and heat pump unit load full-temperature field monitoring data is analyzed to obtain the power response coupling correlation, including:

[0112] Acquire the first output amplitude of each first time unit corresponding to photovoltaic power output fluctuation monitoring data under different meteorological environmental levels, and the second energy consumption amplitude of each second time unit corresponding to heat pump unit load full temperature field monitoring data under different meteorological environmental levels;

[0113] Consistency analysis was performed on the changes in the first output amplitude and the second energy consumption amplitude to obtain the power response coupling correlation.

[0114] The different meteorological environment levels can be classification conditions of operating environments based on meteorological parameters such as cloud cover, irradiance, and temperature. These can be used as contextual constraints for power response coupling analysis to ensure that the amplitude extraction matches the actual disturbance scenario. In this embodiment, different meteorological environment levels may include, but are not limited to, clear and stable high irradiance level, cloudy and fluctuating level, and cloudy and rainy low energy level. The first output amplitude can be a representative power output intensity index of photovoltaic power output fluctuation monitoring data within a first time unit under a specific meteorological environment level. It can be used to characterize the actual energy supply capacity level of photovoltaic under the current meteorological disturbance. For example, the first output amplitude can be obtained by taking the power mean, peak value, or effective value within the time unit and normalizing or weighting it in combination with the meteorological level. The second energy consumption amplitude can be an equivalent electrical power consumption intensity index mapped from the full temperature field monitoring data of the heat pump unit load within a second time unit under the same meteorological environment level. It can be used to reflect the actual energy consumption intensity of the heat pump under the current environmental and load conditions. In one exemplary embodiment, the second energy consumption amplitude can be derived by converting the heat load into equivalent input electrical power based on a heat pump COP model and aggregating it over a time unit.

[0115] Obtaining the first output amplitude of photovoltaic power output fluctuation monitoring data for each first time unit under different meteorological environmental levels can be achieved by filtering photovoltaic data according to meteorological environmental level labels and extracting representative power amplitudes within the corresponding first time unit. Further, this operation can be implemented by using the root mean square value of power within the time unit as the amplitude, or by using the median after removing outliers as the robust amplitude, thereby establishing a characteristic sequence of photovoltaic power supply intensity under meteorological constraints. Obtaining the second energy consumption amplitude of heat pump unit load full temperature field monitoring data for each second time unit under different meteorological environmental levels can be achieved by converting the heat pump full temperature field data into equivalent electrical power under the same meteorological level and aggregating it within the second time unit to obtain the energy consumption amplitude. In a specific embodiment, this operation can be achieved by calculating the instantaneous COP in real time based on a thermodynamic model and integrating to obtain the energy consumption amplitude, or by using a historical operating condition lookup table method to quickly estimate the equivalent power, thereby constructing a characteristic sequence of heat pump energy consumption intensity aligned with photovoltaic power supply. Consistency analysis of the changes in the first output amplitude and the second energy consumption amplitude yields the power response coupling correlation, which can be used to dynamically measure the changing trends of the two amplitude sequences under the same meteorological level. Furthermore, this operation can be achieved by calculating the Pearson correlation coefficient with hysteresis or by using dynamic time warping distance to measure the sequence morphological similarity, thereby quantifying the degree of power-level synergy between photovoltaic power supply fluctuations and heat pump energy consumption response.

[0116] For example, in the scenario of optimizing the operation of an energy station in a university campus under variable weather conditions in spring, the intelligent management system of the integrated energy station in this embodiment could be as follows: A university campus experiences continuous weather changes from sunny to cloudy and then to overcast in the afternoon during spring. The system divides the data into three time periods according to the meteorological environment level: sunny and stable, cloudy and fluctuating, and overcast and rainy with low energy consumption. Under the cloudy and fluctuating level, the first output amplitude of photovoltaic power in the first time unit drops sharply from 80kW to 45kW; simultaneously, due to the decrease in outdoor temperature, the second energy consumption amplitude of the heat pump drops from 30kW to 22kW. Consistency analysis shows that the two are significantly negatively correlated (due to the reduction in heating demand), and the power response coupling correlation is strong. Based on this, the strategy generation module determines that there is no need to activate energy storage compensation, but rather to reduce the operating frequency of the heat pump to match the low energy supply state, thereby reducing power consumption and improving the local photovoltaic consumption rate.

[0117] In one embodiment, a collaborative control optimization strategy for the integrated energy station is determined based on its energy efficiency optimization potential, including:

[0118] The potential for energy efficiency optimization is compared with the preset energy efficiency optimization threshold;

[0119] When the potential for energy efficiency optimization is greater than or equal to the preset energy efficiency optimization threshold, the charging and discharging regulation and control of the energy storage system is triggered and the multi-energy complementary collaborative operation plan is initiated.

[0120] The preset energy efficiency optimization threshold can be a critical value for determining the energy efficiency optimization potential to initiate coordinated control actions. It can serve as a decision boundary for triggering energy storage regulation and coordinated control plans, avoiding low-yield regulation. In this embodiment, the preset energy efficiency optimization threshold can be categorized according to control objectives, including local consumption guidance thresholds, energy storage lifetime protection thresholds, and carbon emission intensity constraint thresholds. Comparing the energy efficiency optimization potential with the preset energy efficiency optimization threshold can be achieved by comparing the real-time calculated energy efficiency optimization potential value with the system's preset threshold. Furthermore, this comparison can be implemented using a static comparison with a fixed threshold or by dynamically adjusting the threshold based on meteorological levels, thereby determining whether the current system state is valuable for implementing coordinated control.

[0121] Energy storage system charge / discharge regulation and control can be a set of operations for dynamically scheduling and limiting the charging and discharging power and status of energy storage units. It can be used to smooth power fluctuations caused by sudden changes in photovoltaic output, support heat pump load response, and improve local consumption rate. For example, energy storage system charge / discharge regulation and control can be triggered when energy efficiency optimization potential is met, and the charge / discharge strategy can be adjusted based on previous coupling analysis results to match source-load dynamics. Multi-energy complementary collaborative operation plan can be a pre-defined set of mode switching and parameter configuration schemes for the joint operation of photovoltaic, energy storage, and heat pumps. It can be used to achieve deep coupling operation of multi-energy units in scenarios with high optimization potential. In an exemplary embodiment, the multi-energy complementary collaborative operation plan can be divided according to operating modes, including peak shaving and valley filling collaborative mode, combined heat and power optimization mode, and emergency power support mode. When the energy efficiency optimization potential is greater than or equal to a preset energy efficiency optimization threshold, the energy storage system charge / discharge regulation and control is triggered, and the multi-energy complementary collaborative operation plan is activated. Alternatively, when the comparison results meet the conditions, the energy storage control command and multi-energy collaborative operation mode can be activated. In one specific embodiment, the operation can be implemented by directly calling a preset control script through an event-driven mechanism, sending an advanced scheduling request to the energy management system for parsing and executing a specific instruction sequence, thereby ensuring that the control action has a clear energy efficiency gain and reducing equipment damage and energy waste caused by ineffective operations.

[0122] For example, in a scenario where the integrated energy station operates under cloudy weather conditions in spring, the intelligent management system of the integrated energy station in this embodiment could be as follows: On a certain afternoon, the clouds over the park move rapidly, causing frequent fluctuations in photovoltaic output. After fluctuation analysis and coupling analysis, the system calculates the energy efficiency optimization potential to be 0.72, exceeding the preset energy efficiency optimization threshold of 0.65. The strategy generation module then triggers the charging and discharging regulation and control of the energy storage system, instructing the energy storage unit to discharge to make up for the shortfall when photovoltaic power drops sharply and to temporarily suspend charging when irradiance recovers; at the same time, it activates the peak shaving and valley filling mode in the multi-energy complementary collaborative operation plan, guiding the heat pump to prioritize heating and energy storage during periods of high photovoltaic power generation. This process avoids frequent start-up and shutdown of energy storage when potential is insufficient, intervening only during high-value windows, significantly improving the local photovoltaic absorption rate and reducing power fluctuations in the grid.

[0123] In one embodiment, triggering the charging and discharging regulation and control of the energy storage system and initiating a multi-energy complementary collaborative operation plan includes:

[0124] When the photovoltaic output exceeds the load demand of the heat pump unit and reaches the preset charging threshold, the energy storage system is controlled to perform charging operation and convert the excess electrical energy into heat energy and store it in the heat storage device.

[0125] When the photovoltaic output is less than the load demand of the heat pump unit and reaches the preset discharge threshold, the energy storage system is controlled to discharge, and the auxiliary heating equipment is coordinated to maintain the stable inlet water temperature of the heat pump unit, so as to realize the full absorption of photovoltaic power generation and the low energy consumption operation of the heat pump unit.

[0126] In this context, photovoltaic (PV) output can be the electrical power output of the PV power generation system at the current moment, which can be used as the main renewable energy input of the system and participate in supply and demand balance judgment. Heat pump unit load demand can be the electrical power consumed by the heat pump to meet heating or cooling targets, which can be used to characterize the main electrical load of the system and compared with PV output to determine energy surplus or deficit status. The preset charging threshold can be the critical value of PV output surplus that triggers the charging action of the energy storage system, which can be used to avoid frequent charging starts when there is only a small surplus, improving the economic efficiency of regulation. In an exemplary embodiment, the preset charging threshold may include, but is not limited to, a basic absorption threshold, a thermal energy storage linkage threshold, and a grid interaction suppression threshold. The energy storage system can be an electrochemical or physical energy storage device for storing and releasing electrical energy, which can be used to smooth PV output fluctuations, supplement power gaps, and support stable system operation. Furthermore, the energy storage system can receive excess electrical energy during the charging phase and work with auxiliary heating equipment to maintain the heat pump inlet water temperature during the discharging phase. The charging operation can be the control behavior of inputting electrical energy into the energy storage system for storage, which can be used to absorb excess PV power and reduce curtailment. Excess electrical energy can be surplus electrical energy generated when photovoltaic output exceeds the load demand of the heat pump unit and meets the charging threshold conditions. It can be used as an energy source for thermal energy storage and conversion, expanding local consumption pathways. Thermal energy can be a form of energy that can be used for heating, converted from electrical energy through an electro-thermal conversion device. It can be stored in a thermal energy storage device as sensible or latent heat, enabling inter-period thermal energy dispatch. The thermal energy storage device can be a medium container or phase change material system for storing thermal energy. It can be used to receive thermal energy converted from excess electrical energy, alleviating the limitations of electrical energy storage capacity and improving overall energy utilization efficiency. For example, the thermal energy storage device can receive thermal energy converted from excess electrical energy and release it in subsequent periods to support the heat pump or provide direct heating.

[0127] When the photovoltaic output exceeds the load demand of the heat pump unit and reaches a preset charging threshold, the energy storage system is controlled to charge, and the excess electrical energy is converted into heat energy and stored in the thermal storage device. This can be achieved by simultaneously initiating electric energy storage charging and electro-thermal conversion storage when the photovoltaic output is detected to be consistently higher than the heat pump load and exceeding the charging threshold. Furthermore, this operation can be achieved by prioritizing charging to the upper limit of the State of Charge (SOC) before activating the electric boiler's thermal storage, and proportionally allocating excess electrical energy to the electric and thermal storage channels. This expands the pathways for renewable energy consumption, alleviates the bottleneck of battery energy storage capacity, and reduces curtailment of solar power.

[0128] A preset discharge threshold can be a critical value for the photovoltaic output shortfall that triggers the energy storage system's discharge action. It can be used to prevent premature discharge under slight insufficient power, thus extending the energy storage lifespan. In one specific embodiment, the preset discharge threshold may include, but is not limited to, a heat pump COP maintenance threshold, a user comfort guarantee threshold, and a system frequency support threshold. Discharge operation can be a control behavior that outputs electrical energy from the energy storage system to supplement the power shortage. It can be used to compensate for insufficient photovoltaic output and maintain the continuity of system power supply. Auxiliary heating equipment can be an electric or thermally driven device used to supplement the heat source to maintain the heat pump inlet water temperature. It can be used to coordinate with energy storage discharge when photovoltaic power is insufficient, preventing the heat pump from experiencing efficiency degradation due to low-temperature inlet water. For example, auxiliary heating equipment may include, but is not limited to, electric auxiliary heating elements, gas combustion supplementation devices, and industrial waste heat recovery interfaces. The heat pump unit inlet water temperature can be the temperature of the heat transfer medium entering the heat pump evaporator or condenser. It can directly affect the heat pump COP and is a key control parameter for low-energy operation. Full grid integration of photovoltaic (PV) power generation means using or storing all the generated energy locally, without feeding it back to the grid or curtailing it. This can maximize the utilization of renewable energy and reduce carbon emissions. Low-energy operation of heat pump units can achieve a high COP (Coefficient of Performance) under reasonable inlet water temperature and load matching conditions. This can reduce the overall power consumption of the system and improve comprehensive energy efficiency.

[0129] When the photovoltaic output is less than the load demand of the heat pump unit and reaches the preset discharge threshold, the energy storage system is controlled to discharge, and the auxiliary heating equipment is coordinated to maintain a stable inlet water temperature for the heat pump unit. Specifically, if the photovoltaic output remains below the heat pump load and falls below the discharge threshold, the energy storage system is activated to discharge and replenish power, while simultaneously activating the auxiliary heating equipment to raise the inlet water temperature. Furthermore, this operation can be achieved by dynamically adjusting the auxiliary heating power based on the inlet water temperature deviation and prioritizing the use of heat released from the thermal storage device to replace some of the electric auxiliary heating. This prevents a sharp drop in COP due to low-temperature inlet water, ensuring efficient and low-consumption operation of the heat pump.

[0130] For example, in the scenario of a commercial complex energy station operating under alternating cloudy and sunny weather conditions in winter, the intelligent management system of the integrated energy station in this embodiment could be as follows: One afternoon, the clouds briefly disperse, and the photovoltaic output rapidly increases to 300kW, while the heat pump heating load only requires 180kW, exceeding the preset charging threshold. The system immediately starts energy storage charging and converts 40kW of the remaining 90kW of electrical energy into hot water via an electric boiler and stores it in the heat storage tank. Subsequently, the clouds cover the area again, and the photovoltaic output drops sharply to 100kW, below the heat pump load and reaching the preset discharge threshold. The energy storage system begins discharging to make up for the 80kW shortfall, while the auxiliary electric heater starts, raising the heat pump inlet water temperature from 35°C to 42°C, preventing the COP from dropping from 3.8 to 2.9. Throughout the entire process, the photovoltaic power is fully absorbed locally, and the heat pump operates efficiently without relying on external power grid supplementation.

[0131] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A smart management system for an integrated energy station, characterized in that, The system includes: The data sensing module is used to acquire photovoltaic output fluctuation monitoring data and heat pump unit load full temperature field monitoring data during the operation of the integrated energy station; The fluctuation analysis module is used to analyze the power output fluctuation monitoring data caused by changes in cloud cover under different meteorological environmental conditions, and to obtain the degree of instability of the photovoltaic power output of the integrated energy station. The coupling analysis module is used to analyze the energy supply and demand synchronization of the photovoltaic power output fluctuation monitoring data and the heat pump unit load full temperature field monitoring data to obtain the supply and demand matching overlap, and to analyze the power change consistency of the photovoltaic power output fluctuation monitoring data and the heat pump unit load full temperature field monitoring data to obtain the power response coupling correlation. The strategy generation module is used to determine the energy efficiency optimization potential of the integrated energy station based on the degree of instability of the photovoltaic output, the degree of overlap of supply and demand matching, and the coupling correlation of the power response, and to determine the collaborative control optimization strategy of the integrated energy station based on the energy efficiency optimization potential.

2. The intelligent management system for the integrated energy station as described in claim 1, characterized in that, The acquisition of photovoltaic output fluctuation monitoring data and heat pump unit load full temperature field monitoring data during the operation of the integrated energy station includes: Real-time power output fluctuation monitoring data of photovoltaics during operation is obtained by power sensors pre-installed in the photovoltaic array combiner box and inverter; By pre-deploying a temperature sensing network in different temperature zones of the evaporator and condenser of the heat pump unit, real-time monitoring data of the entire temperature field of the unit load is obtained.

3. The intelligent management system for the integrated energy station as described in claim 1, characterized in that, The method involves analyzing the power output fluctuation monitoring data under different meteorological environmental conditions, focusing on the sudden changes in power output caused by cloud cover variations, to determine the degree of instability in the photovoltaic output of the integrated energy station. This analysis includes: Based on the meteorological environment level, the photovoltaic power output fluctuation monitoring data is divided into multiple power output monitoring sub-data. By analyzing the power mutations corresponding to the inherent mode function components in the multiple power output monitoring sub-data, the degree of instability of the photovoltaic power output of the integrated energy station is obtained.

4. The intelligent management system for the integrated energy station as described in claim 3, characterized in that, The analysis of power abrupt changes corresponding to the inherent mode function components in the multiple power output monitoring sub-data to obtain the instability level of the photovoltaic output of the integrated energy station includes: For the multiple output monitoring sub-data, extract the first instability degree corresponding to the high-frequency intrinsic mode function component; For the multiple output monitoring sub-data, extract the second instability degree corresponding to the low-frequency intrinsic mode function component; By combining the first degree of instability and the second degree of instability, the degree of instability of the photovoltaic output of the integrated energy station is obtained.

5. The intelligent management system for the integrated energy station as described in claim 4, characterized in that, The step of extracting the first instability degree corresponding to the high-frequency intrinsic mode function component from the plurality of power output monitoring sub-data includes: Based on the multiple power output monitoring sub-data, the number of high-frequency intrinsic mode function components and the entropy corresponding to the high-frequency intrinsic mode function components in each of the power output monitoring sub-data are obtained; Based on the entropy corresponding to the high-frequency intrinsic mode function components in each of the power output monitoring sub-data, determine the trend of entropy change with the meteorological environment level; Based on the trend of entropy change with the meteorological environment level and the number of high-frequency intrinsic mode function components, the first degree of instability corresponding to the high-frequency intrinsic mode function components is determined.

6. The intelligent management system for the integrated energy station as described in claim 5, characterized in that, The step of extracting the second instability degree corresponding to the low-frequency intrinsic mode function component from the plurality of power output monitoring sub-data includes: Based on the multiple power output monitoring sub-data, the number of low-frequency intrinsic mode function components and the energy corresponding to the low-frequency intrinsic mode function components in each of the power output monitoring sub-data are obtained; Based on the energy corresponding to the low-frequency intrinsic mode function component in each of the power output monitoring sub-data, determine the trend of energy variation with the meteorological environment level; Based on the changing trends of the entropy with the meteorological environment level, the changing trends of the energy with the meteorological environment level, and the number of low-frequency intrinsic mode function components, the second degree of instability corresponding to the low-frequency intrinsic mode function components is determined.

7. The intelligent management system for the integrated energy station as described in claim 1, characterized in that, The analysis of the energy supply and demand synchronization of the photovoltaic output fluctuation monitoring data and the heat pump unit load full temperature field monitoring data yields the supply and demand matching overlap degree, including: According to the preset time series step size, the photovoltaic power output fluctuation monitoring data is divided into multiple first time units, and the heat pump unit load full temperature field monitoring data is divided into multiple second time units. For each of the first time units, the power abrupt change point of the photovoltaic output is determined by the kurtosis peak value; For each of the second time units, differential monitoring is used to determine the response point corresponding to the maximum rate of change of heat load in the full temperature field monitoring data; The energy supply and demand synchronicity between the power mutation point and the response point is analyzed to obtain the supply and demand matching overlap degree.

8. The intelligent management system for the integrated energy station as described in claim 7, characterized in that, The analysis of the consistency of power change between the photovoltaic power output fluctuation monitoring data and the heat pump unit load full temperature field monitoring data yields the power response coupling correlation, including: The first output amplitude of the photovoltaic power output fluctuation monitoring data for each first time unit at different meteorological environmental levels is obtained, and the second energy consumption amplitude of the heat pump unit load full temperature field monitoring data for each second time unit at different meteorological environmental levels is obtained. Consistency analysis is performed on the changes in the first output amplitude and the second energy consumption amplitude to obtain the power response coupling correlation.

9. The intelligent management system for the integrated energy station as described in claim 1, characterized in that, The determination of the collaborative control optimization strategy for the integrated energy station based on the energy efficiency optimization potential includes: The energy efficiency optimization potential is compared with a preset energy efficiency optimization threshold; When the energy efficiency optimization potential is greater than or equal to the preset energy efficiency optimization threshold, the energy storage system's charge and discharge regulation and control are triggered, and the multi-energy complementary collaborative operation plan is initiated.

10. The intelligent management system for the integrated energy station as described in claim 9, characterized in that, The triggering of the energy storage system's charge and discharge regulation and control, and the initiation of the multi-energy complementary collaborative operation plan, includes: When the photovoltaic output exceeds the load demand of the heat pump unit and reaches the preset charging threshold, the energy storage system is controlled to perform charging operation and convert the excess electrical energy into heat energy and store it in the heat storage device. When the photovoltaic output is less than the load demand of the heat pump unit and reaches the preset discharge threshold, the energy storage system is controlled to discharge, and the auxiliary heating equipment is coordinated to maintain the stable inlet water temperature of the heat pump unit, so as to realize the full absorption of photovoltaic power generation and the low energy consumption operation of the heat pump unit.

Images