Energy storage coupled solar high temperature heat pump steam supply digital twin system and method
By constructing a digital twin system for data prediction and control optimization of solar high-temperature heat pump steam supply, the problems of system control lag and energy mismatch were solved, and the stable and efficient operation of the solar heat pump system was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XIAN THERMAL POWER RES INST CO LTD
- Filing Date
- 2026-03-23
- Publication Date
- 2026-06-30
AI Technical Summary
Existing solar high-temperature heat pump steam supply systems lack numerical modeling and control methods, making it difficult for the system to maintain optimal operation in the long term. Furthermore, the intraday fluctuations and seasonality of solar energy resources make it difficult to meet the continuous, stable, and adjustable steam load demand.
A solar high-temperature heat pump steam supply digital twin system with energy storage coupling is constructed, including a data acquisition module, a time series prediction module, a digital twin module, a system control module, and a feedback verification module. Data prediction and control optimization are performed through machine learning and digital twin models to achieve proactive and advance control.
It enables real-time observation and interactive control of solar heat pump systems, solves the problems of control lag and energy-saving optimization, is suitable for industrial steam production, and achieves stability and high efficiency in steam supply.
Smart Images
Figure CN122305672A_ABST
Abstract
Description
Technical Field
[0001] The embodiments disclosed herein belong to the field of solar high-temperature heat pump technology, specifically relating to a digital twin system and method for energy storage coupled with solar high-temperature heat pump steam supply. Background Technology
[0002] A solar-powered high-temperature heat pump is a composite energy supply system that integrates solar thermal collection technology and high-temperature heat pump technology. It utilizes solar energy as a clean heating energy source to produce steam, and can also produce high-temperature steam through a heat pump system.
[0003] Solar energy resources offer advantages such as cleanliness and low marginal cost, but the intense intraday fluctuations in irradiance and the mismatch with seasonality make it difficult to directly meet the "continuous, stable, and adjustable" steam load requirements. Therefore, energy storage (sensible heat / latent heat / thermochemical or water tank / molten salt / phase change materials, etc.) and adjustable heat pump systems must be introduced to time-shift energy and improve its quality. Furthermore, the "solar energy-energy storage-high-temperature heat pump-steam generation / heat exchange-steam load" is a complex system with strong coupling, strong nonlinearity, and strong time variation, making it difficult to maintain optimal system operation in the long term using only fixed thresholds or empirical switching.
[0004] However, existing research focuses on the construction of novel systems and lacks numerical modeling and control methods for solar high-temperature heat pump steam supply systems coupled with energy storage. Summary of the Invention
[0005] The embodiments disclosed herein aim to at least solve one of the technical problems existing in the prior art, and provide a digital twin system and method for energy storage coupled with solar high-temperature heat pump steam supply.
[0006] One aspect of this disclosure provides a solar high-temperature heat pump steam supply digital twin system coupled with energy storage, the system comprising a data acquisition module, a time series prediction module, a digital twin module, a system control module, and a feedback verification module; The data acquisition module is used to collect and preprocess historical and real-time multi-source data, and output them to the time series prediction module and the digital twin module respectively. The time-series prediction module is used to receive historical multi-source data, predict multi-source data at multiple future times based on machine learning, and output the prediction data to the digital twin module. The digital twin module is used to receive real-time multi-source data and future multi-source data, calculate and output predicted terminal parameters based on the digital twin model to the system control module; The system control module is used to receive the predicted terminal parameters, generate a heat pump advance control scheme based on the preset optimization target, and execute it. The feedback verification module is used to collect the actual terminal parameters after the control scheme is executed, compare them with the predicted terminal parameters, and correct the digital twin model based on the comparison differences.
[0007] Furthermore, the multi-source data includes meteorological information, sunshine duration, solar irradiance, cloud information characteristics, ambient temperature and humidity, inlet and outlet temperatures and flow rates of the heat collector circuit, temperature, pressure and power on the heat pump side, state of charge of the heat storage unit, steam pressure and temperature on the steam side, and steam consumption.
[0008] Furthermore, the time-series prediction module is specifically used for: Calculate historical heat consumption based on historical steam consumption, steam pressure, and temperature; Using historical meteorological information, sunshine duration, solar irradiance, cloud information characteristics, and historical heat consumption as input variables, a long short-term memory network model is used to predict solar irradiance and heat consumption at multiple future moments.
[0009] Furthermore, the digital twin model includes physical models of a solar thermal collector circuit, an energy storage unit, a high-temperature heat pump system, and a steam generation and heat exchange unit connected by mass and energy conservation principles.
[0010] Furthermore, the digital twin module is configured as follows: Historical multi-source data is written into the digital twin model for initial calibration; Real-time multi-source data is written into the initialized and calibrated digital twin model for online parameter identification and offset correction.
[0011] Furthermore, the optimization objective is: Minimize heat pump energy consumption and compressor start-up / shutdown penalties while satisfying the steam parameters entering the storage tank.
[0012] Furthermore, the feedback verification module is also used for A credibility assessment is introduced based on the parameter residuals generated by the comparison; when the credibility assessment fails, an alarm is triggered and the correction is stopped.
[0013] Another aspect of this disclosure provides a digital twin method for energy storage-coupled solar high-temperature heat pump steam supply, based on the energy storage-coupled solar high-temperature heat pump steam supply digital twin system described above, the method comprising: Collect and preprocess historical and real-time multi-source data; Based on historical multi-source data, predict multi-source data for multiple future moments using machine learning; Based on real-time multi-source data and future multi-source data, predictive terminal parameters are calculated and output using a digital twin model. Based on the predicted terminal parameters, a heat pump advance control scheme is generated and executed based on a preset optimization target; The actual terminal parameters after the control scheme is executed are collected and compared with the predicted terminal parameters. The digital twin model is then corrected based on the differences in the comparison.
[0014] Furthermore, the method of predicting multi-source data at multiple future moments based on historical multi-source data and machine learning includes: Calculate historical heat consumption based on historical steam consumption, steam pressure, and temperature; Using historical meteorological information, sunshine duration, solar irradiance, cloud information characteristics, and historical heat consumption as input variables, a long short-term memory network model is used to predict solar irradiance and heat consumption at multiple future moments.
[0015] Furthermore, after correcting the digital twin model based on the comparative differences, the method further includes: A credibility assessment is introduced based on the parameter residuals generated by the comparison; when the credibility assessment fails, an alarm is triggered and the correction is stopped.
[0016] This disclosure discloses a digital twin system and method for energy storage coupled solar high-temperature heat pump steam supply. By constructing a complete digital twin system for energy storage coupled solar heat pump steam system, it predicts the required operating scenarios in the future based on historical and real-time multi-source data, and realizes closed-loop optimization iteration based on real data to complete autonomous learning optimization. It is applicable to the solar high-temperature heat pump control system for industrial steam production, realizes proactive advance control, and completes data acquisition, prediction, interaction and control. It can solve problems such as control lag and energy-saving optimization of solar heat pump system, and can be observed and interactively controlled in real time. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the structure of a solar high-temperature heat pump steam supply digital twin system coupled with energy storage according to an embodiment of the present disclosure. Figure 2 This is a schematic flowchart of a digital twin method for supplying steam to a solar high-temperature heat pump coupled with energy storage, according to another embodiment of this disclosure. Detailed Implementation
[0018] The technical solutions of the embodiments of this disclosure 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 this disclosure, and not all of them. Based on the embodiments of this disclosure, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this disclosure.
[0019] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.
[0020] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.
[0021] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various components, these components should not be limited by these terms. These terms are used to distinguish one component from another. Therefore, the first component discussed below may be referred to as the second component without departing from the teachings of this disclosure. As used in this disclosure, the term "and / or" includes all combinations of any and more of the associated listed items.
[0022] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of exemplary embodiments, and the modules or processes in the drawings are not necessarily necessary for implementing this disclosure, and therefore cannot be used to limit the scope of protection of this disclosure.
[0023] like Figure 1 As shown, one embodiment of this disclosure provides a solar high-temperature heat pump steam supply digital twin system coupled with energy storage. The system includes a data acquisition module 110, a time-series prediction module 120, a digital twin module 130, a system control module 140, and a feedback verification module 150. The data acquisition module 110 collects and preprocesses historical and real-time multi-source data, and outputs it to the time-series prediction module 120 and the digital twin module 130, respectively. The time-series prediction module 120 receives historical multi-source data and predicts future timeframes based on machine learning. The system receives multi-source data and outputs it to the digital twin module 130. The digital twin module 130 receives real-time multi-source data and future multi-source data, calculates and outputs predicted terminal parameters to the system control module 140 based on the digital twin model. The system control module 140 receives the predicted terminal parameters, generates a heat pump advance control scheme based on a preset optimization target, and executes it. The feedback verification module 150 collects the actual terminal parameters after the control scheme is executed, compares them with the predicted terminal parameters, and corrects the digital twin model based on the comparison differences.
[0024] Specifically, the data acquisition module 110 collects data such as solar irradiance, ambient temperature and humidity, inlet and outlet temperatures and flow rates of the heat collector circuit, temperature, pressure and power of the heat pump side, SOC of the heat storage unit, pressure and temperature of the steam side, and terminal steam consumption from temperature and humidity sensors and flow sensors via the Internet of Things, as well as meteorological data, energy parameters (electricity price, etc.), terminal parameter data, heat pump system operating parameters, and energy storage unit parameters of the solar high-temperature heat pump steam supply system coupled with energy storage. It obtains historical multi-source data and real-time multi-source data, and performs data preprocessing (time alignment, removal of outliers, compensation for missing values, etc.) to form a real-time data stream, which is then output to the time series prediction module 120 and the digital twin module 130.
[0025] The time-series prediction module 120 receives historical multi-source data from the data acquisition module 110. It incorporates built-in machine learning algorithms (LSTM, Transformer, etc.) for time-series prediction to forecast future multi-source data and outputs this forecast to the digital twin module 130. By inputting the historical and real-time multi-source data into the time-series prediction module 120, future multi-source data is predicted, determining the solar irradiance and sunshine duration at multiple future moments, and correspondingly predicting the steam consumption at multiple future moments. Specifically, this includes: Calculate historical heat consumption based on historical steam consumption, steam pressure, and temperature; Using historical meteorological information, sunshine duration, solar irradiance, cloud information characteristics, and historical heat consumption as input variables, a long short-term memory network model is used to predict solar irradiance and heat consumption at multiple future moments.
[0026] Taking the Long Short-Term Memory (LSTM) network model as an example, historical multi-source datasets are used. h t-1 and real-time multi-source datasets x t As the input to the LSTM forget gate, σ ( W f · [ h t-1 , x t ]+ b f ) calculate f t The output of the forget gate is the multi-source dataset that needs to be ignored. Similarly, the historical multi-source dataset... h t-1 and real-time multi-source datasets x t As the input to the LSTM input gate, σ ( W i ·[ h t-1 , x t ]+ b i ) calculate i t As the output of the input gate, and expressed in tanh( W c ·[ h t-1 , x t ]+ b c ) calculate t The cell state output, which serves as the input gate, is the input feature of the LSTM algorithm. The aforementioned... f t and i t by f t × C t-1 + i t × t calculate C t This is used to update cell states, i.e., the carrier of the multi-source dataset. Finally, based on the historical multi-source dataset... h t-1 and real-time multi-source datasets x t As the input to the LSTM output gate, σ ( W o ·[ h t-1 , x t ]+ b o Calculate the output of the output gate. o t Based on cell state C t by o t × tanh( C t ) Calculate the future multi-source dataset h t This enables future predictions based on multi-source data.
[0027] The digital twin module 130 is the core module of this embodiment, mainly including a physical information digital twin model (a physical model of the solar collector circuit, energy storage unit, high-temperature heat pump system, steam generation and heat exchange unit connected by mass and energy conservation) and a specific geometric model (the actual dimensions of each component) constructed based on the actual heat pump system, as well as defined state variables, adjustable parameters and boundary condition matrices. The historical multi-source data is input into the digital twin module to initialize and calibrate the digital twin model, and online identification and offset correction of multiple parameters of the digital twin model are completed based on real-time multi-source data. The digital twin model processes the data received from the data acquisition module 110 and the time series prediction module 120, converting it into real-time and future terminal parameters (including steam temperature and pressure, heat pump performance parameters, energy storage unit SOC, etc.). Specifically, this includes: Real-time multi-source data is input into the digital twin model to calculate the SOC and temperature field of the energy storage unit, key state parameters on the steam side, and operating parameters of the heat pump system. By inputting data such as solar irradiance, sunshine duration, terminal steam consumption, and SOC of energy storage units at multiple future moments into a digital twin model, forward simulations are performed for different heat supply and demand scenarios. The digital twin model calculates terminal parameters at multiple future moments, such as the SOC change trajectory of energy storage units, steam-side parameter curves, and system energy consumption / efficiency indicators, thus completing multi-scenario simulations and future operating condition simulations.
[0028] The terminal parameters for multiple future time periods are output to the system control module 140. The system control module 140 has built-in constraints and optimization algorithms. Under the premise of meeting the constraints, it actively optimizes the operating parameters of the heat pump system to achieve optimal system performance. Specifically, if the SOC of the energy storage unit is insufficient for subsequent periods, then, under the constraint of the steam parameters entering the storage tank, the optimization objective is to minimize the heat pump energy consumption and compressor start-up / stop penalty. This aims to increase the heat pump's heat output, increase steam production to increase the SOC of the energy storage unit, generate a heat pump advance control scheme, and achieve proactive and precise advance control of the heat pump.
[0029] The feedback verification module 150 compares the changes in terminal parameters caused by changes in control variables with the real-time acquired terminal parameters, performing closed-loop rolling optimization of the digital twin model. Specifically, it collects real-time multi-source data of the heat pump system after the control scheme is executed, compares it with the simulation results of future operating conditions, calculates the residuals of each parameter for model verification, and introduces a credibility assessment based on the residuals to update the digital twin model online, achieving self-closed-loop updates. When the credibility is lower than a preset threshold and the credibility assessment fails, an alarm is triggered and correction stops, switching to manual operation.
[0030] This disclosure discloses a digital twin system for solar high-temperature heat pump steam supply coupled with energy storage. By constructing a complete digital twin system for solar heat pump steam systems coupled with energy storage, it predicts the required operating scenarios in the future based on historical and real-time multi-source data, and performs closed-loop optimization iteration based on real data to complete autonomous learning optimization. This system is suitable for solar high-temperature heat pump control systems for industrial steam production, realizing proactive advance control and completing data acquisition, prediction, interaction, and control. It can solve problems such as control lag and energy-saving optimization in solar heat pump systems, and can be observed and interactively controlled in real time.
[0031] like Figure 2 As shown, another embodiment of this disclosure provides a digital twin method for energy storage coupled with solar high-temperature heat pump steam supply, based on the energy storage coupled solar high-temperature heat pump steam supply digital twin system described in the previous embodiment. The method includes: Step S1: Collect and preprocess historical and real-time multi-source data; Step S2: Based on historical multi-source data, predict multi-source data for multiple future moments using machine learning; Step S3: Based on real-time multi-source data and future multi-source data, calculate and output the predicted terminal parameters based on the digital twin model; Step S4: Based on the predicted terminal parameters, generate and execute a heat pump advance control scheme based on a preset optimization target; Step S5: Collect the actual terminal parameters after the control scheme is executed, compare them with the predicted terminal parameters, and correct the digital twin model based on the comparison differences.
[0032] For example, the step of predicting multi-source data for multiple future moments based on historical multi-source data and machine learning includes: Calculate historical heat consumption based on historical steam consumption, steam pressure, and temperature; Using historical meteorological information, sunshine duration, solar irradiance, cloud information characteristics, and historical heat consumption as input variables, a long short-term memory network model is used to predict solar irradiance and heat consumption at multiple future moments.
[0033] For example, after correcting the digital twin model based on the contrast differences, the method further includes: A credibility assessment is introduced based on the parameter residuals generated by the comparison; when the credibility assessment fails, an alarm is triggered and the correction is stopped.
[0034] Specifically, the energy storage coupled solar high-temperature heat pump steam supply digital twin method of this disclosure is based on the energy storage coupled solar high-temperature heat pump steam supply digital twin system described in the above embodiments. The specific implementation process has been described in detail in the above embodiments and will not be repeated here.
[0035] It is understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of this disclosure, and this disclosure is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this disclosure, and these modifications and improvements are also considered to be within the scope of protection of this disclosure.
Claims
1. A solar-powered high-temperature heat pump steam supply digital twin system with energy storage coupling, characterized in that, The system includes a data acquisition module, a time series prediction module, a digital twin module, a system control module, and a feedback verification module. The data acquisition module is used to collect and preprocess historical and real-time multi-source data, and output them to the time series prediction module and the digital twin module respectively. The time-series prediction module is used to receive historical multi-source data, predict multi-source data at multiple future times based on machine learning, and output the prediction data to the digital twin module. The digital twin module is used to receive real-time multi-source data and future multi-source data, calculate and output predicted terminal parameters based on the digital twin model to the system control module; The system control module is used to receive the predicted terminal parameters, generate a heat pump advance control scheme based on the preset optimization target, and execute it. The feedback verification module is used to collect the actual terminal parameters after the control scheme is executed, compare them with the predicted terminal parameters, and correct the digital twin model based on the comparison differences.
2. The energy storage coupled solar high-temperature heat pump steam supply digital twin system according to claim 1, characterized in that, The multi-source data includes meteorological information, sunshine duration, solar irradiance, cloud information characteristics, ambient temperature and humidity, inlet and outlet temperatures and flow rates of the heat collector circuit, temperature, pressure and power of the heat pump side, state of charge of the heat storage unit, steam side pressure and temperature, and steam consumption.
3. The energy storage coupled solar high-temperature heat pump steam supply digital twin system according to claim 2, characterized in that, The time series prediction module is specifically used for: Calculate historical heat consumption based on historical steam consumption, steam pressure, and temperature; Using historical meteorological information, sunshine duration, solar irradiance, cloud information characteristics, and historical heat consumption as input variables, a long short-term memory network model is used to predict solar irradiance and heat consumption at multiple future moments.
4. The energy storage coupled solar high-temperature heat pump steam supply digital twin system according to claim 1, characterized in that, The digital twin model includes physical models of a solar thermal collector circuit, an energy storage unit, a high-temperature heat pump system, and a steam generation and heat exchange unit connected by mass and energy conservation principles.
5. The energy storage coupled solar high-temperature heat pump steam supply digital twin system according to claim 4, characterized in that, The digital twin module is configured as follows: Historical multi-source data is written into the digital twin model for initial calibration; Real-time multi-source data is written into the initialized and calibrated digital twin model for online parameter identification and offset correction.
6. The energy storage coupled solar high-temperature heat pump steam supply digital twin system according to claim 1, characterized in that, The optimization objective is: Minimize heat pump energy consumption and compressor start-up / shutdown penalties while satisfying the steam parameters entering the storage tank.
7. The energy storage coupled solar high-temperature heat pump steam supply digital twin system according to claim 1, characterized in that, The return verification module is also used for A credibility assessment is introduced based on the parameter residuals generated by the comparison; when the credibility assessment fails, an alarm is triggered and the correction is stopped.
8. A digital twin method for energy storage coupled with solar high-temperature heat pump steam supply, based on the energy storage coupled solar high-temperature heat pump steam supply digital twin system according to any one of claims 1 to 7, characterized in that, The method includes: Collect and preprocess historical and real-time multi-source data; Based on historical multi-source data, predict multi-source data for multiple future moments using machine learning; Based on real-time multi-source data and future multi-source data, predictive terminal parameters are calculated and output using a digital twin model. Based on the predicted terminal parameters, a heat pump advance control scheme is generated and executed based on a preset optimization target; The actual terminal parameters after the control scheme is executed are collected and compared with the predicted terminal parameters. The digital twin model is then corrected based on the differences in the comparison.
9. The digital twin method for solar high-temperature heat pump steam supply coupled with energy storage as described in claim 8, characterized in that, The method of predicting future multi-source data based on historical multi-source data and machine learning includes: Calculate historical heat consumption based on historical steam consumption, steam pressure, and temperature; Using historical meteorological information, sunshine duration, solar irradiance, cloud information characteristics, and historical heat consumption as input variables, a long short-term memory network model is used to predict solar irradiance and heat consumption at multiple future moments.
10. The method for digital twin of solar high-temperature heat pump steam supply coupled with energy storage according to claim 8, characterized in that, After correcting the digital twin model based on the comparative differences, the method further includes: A credibility assessment is introduced based on the parameter residuals generated by the comparison; when the credibility assessment fails, an alarm is triggered and the correction is stopped.