Multi-energy complementary cross-seasonal heat storage and supply system and control method thereof

Abstract

The invention provides a multi-energy complementary cross-seasonal heat storage and supply system and a control method thereof. The system comprises an air source heat pump, a ground source heat pump, a buried pipe, an energy station, a first heat supply loop used for communicating the buried pipe with the evaporation side of the ground source heat pump, and a second heat supply loop used for communicating the air source heat pump with the evaporation side of the ground source heat pump. By means of the two-stage heating structure, the air source heat pump or the buried pipe is used for providing a low-level heat source for the ground source heat pump, and the technical problem that high-temperature hot water needed by primary network heat supply is difficult to meet by a traditional heat pump is solved. The control method comprises the steps of monitoring first system energy consumption of the first heat supply loop and second system energy consumption of the second heat supply loop, and switching between the two heat source loops based on an energy consumption comparison result. The invention also relates to an extended system utilizing PVT, industrial waste heat and a steam heat pump. The system is high in energy efficiency, low in investment, sufficient in renewable energy source utilization and suitable for replacing a northern centralized heating primary network.

Description

[0001] A multi-energy complementary cross-seasonal heat storage and supply system and a control method thereof TECHNICAL FIELD

[0002] The present application relates to the technical field of heat supply and energy comprehensive utilization, in particular to a multi-energy complementary cross-seasonal heat storage and supply system and a control method thereof. BACKGROUND

[0004] To achieve deep decarbonization in the field of heat supply, using heat pump technology to efficiently extract renewable energy such as air energy and shallow geothermal energy for heat supply has become an important technical approach to replace fossil energy and reduce carbon emissions. Among the many forms of heat pump applications, the cross-seasonal heat storage and supply system has unique advantages because it can achieve "peak load shifting" in the time scale. Compared with simply relying on heat pumps to extract heat immediately in winter, the cross-seasonal heat storage and supply system has the advantages of efficient comprehensive energy utilization, low-cost long-term energy storage, the ability to provide more stable heat protection, and significantly reducing the carbon emissions of the whole life cycle, which shows that it is of great engineering significance and social value to carry out research on cross-seasonal heat storage technology.

[0005] However, the existing cross-seasonal heat storage and supply technology still has many deficiencies in practical application. The main deficiencies are as follows: first, the combination of solar heat collection devices and ground source heat pumps, although cross-seasonal heat storage is achieved, solar energy is greatly affected by weather and day and night, and the energy flow density is low, resulting in unstable system heat storage and limited heat collection site; second, air source heat pumps are used to assist ground source heat pumps, but the existing design often fails to fully tap the heat storage potential of air source heat pumps in the non-heating season and the flexible heating capacity in different periods of the heating season, resulting in low overall energy efficiency of the system and long equipment investment return period; third, simple electric heat conversion and heat storage, although it uses abandoned wind and light power, but it is highly dependent on large heat storage tanks, and the system is complex and the initial investment is high.

[0006] In summary, the existing technology generally has the defects of single energy utilization form, rigid system operation strategy, and insufficient renewable energy consumption. Therefore, it is urgent to develop a new type of cross-seasonal heat storage and supply system and a control method thereof which can replace traditional high-carbon heat sources, realize multi-energy complementation, have flexible operation mode, and further improve energy efficiency and reduce carbon emissions. SUMMARY

[0007] One of the purposes of the present application is to provide a multi-energy complementary cross-seasonal heat storage and supply system and a control method thereof to solve the problems pointed out in the background art.

[0008] In a first aspect, the present application provides a multi-energy complementary cross-seasonal heat storage and supply system, comprising: an air source heat pump; A ground source heat pump having an evaporation side and a condensation side; A ground pipe; An energy station in heat exchange connection with the condensation side of the ground source heat pump for supplying heat to the heat exchange station; A first heat supply loop for communicating an outlet of the ground pipe with the evaporation side of the ground source heat pump; A second heat supply loop for communicating an outlet of the air source heat pump with the evaporation side of the ground source heat pump; and A control device configured to at least control switching between the first heat supply loop and the second heat supply loop to selectively provide a low-grade heat source for the evaporation side of the ground source heat pump.

[0009] Optionally, the system further comprises a third heat source loop for communicating the outlet of the air source heat pump with the energy station; The control device is further configured to control the third heat source loop to be turned on to make the air source heat pump directly supply heat to the energy station when in a transition season or a period of higher outdoor temperature in a heating season.

[0010] Optionally, the system further comprises a thermal storage loop for communicating the outlet of the air source heat pump with an inlet of the ground pipe; The control device is further configured to control the thermal storage loop to be turned on to make the air source heat pump store heat in the soil surrounding the ground pipe when in a non-heating season.

[0011] Optionally, the control device is configured to: Monitor a first system energy consumption when operating with the first heat supply loop; Monitor a second system energy consumption when operating with the second heat supply loop; Based on a comparison result of the first system energy consumption and the second system energy consumption, perform switching between the first heat supply loop and the second heat supply loop.

[0012] Optionally, the system further comprises: A solar photovoltaic panel; The solar photovoltaic panel is electrically connected with an electrical device of the system for supplying power to the system with solar energy.

[0013] Optionally, the solar photovoltaic panel is a photovoltaic-photothermal integrated PVT assembly; The system further comprises: A water tank; and a first plate heat exchanger; The water tank is in heat exchange connection with the PVT assembly for storing heat collected by the PVT assembly; the water tank is in heat exchange connection with the first heat supply loop or the second heat supply loop via the first plate heat exchanger for preheating low-grade heat source entering the evaporation side of the ground source heat pump in the heating season. Optionally, the system further comprises: an industrial waste heat source; a second plate heat exchanger; The industrial waste heat source is in heat exchange connection with the first heat supply loop or the second heat supply loop via the second plate heat exchanger for preheating low-grade heat source entering the evaporation side of the ground source heat pump.

[0014] Optionally, the control device is further configured to: when the temperature of the industrial waste heat source is greater than a preset high temperature threshold, control the industrial waste heat source to be connected in parallel with the condensation side of the ground source heat pump to supply heat to the energy station together; and control the industrial waste heat return water from the energy station to enter the second plate heat exchanger to preheat the low-grade heat source.

[0015] Optionally, the system further comprises: a steam heat pump, the steam heat pump having an evaporation side and a condensation side; the evaporation side of the steam heat pump is in heat exchange connection with the energy station; the condensation side of the steam heat pump is used to supply steam to steam users.

[0016] Optionally, the control device is configured to implement a hierarchical control strategy, which includes a supervisory controller and a local controller; The supervisory controller is configured to: optimize the cross-seasonal heat balance of the ground heat exchanger based on a long-term time scale and a dynamic soil heat model; and based on the optimization result, determine and output a long-term operation constraint for the first heat supply loop to the local controller; The local controller is configured to: obtain predicted electricity price data and predicted heat load data in a short-term prediction time domain; strictly meet the long-term operation constraint output by the supervisory controller; optimize a short-term multi-objective cost function using a model predictive control algorithm, the function including at least: future operation electricity cost calculated based on the predicted electricity price data; and a preset depreciation cost associated with switching actions between the first heat supply loop and the second heat supply loop; and based on the result of the optimization of the short-term multi-objective cost function, switching between the first heat supply loop and the second heat supply loop is performed to minimize the short-term multi-objective cost function.

[0017] Optionally, the supervisory controller in the control device is further configured to periodically execute a model calibration subprogram to correct long-term model drift in the dynamic soil thermal model caused by time variation of the groundwater seepage field, the model calibration subprogram comprising: dynamically designating at least one of the ground heat exchangers as an excitation ground heat exchanger and a plurality of other ground heat exchangers as sensing ground heat exchangers; controlling the air source heat pump or other heat source of the system to inject an active and metered heat pulse into the ground via the excitation ground heat exchanger; high-frequency monitoring and collecting a set of asymmetric thermal response time series data caused by the heat pulse and thermal advection via the plurality of sensing ground heat exchangers; executing a preset thermal advection tomography inversion algorithm to inversely solve and generate a quantitative three-dimensional seepage vector field map using the asymmetric thermal response time series data as input; using the generated three-dimensional seepage vector field map to cover or correct the original static parameters in the dynamic soil thermal model for characterizing the groundwater seepage around the ground heat exchangers.

[0018] In a second aspect, the embodiments of the present application provide a control method of a multi-capacity complementary cross-seasonal heat storage and supply system, the system comprising a ground source heat pump, a first heat supply loop and a second heat supply loop in communication with an evaporation side of the ground source heat pump, the first heat supply loop using a ground heat exchanger for heat supply, and the second heat supply loop using an air source heat pump for heat supply, the method comprising: in a first heat supply mode, supplying heat to the evaporation side of the ground source heat pump through the first heat supply loop; monitoring first system energy consumption in the first heat supply mode; in a second heat supply mode, supplying heat to the evaporation side of the ground source heat pump through the second heat supply loop; monitoring second system energy consumption in the second heat supply mode; based on a comparison result of the first system energy consumption and the second system energy consumption, switching between the first heat supply mode and the second heat supply mode.

[0019] The present application has the following beneficial effects: Significant reduction in initial investment cost: The present invention improves the operating conditions of the ground source heat pump by raising the water inlet temperature of the evaporating side of the ground source heat pump, thereby significantly reducing the demand for the number of ground heat exchanger pipes and the drilling area, and reducing the initial investment cost of the ground source side. At the same time, the present system aims to replace the traditional heat source of the heat and power plant (primary network), and the existing heat exchange station and secondary network pipeline facilities can continue to be used without the need for large-scale modification, thereby maximizing the cost savings of the project construction.

[0020] Solve the problem of high-temperature heating of the primary network and realize secondary heating: In view of the pain point that the traditional single heat pump cannot meet the high-temperature demand of the primary network of central heating, the present invention uses a secondary heating structure combining air source heat pump and ground source heat pump to provide low-temperature heat source for the evaporating side of the ground source heat pump using air source heat pump or ground heat exchanger, thereby effectively improving the condensing temperature and heating performance of the ground source heat pump, so that high-temperature hot water meeting the delivery requirements of the primary network can be efficiently and stably prepared.

[0021] Significant reduction in carbon emissions: The present system deeply integrates various renewable energy sources such as air energy and shallow geothermal energy, and can be extended to integrate solar photovoltaic-thermal (PVT) and industrial waste heat, realizing multi-energy complementation. By replacing coal-fired or gas-fired boilers, the present system significantly reduces the consumption of fossil energy, realizes low-carbon or even zero-carbon operation in the whole life cycle, and has important environmental protection significance for promoting the central heating industry to achieve the "double carbon" goal.

[0022] Flexible operation mode, efficient use of air source heat pump all year round: The system flow design is optimized to organically combine the power supply mode and the heating mode. In terms of heating, different heating loops and heat storage loops are set to make the air source heat pump used for soil heat storage in the non-heating season; in the high-temperature period of the heating season, it can directly provide heating; in the low-temperature period of the heating season, it can be used as an efficient low-temperature heat source (secondary heating) for the ground source heat pump, thereby greatly improving the annual utilization rate of the equipment and the flexibility of the system operation.

[0023] Advanced control logic ensures cross-season heat balance: The present invention adopts a two-level control framework of "supervisory controller" and "field controller". Not only can it dynamically select the optimal heating loop through energy consumption comparison logic, but also can plan cross-season based on long-term dynamic soil heat model, effectively avoiding soil heat imbalance problem, ensuring long-term stable operation of the system.

[0024] Low operating cost: The present invention discards the traditional fixed temperature threshold switching method and adopts a linkage control strategy based on real-time "energy consumption comparison". The system can dynamically calculate and compare the real-time system overall power consumption under different heating loop combinations, and always automatically lock and switch to the operation mode with the lowest economic cost under the current operating conditions, thereby minimizing the daily operation cost of the system.

[0025] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0026] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0027] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the structure of a multi-energy complementary cross-seasonal thermal storage and heating system (basic system) in an embodiment of the present invention.

[0028] Figure 2 This is a schematic diagram of the structure of a multi-energy complementary cross-seasonal thermal storage and heating system (extended system) according to an embodiment of the present invention.

[0029] Figure 3 This is a schematic flowchart of a multi-energy complementary cross-seasonal thermal storage and heating control method in an embodiment of the present invention. Detailed Implementation

[0030] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0031] Example 1: Basic System like Figure 1 As shown, the basic system of this embodiment of the invention aims to "serve centralized heating in northern China and replace traditional thermal power plants and large-scale heating network energy stations", and includes an air source heat pump 8, a buried pipe 9, a ground source heat pump 10, an energy station 11, a heat exchange station 12, a solar photovoltaic panel 13, and a power grid 14.

[0032] The system is equipped with circulating water pumps A, B, C, and D; valves 1, 2, 3, 4, 5, 6, 7, and 15 (located at the outlet of the buried pipe); and connecting pipes J1 to J15.

[0033] Static structural connection relationships: Air source heat pump 8: Outlet connection pipe J1, inlet connection pipe J2.

[0034] Buried pipe 9: Outlet connection pipe J3, inlet connection pipe J4 (valve 15 is located here).

[0035] Ground source heat pump 10: Evaporation side inlet connection pipe J5, evaporation side outlet connection pipe J6; condensation side outlet connection pipe J9, condensation side inlet connection pipe J10.

[0036] Energy station 11: Heat input side (for example, from the ground source heat pump condensation side) connection pipe J9, heat output side (return water) connection pipe J10. The energy station 11 supplies heat to the heat exchange station 12 through the pipe J11, the circulating water pump D, and the return water of the heat exchange station 12 returns to the energy station 11 through the pipe J12.

[0037] Key loop: First heat supply loop (ground pipe heat supply): Ground pipe 9 outlet→J3→valve 3→J5→valve 6→circulating water pump B→ground source heat pump 10 evaporation side.

[0038] Second heat supply loop (air source heat supply): Air source heat pump 8 outlet→J1→valve 2→J3→J5→valve 6→circulating water pump B or water pump A drive→ground source heat pump 10 evaporation side.

[0039] Third heat source loop (ASHP direct supply): Air source heat pump 8 outlet→J1→pipe J7→valve 1→pipe J9→through circulating water pump C→energy station 11. Energy station 11 return water→J10→pipe J8→valve 4→J2→circulating water pump A→air source heat pump 8.

[0040] Heat storage loop: Air source heat pump 8 outlet→J1→valve 2 and 3→J3→ground pipe 9 inlet. Ground pipe 9 outlet→J4→valve 5 and 15→through circulating water pump A→J2→air source heat pump 8.

[0041] Electrical connection: Solar photovoltaic panel 13 preferentially supplies power to system equipment (heat pump, water pump, etc.) through line J13; when there is excess power generation, the excess power is fed into the grid through line J15; when the power generation is insufficient, the power is purchased from the grid 14 through line J14.

[0042] Dynamic operation mode: The system cooperatively controls each heat supply loop and power supply loop through a control device (for example, a PLC controller), and divides the system operation into a heat supply operation mode and a power supply operation mode, and the specific control logic is as follows: I. Heat supply operation mode The heat supply operation of the system strictly follows the time logic of "cross-season", and is divided into a heat storage working condition in a non-heating season and a heat supply working condition in a heating season.

[0043] 1. Non-heating season (cross-season heat storage mode) In the non-heating season (for example, spring, summer, and autumn), the core task of the system is to restore the soil heat balance. The control device controls the heat storage loop to be turned on and operates in the "soil heat storage working condition" (see Table 1 for details). At this time, the air source heat pump 8 collects heat energy from the air, or the solar photovoltaic panel 13 drives the system to inject heat into the soil around the ground buried pipe 9, reserves heat energy for winter heating, and prevents the ground temperature from decreasing year by year.

[0044] 2. Heating season (cascade heating mode) In the heating season (winter), the system flexibly switches among the three sub-modes according to the changes of outdoor environment temperature and ground buried pipe outlet water temperature, to ensure heating and reduce energy consumption: (1) Direct heating working condition (see Table 2 for details): In the transition season or when the outdoor temperature is relatively high in the heating season, the air source heat pump 8 has high energy efficiency. At this time, the third heating loop is controlled to be turned on, and the air source heat pump 8 directly supplies heat to the energy station 11, which also serves as a peak shaving or standby heat source.

[0045] (2) Ground source heat pump main supply working condition (see Table 3 for details): In the heating season, when the outdoor temperature is relatively low (such as in winter), the first heating loop is controlled to be turned on. The ground buried pipe 9 extracts the heat stored in the soil across seasons to provide a stable low-temperature heat source for the evaporation side of the ground source heat pump 10, and the ground source heat pump 10 supplies heat to the energy station 11.

[0046] (3) Air source secondary heating working condition (see Table 4 for details): When the ground buried pipe 9 continuously extracts heat, causing the outlet water temperature to decrease and the energy efficiency of the ground source heat pump 10 to decrease, the second heating loop is controlled to be turned on. The air source heat pump 8 is started as a "primary heating source" to increase the inlet water temperature of the evaporation side of the ground source heat pump 10 (secondary heating), thereby maintaining the high efficiency of the system.

[0047] II. Power supply operation mode The system uses the solar photovoltaic panel 13 to generate electricity, and cooperates with the power grid 14 to realize multi-energy complementary power supply. The control device implements differentiated power dispatching strategies according to the characteristics of different seasons: 1. Non-heating season (power generation and grid-connected strategy) In the non-heating season, the system only operates in the heat storage mode or is in standby state, and the overall power consumption is low. At this time, the electric energy generated by the solar photovoltaic panel 13 is mainly in the "full grid-connected" or "surplus grid-connected" mode, and is transmitted to the power grid 14 to obtain electricity sales revenue and improve the economic return rate of the system, in addition to a small amount of power supply for maintaining the basic operation of the system.

[0048] 2. Heating season (self-generation and self-use strategy) In the heating season, the heat pump unit and circulating water pump are in high-load operation state, and the system consumes a large amount of electricity. At this time, the electric energy generated by the solar photovoltaic panel 13 strictly implements the "self-generation and self-use" strategy, and is preferentially supplied to the air source heat pump 8, the ground source heat pump 10, and various circulating water pumps.

[0049] If the photovoltaic power generation is greater than the instantaneous power consumption of the system, the excess power is transmitted to the power grid 14 (surplus power on the grid); If the photovoltaic power generation is insufficient to support system operation (such as at night or on cloudy days), the control line J14 automatically purchases power from the power grid 14 to supplement it, ensuring uninterrupted heating.

[0050]

[0051]

[0052] Example Two: Extended System As shown in Figure 2 , the extended system of the present application is built on the basis of the basic system, further integrating various forms of energy utilization to achieve more efficient cascade utilization and functional expansion.

[0053] New static structural components: New equipment: air source heat pump 15 (corresponding to 8 of the basic system), ground heat exchanger 16 (corresponding to 9), industrial waste heat source 17, water tank 19, energy station 20 (for steam) and energy station 21 (for hot water), steam heat pump 22, steam user 23, battery 25, PVT component 27, plate heat exchanger 28 (connecting waste heat), plate heat exchanger 29 (connecting water tank).

[0054] New water pumps: circulating water pump G (water tank-plate exchanger 29), circulating water pump H (PVT-water tank), circulating water pump I (industrial waste heat). (Pumps E, F are steam circuit water pumps).

[0055] New valves: valves 8, 9, 10, 11, 12, 13, 14, 30 (ground heat exchanger outlet).

[0056] New pipelines / lines: J11-J27.

[0057] Key loops: PVT and water tank: PVT 27→J11→circulating water pump H→water tank 19→J12→PVT 27.

[0058] Water tank preheating: water tank 19→J15→circulating water pump G→plate heat exchanger 29.

[0059] Waste heat preheating: industrial waste heat source 17→J13→circulating water pump I→valve 13→J14→plate heat exchanger 28.

[0060] Ground loop secondary heat extraction loop: Ground loop 16→J4→valve 30→J5→valve 7→circulating water pump B→valve 8→plate heat exchanger 28 (industrial waste heat primary heat extraction)→valve 9→valve 10→plate heat exchanger 29 (PVT water tank secondary heat extraction)→valve 11→ground source heat pump evaporation side→J6→valve 6→J3 valve 3→ground loop 16.

[0061] Steam supply: Energy station 20→J19→circulating water pump E→steam heat pump 22 evaporation side→J20→energy station 20. Steam heat pump 22 condensation side→J22→circulating water pump F→steam user 23.

[0062] Waste heat cascade utilization: When the temperature of waste heat 17 is high (such as ≥ 70℃), it flows through J13→water pump I→J9→and is supplied to energy stations 20, 21 in parallel. Its return water→J10→J14→enters the plate heat exchanger 28 again to preheat the ground loop outlet water.

[0063] Electrical connection of the extended system: The electrical logic of the extended system introduces the battery 25 as a buffer.

[0064] Power supply priority: PVT 27 power generation J23→preferentially supplies each component of the system J23→if there is excess power, it is preferentially stored in the battery 25 through the line J25→if the battery 25 is full, the excess power is fed into the grid J26.

[0065] Power consumption priority: preferentially use PVT 27 power generation J23→if the power is insufficient, call the battery 25 power J24→if the battery 25 is exhausted, purchase power from the grid J27.

[0066] Dynamic operation mode of the extended system: Based on the increased PVT components, industrial waste heat and battery energy storage units, the operation mode of the embodiment is further refined as: I. Heating operation mode 1. Non-heating season (multi-energy storage mode) system actively heats the soil using idle heat sources: (1) Waste heat + PVT collaborative heat storage (see Table 5 for details): When industrial waste heat is sufficient and the PVT water tank temperature is relatively high, both heat sources are used to heat and store heat in the ground loop.

[0067] (2) ASHP alone heat storage (see Table 6 for details): In the case of insufficient light or no industrial waste heat, an air source heat pump is enabled to prevent the soil temperature from falling.

[0068] 2. Heating season (cascade utilization heating mode) according to the grade of heat source: (1) Heating mode 1 (ASHP direct heating, see Table 7 for details): When the air temperature is high, the air source heat pump directly heats.

[0069] (2) Heating mode 2 (high-temperature waste heat cascade utilization, see Table 8 for details): when the industrial waste heat temperature is ≥70℃, direct parallel heating, and the return water is used for preheating the ground buried pipe.

[0070] (3) Heating mode 3 (low-temperature waste heat utilization, see Table 9 for details): when the industrial waste heat temperature is <70℃, only used for preheating the ground buried pipe outlet water.

[0071] (4) Heating mode 4 (ASHP secondary heating, see Table 10 for details): when there is no waste heat and the ground temperature is low, secondary heating is provided by an air source heat pump.

[0072] II. Power supply operation mode The extended system introduces a battery 25 as an energy buffer, and the power supply strategy is adjusted as follows: 1. Non-heating season (energy storage and grid access) At this time, the system heat load is low. The electricity generated by the PVT component 27 is stored in the battery 25 for future use or night heat storage; when the battery is full, the excess electricity is all transmitted to the power grid (grid access).

[0073] 2. Heating season (peak shaving and self-use) At this time, the system heat load is high. The electricity generated by the PVT component 27 is preferentially supplied to the system components for "self-generation and self-use".

[0074] When the photovoltaic power is insufficient, the control device preferentially calls the stored electricity in the battery 25 for supplementation (peak shaving); When the battery power is depleted, electricity is purchased from the power grid.

[0075] If there is excess electricity, it is stored in the battery or transmitted to the power grid.

[0076]

[0077] Example Three: Cross-season operation optimization based on level control strategy For clarity, the control strategy (or method) of this embodiment will be described in conjunction with the basic system shown in Figure 1 , such as air source heat pump 8, ground buried pipe 9, and ground source heat pump 10. However, those skilled in the art should understand that the control logic and algorithm described in this embodiment are also applicable to the extended system shown in Figure 2 , such as controlling air source heat pump 15, ground buried pipe 16, and industrial waste heat source 17, and do not constitute a limitation of the present application.

[0078] This embodiment describes an advanced control device for the system of the present invention and its working principle. The control device implements a hierarchical control strategy aiming at ensuring the operational sustainability of the system (especially the seasonal heat balance of the ground loop 9) throughout its life cycle (e.g. twenty years), while minimizing the economic cost of short-term operation (e.g. the next twenty-four hours) subject to the long-term constraint.

[0079] In this embodiment, the control device is physically and logically implemented as a two-level distributed control system, which includes a supervisory controller and a field controller.

[0080] Supervisory controller: The supervisory controller can be physically deployed in a remote cloud server or the data center of a district energy management center. It utilizes powerful cloud computing resources to perform complex, computationally intensive, non-real-time simulation and optimization tasks. Its control time scale is very long, e.g. planning and simulation in years or seasons. The main task of the supervisory controller is to analyze the long-term sustainability of the system, especially the soil heat balance around the ground loop 9.

[0081] Field controller: The field controller is physically deployed on site at the energy station where the heating system is located, e.g. a high-performance industrial programmable logic controller (PLC) or edge computing gateway. It is responsible for executing real-time, short-cycle control decisions. Its control time scale is very short, e.g. fifteen minutes or one hour as a control step, and looks ahead to the next twenty-four to seventy-two hours of operation.

[0082] Data flow and coordination: The two controller levels interact through a secure communication network (e.g. industrial Ethernet or cellular Internet of Things private network). This hierarchical architecture is the key to realizing complex system optimization: the supervisory controller is responsible for calculating and formulating a long-term operational constraint, and periodically (e.g. every month or at the beginning of each heating season) issuing this constraint to the field controller. The field controller then autonomously performs optimal economic dispatch in the short term while strictly adhering to the high-level issued constraint.

[0083] The core of the supervisory controller's work is to optimize the seasonal heat balance of the ground loop 9 based on a dynamic soil heat model.

[0084] Dynamic soil thermal model construction: At the initial deployment of the system, the high-level controller will integrate a high-precision three-dimensional dynamic soil thermal model. This model is a digital twin of the local ground loop 9 area. Building this model requires input of detailed on-site geological exploration data, which at least includes: the types, densities, specific heat capacities, thermal conductivities, porosities of the soil and rock mass, and (most importantly) the average seepage velocity and direction of the underground water. These thermal physical parameters are the basis for ensuring that the model accurately predicts the effects of heat conduction, convection, and accumulation. Those skilled in the art will understand that this model is functionally a numerical solver of the three-dimensional partial differential heat conduction and heat convection equation set, which is configured to: (a) receive real-time operation data of the ground loop 9 (e.g., inlet / outlet temperature, flow of circulating fluid) as heat source / sink terms; (b) receive meteorological data (e.g., ground surface temperature, precipitation) as upper boundary conditions; (c) receive underground water data (e.g., seepage velocity) as convection terms; (d) and based on these inputs, solve the distribution and evolution of the underground soil temperature field for one or more time steps in the future (e.g., 20 years in the future). The supervisory controller uses this model as a simulation platform.

[0085] Optimization process for cross-seasonal thermal balance: The purpose of optimizing cross-seasonal thermal balance in this invention is to avoid a fatal flaw that may occur in the long-term operation of the ground-source heat pump 10 system, i.e., heat imbalance. Without long-term management, the soil temperature around the ground loop 9 may experience an irreversible annual decline (e.g., from the initial 14.78 degrees Celsius to 13.00 degrees Celsius) in just a few years (e.g., six to eight years), which will cause the heating efficiency of the ground-source heat pump 10 to deteriorate year by year, eventually leading to the system failing to extract heat during the winter peak period.

[0086] The supervisory controller solves this problem by: It uses the dynamic soil thermal model to perform accelerated simulation on a long-term time scale (e.g., a scale of twenty years in the future). The controller will evaluate a variety of preset annual operation strategies, such as: Strategy A: unrestricted use of the first heating loop (ground loop 9) throughout the heating season.

[0087] Strategy B: at the end of the heating season (e.g., March), forcibly switch to the second heating loop (air-source heat pump 8) to protect soil heat.

[0088] Strategy C: in the non-heating season, forcibly supplement a certain amount of heat to the soil using the heat storage loop.

[0089] The high-level controller predicts the evolution of the soil temperature field over the next 20 years for each strategy by simulating the above strategies. Its optimization goal is to screen out a sustainable operation strategy combination that can ensure the annual average soil temperature to be basically flat with the initial soil temperature after 20 years (i.e. to achieve cross-seasonal thermal balance).

[0090] The optimization result of the supervisory controller is finally quantified as one or a set of specific long-term operation constraints and issued as instructions to the field controller.

[0091] This constraint is the red line to ensure the long-term survival of the system. It is not a fixed value, but usually a set of boundary conditions that dynamically change over time. This embodiment provides two specific constraint forms: Constraint Form One: Dynamic Hard Lower Limit of the Borehole 9 Outlet Temperature The high-level controller calculates a minimum temperature threshold of the circulating liquid at the borehole 9 outlet in different months of the heating season according to its long-term thermal balance simulation.

[0092] For example, the constraint file issued by the high-level controller to the bottom layer may stipulate: November to December (soil heat is sufficient): the minimum allowable outlet temperature is 5.0 degrees Celsius.

[0093] January to February (peak heat extraction period): the minimum allowable outlet temperature is raised to 5.5 degrees Celsius.

[0094] March (end of heating season, protective recovery period): the minimum allowable outlet temperature is further raised to 6.5 degrees Celsius to ensure sufficient heat base for soil heat recovery in the non-heating season.

[0095] Constraint Form Two: Upper Limit of Daily Heat Extraction of the First Heating Loop Alternatively or in addition, the high-level controller can also issue a daily maximum allowable heat extraction amount (e.g. in gigajoules / day) based on its simulation results in different stages of the heating season. For example, it is stipulated that the total heat extracted from the soil through the first heating loop per day in the later stage of the heating season cannot exceed a specific quota.

[0096] The field controller must take these dynamic thresholds or quotas as insurmountable boundaries when performing its short-term economic dispatch.

[0097] The field controller is the executor of the daily operation of the system. It uses the Model Predictive Control (MPC) algorithm to minimize the total operation cost in the future short-term prediction time domain while strictly complying with the long-term operation constraints issued by the high-level.

[0098] Data Acquisition: At the beginning of each control cycle (e.g. every fifteen minutes), the field controller will proactively acquire rolling prediction data for a short-term prediction horizon (e.g. forty-eight hours ahead) from multiple external data sources through its data interfaces. These data include at least: Forecasted electricity price data: forecasted time-of-use or real-time electricity price curve for the next forty-eight hours from the local grid operator or electricity retailer.

[0099] Forecasted weather data: high-precision forecast of outdoor ambient temperature, humidity, wind speed, solar radiation intensity, etc. for the next forty-eight hours from a weather service agency.

[0100] Forecasted heat load data: a heat load prediction module inside the controller that automatically generates a forty-eight-hour ahead, fifteen-minute resolution heat load demand curve based on acquired weather forecast data, current date (weekday or holiday), and historical heat load data.

[0101] Execution logic of the model predictive control (MPC) algorithm: The core of the MPC algorithm in this embodiment is a rolling optimization and feedback correction control strategy.

[0102] Suppose the current time is eight o'clock in the morning, the control cycle is fifteen minutes, and the prediction horizon is twenty-four hours.

[0103] Prediction: the controller acquires all forecasted electricity price data, forecasted heat load data, and weather data from eight o'clock in the morning to eight o'clock the next morning.

[0104] Optimization: At the current time (8am), the controller uses its internal system performance model to compute a complete, 24-hour optimal control sequence. The internal system performance model is the core of the control device of the present invention, a dynamic prediction model based on physics or data-driven. This model is pre-established and validated at the time of system deployment. Its function is: (a) receive the current measurable states of the system (e.g. outdoor temperature, borehole outlet temperature, current thermal load, energy station return water temperature) and a candidate control sequence (e.g. for the next 24 hours, every 15 minutes, use loop 1 or loop 2); (b) in response, the model is configured to predict and output the future system states that the control sequence will lead to. The future system states include at least: the predicted system total power consumption at every future time step (for the calculation of cost component one), and the borehole outlet temperature at every future time step (for the checking of high-level constraints). The optimization step of the MPC algorithm is essentially to repeatedly call this performance model to iteratively search for that optimal control sequence which minimizes the total value of the cost function (electricity + depreciation), and whose predicted borehole outlet temperature does not violate the high-level constraints. This sequence will specify in detail whether the first heating loop, the second heating loop, or both should be operated at every future 15-minute step (96 steps in total). Its optimality criterion is to minimize the total value of a short-term multi-objective cost function over the next 24 hours.

[0105] Execution: The controller does not execute this complete 24-hour plan. It only adopts the first step of the plan, the control decision from 8am to 8:15am (e.g. decide to start the second heating loop), and immediately executes this decision.

[0106] Feedback and correction: At the next time (8:15am), the controller discards all the previously computed 95 remaining steps. It acquires the latest actual states of the system (e.g. actual outlet water temperature of the energy station, actual thermal load response, this is the feedback correction), and acquires new 24-hour prediction data updated from 8:15am to 8:15am the next day.

[0107] Rolling: The controller repeats the above optimization step to re-compute a completely new, 24-hour optimal control sequence, and only executes the first step of the new sequence (i.e. the decision from 8:15am to 8:30am).

[0108] This cycle from prediction, optimization, execution, feedback to re-prediction is rolled out every 15 minutes, ensuring that the control decisions of the system can respond in real-time to changes in prediction data (e.g. sudden jump in electricity price) and deviations in actual load.

[0109] The core of the field controller in performing the optimization step of the MPC algorithm is to solve the minimum of a short-term multi-objective cost function. This function is a quantitative representation of the short-term operation economy of the system. This function is at least a weighted sum of the following two cost components: Cost component one: future operation electricity cost calculated based on the predicted electricity price data This is the main economic objective of the optimization. In the optimization simulation of the MPC, the controller will calculate the cost for each time step (e.g. fifteen minutes) in the prediction horizon (e.g. twenty-four hours): Calculation loop 1 (ground loop 9): the controller calculates the predicted electricity consumption of driving the ground source heat pump 10 and circulating pump B, using its internally stored performance curve (i.e. COP-temperature relationship) of the ground source heat pump 10, based on the predicted heat load of this time period and the current (or predicted) outlet temperature of the ground loop 9. Then it multiplies this electricity consumption by the predicted electricity price of this time period to get the electricity cost of operating loop 1.

[0110] Calculation loop 2 (air source): the controller calculates the predicted total electricity consumption of driving the air source heat pump 8, ground source heat pump 10 and circulating pump A, using its internally stored performance curve (i.e. COP-outdoor temperature relationship) of the air source heat pump 8, based on the predicted heat load of this time period and the outdoor temperature from the weather forecast. Then it multiplies this total electricity consumption by the predicted electricity price of this time period to get the electricity cost of operating loop 2.

[0111] One of the optimization objectives of the MPC algorithm is to select the appropriate loop combination in the prediction horizon so that the total sum of the future operation electricity cost is minimized. For example, use the loop with lower efficiency but cheaper electricity price in the low electricity price period, and lock the loop with the highest efficiency in the peak electricity price period.

[0112] Cost component two: preset depreciation cost associated with switching actions This is a key non-electricity cost item for protecting the life of the equipment and ensuring the stability of the operation.

[0113] Technical problem: if the cost function only contains the above cost component one (electricity cost), then in some operating conditions (e.g. when the outdoor temperature and the outlet temperature of the ground loop 9 make the operation electricity costs of the two loops extremely close), a slight fluctuation in electricity price or a slight disturbance in prediction may cause the MPC algorithm to frequently swing between the most economical loop 1 and the most economical loop 2. This may cause the controller to issue multiple switching instructions in a short period of time (e.g. within one or two hours).

[0114] Technical consequences: such frequent switching actions involve the start-stop and switching of large compressors, high-power circulating pumps and pipe valves. Frequent start-stop will cause huge electrical impact and mechanical wear and tear, greatly shortening the service life of the core equipment, i.e. generating high depreciation cost.

[0115] Solution: To avoid this harmful frequent switching, a cost component two is introduced into the multi-objective cost function of this invention.

[0116] In the optimization simulation of MPC, whenever a switching action occurs in a control sequence, switching from the first heating loop to the second heating loop (or vice versa), a fixed penalty value is artificially and additionally added to the total cost of that sequence. This penalty value represents a preset depreciation cost associated with the switching action between the first and second heating loops.

[0117] By introducing this depreciation cost penalty, the optimization logic of MPC has changed. It will no longer perform a switchover to save a few cents in electricity costs (cost component one); the controller will only consider a switchover economical and optimal if the future electricity cost savings from the switchover are significantly greater than (or equal to) the preset depreciation cost incurred in performing the switchover (cost component two). This ensures that once the system selects an operating loop, it will run stably for a longer period, effectively avoiding harmful frequent start-stop cycles.

[0118] The final control logic in this embodiment combines the long-term operational constraints of the supervisory controller with the short-term multi-objective cost function of the field controller.

[0119] When the field controller performs rolling optimization of MPC, the highest criterion of its solution algorithm (e.g., mixed integer programming or dynamic programming) is that it must minimize the short-term multi-objective cost function while strictly satisfying the long-term operating constraints output by the supervisory controller.

[0120] Workflow example (satisfying all constraints): The following scenario illustrates the complete workflow of this level control strategy: Scenario: Late heating season (March). Outdoor nighttime (3:00 AM), predicted moderate heat load.

[0121] High-level constraint: The monitoring controller issued an instruction a month ago that, to ensure sufficient recovery of soil heat during the non-heating season, the outlet temperature of underground pipe 9 must not be lower than 6.5 degrees Celsius in March. This is a long-term operational constraint.

[0122] Low-level data: Obtained by the field controller: Electricity price forecast: Electricity prices will be at their lowest between 3:00 AM and 5:00 AM.

[0123] Sensor data: The current outlet temperature of buried pipe 9 is 6.6 degrees Celsius, which is close to the constraint red line.

[0124] The underlying MPC has begun rolling optimization: The controller starts a simulation to calculate the optimal solution for the next twenty-four hours. For the decision at 3 AM, it evaluates at least the following options: Simulation option A (cost only): Run the first heating loop (ground loop 9) alone.

[0125] Cost function calculation: Since the COP (coefficient of performance) of the ground source heat pump 10 can still be slightly higher than that of the air source heat pump 8 at this time, and the electricity price is at a low point, the future electricity cost of this option is the lowest.

[0126] Constraint check: The internal model of the controller (i.e., the system performance model) predicts that if this option is executed, the ground loop 9 outlet temperature will drop to 6.4 degrees Celsius in fifteen minutes.

[0127] Simulation option B (constraint first): Switch completely to the second heating loop (air source heat pump 8).

[0128] Cost function calculation: Since the air source heat pump 8 is started, the total electricity cost is higher than option A. But there is no switching (assuming it was on loop 2 a moment ago), and the depreciation cost is zero.

[0129] Constraint check: The ground loop 9 outlet temperature will slowly rise (e.g., to 6.7 degrees Celsius), satisfying the constraint of not being lower than 6.5 degrees Celsius.

[0130] Decision execution: The MPC optimization algorithm of the on-site controller, when evaluating, first determines that simulation option A violates the long-term operation constraints issued by the upper layer. Therefore, even if option A has the lowest short-term electricity cost, it must be discarded by the algorithm as an infeasible solution.

[0131] Then, the algorithm selects the solution that minimizes the total value of the short-term multi-objective cost function (electricity cost + depreciation cost) among all feasible solutions that satisfy the long-term operation constraints (e.g., option B).

[0132] In this scenario, the controller will enforce option B (or other mixed loops that satisfy the constraints), sacrificing short-term electricity economy to strictly meet the long-term sustainability requirements of the system.

[0133] In this way, the hierarchical control strategy described in this embodiment perfectly combines the long-term sustainability planning of the supervisory controller and the short-term economic optimization of the on-site controller. It ensures that the system is running in the most economical (bottom-layer cost function) way at any time (upper-layer constraints).

[0134] Embodiment Four: The embodiment provides a detailed technical solution for implementing long-term model self-adaptive calibration in the supervisory controller on the basis of the hierarchical control strategy (including a supervisory controller and a field controller) described in Embodiment Three.

[0135] The application object of the embodiment is to solve a hidden but major technical problem faced by the supervisory controller (HLC) in performing its long-term optimization task in Embodiment Three. The technical problem is caused by long-term model drift between the dynamic soil heat model relied on by the HLC and the real physical world.

[0136] Specifically, the dynamic soil heat model described in Embodiment Three relies on one-time field geological exploration data when it is constructed at the initial stage of system deployment (i.e., the 0th year). In these data, a parameter that is crucial to the accuracy of the model, i.e., the underground water seepage field around the ground heat pipe array, is initialized as a static or average estimated value in the model.

[0137] However, in the actual operation life cycle of the heating system for 20 years or more, the underground water seepage field in the physical world is by no means static. The real underground water seepage speed and direction will change slowly but significantly over time due to seasonal rainfall changes (e.g., changes in hydraulic gradient caused by wet and dry seasons), engineering construction in the surrounding area of the system (e.g., regional dewatering caused by large foundation pit excavation or tunnel grouting changing rock-soil permeability), urbanization development (e.g., large-area ground hardening changing the groundwater recharge mode), and even long-term heat injection or extraction of the system itself (causing slight changes in the viscosity of underground water).

[0138] The fundamental contradiction between this static model parameter and the dynamic physical reality will inevitably cause the dynamic soil heat model relied on by the HLC to produce irreversible and gradually accumulated deviations between its model prediction values and the physical real values over time (e.g., in the 5th year, the 10th year, and the 15th year). This is the long-term model drift.

[0139] The long-term model drift has a great impact on the hierarchical control strategy described in Embodiment Three. If not calibrated, the actual underground water seepage field in the ground heat pipe area may change due to surrounding construction and other factors as the operation life increases. At this time, if the initial static model parameter is still used, it will cause deviations in the long-term operation constraints (such as the lower limit of the ground heat pipe outlet temperature) calculated by the supervisory controller. The field controller may run based on the wrong constraints, which may cause excessive heat extraction from the soil, destroy the cross-season heat balance, and ultimately reduce the long-term heating efficiency of the system.

[0140] To fundamentally solve the long-term model drift problem mentioned above, the supervisory controller in the control device of the embodiment is further configured to periodically (or intelligently triggered when necessary) execute a model calibration subprogram.

[0141] The subprogram exists as a preset functional module of the supervisory controller in the software architecture, which can be physically deployed in the remote cloud server or data center of the regional energy management center described in Embodiment Three.

[0142] The trigger logic of the subprogram is configured in at least two modes: Periodic timing trigger: HLC is configured to automatically start this calibration subprogram, for example, at the beginning of the non-heating season every year (e.g., fixed May 1st at midnight). The reason for choosing this time point is that: (a) in the non-heating season, the heating system (such as air source heat pump) is in low load or shutdown state, which can be called by HLC as a heat source for injecting heat pulse; (b) at this time, the underground soil temperature field has been naturally restored for a spring, and is relatively stable, without being disturbed by the heat extraction in the heating season or the heat storage in the non-heating season, thus providing a clean thermal background condition for high-precision thermal response measurement.

[0143] Intelligent error trigger: HLC continuously compares the predicted values (e.g., predicted buried pipe outlet temperature, predicted energy station return water temperature) of its dynamic soil heat model with the actual measured values collected from the field sensors during normal operation. When HLC detects that the cumulative error of model prediction (e.g., root mean square error for seventy-two consecutive hours) continuously exceeds a preset error threshold (e.g., five percent), HLC determines it as a sign of model drift. Even if it does not reach the timing period, HLC will actively trigger this calibration subprogram to deal with (for example) the dramatic changes in the seepage field caused by sudden surrounding construction.

[0144] Once triggered, the supervisory controller (HLC) will automatically execute the following five core steps of the model calibration subprogram in sequence: Step (a): dynamically specify at least one of the buried pipes as an excitation buried pipe, and specify a plurality of other buried pipes as sensing buried pipes In this step, HLC first reads the complete three-dimensional topological structure diagram of the entire buried pipe array (e.g., buried pipe 9 in Figure 1 from its configuration library of the dynamic soil heat model described in Embodiment Three. The structure diagram accurately describes the precise coordinate position of each buried pipe (e.g., a 10x10 array, a total of 100 buried pipes) in the underground (X, Y, Z).

[0145] The HLC implements a dynamic designation strategy. Rather than randomly designating, the HLC implements an optimized, iterative test sequence to ensure that subsequent tomographic imaging can be performed from multiple different angles.

[0146] In a typical calibration sequence, the HLC can perform the following iterations: Iteration One (Center Excitation): The HLC designates a certain buried pipe (e.g., Buried Pipe 55) located in the center region of the array as the excitation buried pipe. At the same time, the HLC automatically designates all first-order neighbors (i.e., physically closest) of this buried pipe (e.g., Buried Pipes 45, 54, 56, 65) as the sensor buried pipes according to the three-dimensional topology. All other buried pipes in the array (e.g., Buried Pipes 1 through 44) are temporarily ignored (their valves remain closed) in this iteration.

[0147] Iteration Two (Edge Excitation): After completing the tests and collecting data from Iteration One, the HLC can designate a certain buried pipe (e.g., Buried Pipe 1) located in the edge region of the array as the new excitation buried pipe and designate its neighbors as the new sensor buried pipes, then repeat the subsequent steps, depending on the preset strategy.

[0148] Through this dynamic and iterative designation (rather than testing only once), the HLC can obtain richer, more dimensional underground thermal response data, providing higher-quality input for the thermal plume tomographic inversion algorithm in Step (d), thereby inverting a more accurate three-dimensional seepage field map.

[0149] Step (b): Control the air source heat pump or other heat source of the system to inject an active and metered heat pulse into the ground via the excitation buried pipe After the HLC completes the designation in Step (a) (e.g., designates Buried Pipe 55 as the excitation buried pipe), it will immediately take over and precisely control the hardware devices of the system (refer to Figure 1 ) to perform an active and metered heat pulse injection.

[0150] The following is a typical hardware control timing sequence of the HLC performing this step (taking the base system shown in Figure 1 as an example, assuming that each sub-circuit of Buried Pipe 9 has an HLC-controllable independent valve): Isolate the main loop: The HLC first issues an instruction to close Valves 3, 15 (the buried pipe total outlet valve), 6, and 7 (the ground source heat pump evaporator side valve). The purpose of this action is to completely disconnect the buried pipe array (Buried Pipe 9) from the main loop (i.e., the first heating loop) of the ground source heat pump 10, ensuring that the heat pulse does not interfere with the heating main system.

[0151] Establish the stimulation loop: HLC then controls the specific valve combination to establish a dedicated stimulation loop. HLC opens valve 2 (ASHP outlet to the ground loop) and valve 5 (ground loop outlet to ASHP). It is crucial that HLC only opens the specific sub-loop valve leading to the designated stimulation ground loop (i.e. number 55), while keeping all other 99 ground loop sub-loop valves closed.

[0152] Start the heat source: HLC starts air source heat pump 8 and circulation pump A. In this scenario, air source heat pump 8 plays the role of the other heat source of the system.

[0153] Inject the heat pulse: Circulation pump A drives the circulation medium. The hot water (e.g. 40 degrees Celsius) generated by air source heat pump 8, via pipe J1→ open valve 2→ pipe J3→ only flows into that selected stimulation ground loop (number 55), releases heat underground, and via pipe J4→ open valve 5→ pipe J2→ returns to circulation pump A→ re-enters air source heat pump 8.

[0154] Control the injection time: HLC maintains this stimulation loop running for a preset injection time (e.g. 8 hours of continuous injection), forming a high-intensity heat pulse.

[0155] In this step, it is crucial to be active and to be quantitative: Active: This is not a passive temperature observation. It is an active interference test initiated by HLC, forcibly injecting heat into the ground.

[0156] Quantitative: During the above 8 hours of injection, HLC simultaneously monitors (e.g. every second) the instantaneous flow rate of the circulation water flowing through the stimulation ground loop (number 55) and the instantaneous inlet-outlet temperature difference through the temperature sensors and flow meters deployed at the inlet and outlet of the stimulation ground loop. HLC's internal processor accurately calculates the total heat injected into the ground (e.g. a total of 2.5 gigajoules) by integrating (instantaneous flow rate x instantaneous temperature difference x specific heat capacity of the circulation medium) over the entire injection time (8 hours).

[0157] This accurate, quantitative total heat value is a key, known input parameter (i.e. the strength of the stimulation source) for the inversion algorithm in subsequent step (d) to solve.

[0158] Step (c): High-frequency monitoring and collection of a set of asymmetric heat response time series data through the multiple sensing ground loops caused by the combined action of the heat pulse and heat advection The core of this step is to observe how the heat pulse injected in step (b) propagates underground and capture the key signals caused by underground water flow.

[0159] Under the influence of groundwater thermal advection, the heat diffusion in soil will exhibit asymmetric characteristics. The temperature rise time monitored by the sensor geothermal pipe located downstream of groundwater flow will be earlier than that of the upstream, and the temperature rise amplitude is higher.

[0160] The flow of groundwater (advection) will inevitably lead to the asymmetry of heat diffusion in the ground.

[0161] HLC is just using this physical principle to monitor and collect this asymmetric thermal response at high frequency: Monitoring start and duration: At the same time (0 hours) when the heat pulse starts to be injected in step (b), HLC instructs all designated sensor geothermal pipes (e.g., numbers 45, 54, 56, 65) to start recording their outlet water temperature at high frequency (e.g., every five minutes) from the outlet temperature sensor.

[0162] Monitoring duration: HLC will continue to record these data for a total of 72 hours. This monitoring total duration is much longer than the 8-hour injection time, and the purpose is to completely capture the complete thermal response curve of the heat pulse from the beginning to reach (temperature rise), reach the peak, and finally decay back to the thermal background.

[0163] Capture of asymmetric signals: Assuming that the groundwater is flowing northward at a speed of 0.1 meters per day (which HLC does not yet know). HLC will collect a set of asymmetric data as follows: Downstream sensor (north side, number 45): Since it is located downstream of the groundwater flow, the heat pulse is pushed by the groundwater to accelerate to reach. This sensor will monitor the temperature rise earliest (e.g., 15 hours) and the temperature rise amplitude is the largest (peak highest).

[0164] Upstream sensor (south side, number 65): Since it is located upstream, the groundwater flow carries away part of the heat that should have conducted southward. This sensor will monitor the temperature rise latest (e.g., 30 hours) and the temperature rise amplitude is the smallest (peak lowest).

[0165] Lateral sensors (west side number 54, east side number 56): Their response is between the above two, and may also exhibit a small asymmetry (depending on whether the flow direction is due north).

[0166] Data processing and input preparation: After 72 hours, HLC collects all time series data of all (e.g., 4) sensors (e.g., 4 sensors x 12 points / hour x 72 hours = 3456 data points). The processor of HLC will preprocess this set of raw data, such as signal denoising (filtering out the electronic noise of the sensor itself), baseline correction (uniformly subtracting the soil background temperature before the test (e.g., 14.5 degrees Celsius)), to generate a set of clean, standardized, and only containing temperature rise information curves.

[0167] This set of standardized, asymmetric thermal response time series data, which is the effect observed in the real physical world, is caused by the stimulus (cause) in step (b). It will serve as the core input for the next step of the inversion algorithm.

[0168] Step (d): Execute a pre-set thermal plume tomography inversion algorithm, taking the asymmetric thermal response time series data as input, to inversely solve and generate a quantitative three-dimensional seepage vector field map This step is the core of the model calibration subprogram. HLC will call a dedicated software package, the thermal plume tomography inversion algorithm, which is pre-set in the HLC controller (or its connected cloud server).

[0169] The thermal plume tomography inversion algorithm takes the asymmetric thermal response time series data as observations and the thermal pulse data as the stimulus source, and inversely solves the groundwater seepage vector field through iterative calculation.

[0170] This thermal plume tomography inversion algorithm is exactly the automated calculation tool that reverses the cause from the effect. When HLC calls this algorithm, it will provide it with three key inputs: Input 1 (geometry): The three-dimensional topological structure map determined in step (a), which contains all the stimuli and sensing buried pipes.

[0171] Input 2 (stimulus source): The total heat value injected into the ground accurately measured in step (b) (for example, 2.5 gigajoules).

[0172] Input 3 (observed effect): The set of asymmetric thermal response time series data collected and pre-processed in step (c).

[0173] After receiving these three inputs, the algorithm begins an iterative, convergent calculation process within itself to inversely solve the unknown cause (i.e., the seepage field). This process can be logically divided into the following internal workflow: Step d-1: Establish numerical grid and initial guess: First, the algorithm establishes a high-resolution three-dimensional numerical grid representing the region of the buried pipe array in the HLC's calculation memory according to input 1 (geometry) (for example, a 100x100x100 calculation model containing one million nodes).

[0174] Then, the algorithm must assign an initial guess value to the unknown cause (seepage vector field). For example, in the first iteration, it assigns a zero vector to each of the one million grid nodes (i.e., it guesses that the groundwater flow velocity is zero in all directions at all locations).

[0175] Step d-2: Perform forward calculation (predict the effect): The algorithm calls its internal forward model (a numerical solver for heat conduction and thermal advection). It uses the initial guess vector field from step d-1 (i.e. zero flow velocity) and input 2 (the heat stimulus) to calculate how the heat pulse should have propagated under this zero flow condition, and predicts a theoretical heat response time series (i.e. the temperature rise curve that the sensors 45, 54, 56, 65 should have seen under this zero flow guess).

[0176] ( Result: due to the zero flow guess, this predicted theoretical heat response will be perfectly symmetric).

[0177] Step d-3: Calculate the misfit: This is the key comparison step. The algorithm compares the predicted heat response from step d-2 (that symmetric curve) with input 3 (the actually acquired data) (that highly asymmetric curve) point by point.

[0178] The algorithm calculates a total misfit value (or objective function value) that quantifies the difference between the two. At the first iteration, since the guess (zero flow) is far from reality (there is flow), this misfit value will be very large.

[0179] Step d-4: Iterative optimization and back propagation (correct the cause): This is the smart core of the algorithm. The algorithm analyzes the specific form of this misfit.

[0180] The actually acquired data (input 3) has a peak on the north side (number 45) that is 15 hours earlier and 30% stronger than predicted (step d-2). And the actually acquired data has a peak on the south side (number 65) that is 10 hours later and 50% weaker than predicted. The algorithm calculates the residual of the observed data and the predicted data. When the observed data has a peak time earlier and a higher amplitude than the predicted value (such as the north side sensor), the algorithm corrects the flow velocity vector parameter of the corresponding grid node (for example, increases the velocity component pointing to that side) to minimize the objective function misfit value. The algorithm automatically corrects its internal guess vector field (i.e. the values of the one million nodes) accordingly, for example, by deflecting all node vectors northward by a small angle and assigning a small velocity (e.g. 0.01 meters / day) to form guess #1.

[0181] Step d-5: Loop and convergence judgment: The algorithm discards the previous calculation result and returns to step d-2. It uses guess #1 to re-execute the forward calculation, re-calculates the misfit value in step d-3 (at this time, since guess #2 is closer to reality than guess #1, the new misfit value will be smaller), and further optimizes its guess in step d-4 to generate guess #3.

[0182] The processor of the HLC will automatically repeat this iteration loop (d-2→d-3→d-4) thousands or tens of thousands of times.

[0183] Stopping of the algorithm (i.e. convergence): When the mismatch value calculated by the algorithm in step d-3 is lower than a pre-set convergence threshold of the HLC (e.g. lower than 0.1%), the iteration stops. At this moment, it means that the thermal response curve predicted by the guess vector field inside the algorithm (e.g. guess #10500) has almost perfectly matched the actually collected thermal response curve (input 3).

[0184] The final, optimal guess vector field (guess #10500) stored inside the algorithm at the time of iteration stopping, is the final result of this inversion (i.e. that unknown cause).

[0185] The HLC exports and stores this final, solved result as a standard data file, which is a quantitative, 3D seepage vector field map.

[0186] This data map (e.g. a data table) will precisely describe, for example, that the groundwater seepage velocity at the 3D coordinate (X=10.5m, Y=5.3m, Z=30.0m) is (Vx=0.12, Vy=0.02, Vz=0.00) m / day.

[0187] It describes the complete flow pattern underneath the entire ground heat exchanger array (in X, Y, Z dimensions), not just the flow at a single point.

[0188] Step (e): Using the generated 3D seepage vector field map, to overwrite or correct the original static parameters in the dynamic soil thermal model that were used to characterize the groundwater seepage around the ground heat exchanger This step is the last step of the entire calibration sub-program, i.e. the closed-loop feedback.

[0189] Accessing the model library: The HLC accesses the parameter database of the dynamic soil thermal model as described in embodiment three.

[0190] Locating the static parameters: The HLC locates in the database, those static parameters (i.e. the obsolete average values) that were set in the year 0 of system deployment to characterize the groundwater seepage.

[0191] Performing the overwriting or correction: The HLC performs an overwriting operation. It discards or archives (as historical record) that obsolete, static average parameters.

[0192] Write new parameter: HLC writes the newly generated, measured, high-resolution, quantitative, three-dimensional seepage vector field map in step (d) into the parameter database as the only, up-to-date seepage field basis for the dynamic soil heat model in the future year (or until the next calibration).

[0193] At this point, the model calibration subprogram is complete. The dynamic soil heat model on which the HLC is based is no longer the static model from year 0, but an adaptive, high-precision model that has been calibrated to physical reality.

[0194] Example five: system control method The control device of the present application executes two core algorithms to achieve optimal operation of the system: 1. Pattern matching algorithm This algorithm is the high-level control logic. The controller switches between various operating modes of the system (such as the heat storage, direct supply, two-stage heating modes shown in Tables 1 to 10) according to the input external parameters such as outdoor temperature and heating time (e.g. whether it is in the heating season).

[0195] 2. Linkage control algorithm (energy consumption comparison algorithm) This algorithm is the core control logic of the present application, which is used to determine when to switch between the ground source heat pump main supply working condition (such as Table 3, first heating loop) and the air source heat pump combined heating working condition (such as Table 4, second heating loop).

[0196] The traditional switching logic is based on a fixed buried pipe outlet water temperature threshold, which cannot guarantee the economy of operation. When the buried pipe outlet water temperature decreases, the energy consumption (compressor power consumption) of the ground source heat pump 10 will increase.

[0197] The algorithm of the present application (see Figure 3 ): The present application does not use a fixed temperature threshold, but uses an energy consumption comparison logic.

[0198] Step S301: The system operates in the first heating mode (e.g. Table 3 mode, first heating loop).

[0199] Step S302: The control device monitors or calculates the first system energy consumption in real time. The energy consumption is the total power consumption of the system in the first heating mode, including at least the compressor power consumption of the ground source heat pump 10 and the power consumption of the circulating water pump B.

[0200] Step S303: The control device calculates the second system energy consumption required if switching to the second heating mode (e.g. Table 4 mode, second heating loop) in real time according to the current working condition (e.g. outdoor temperature, ASHP performance curve). The energy consumption is the overall power consumption of the system in the second heating mode, at least including the power consumption of the air source heat pump 8, the compressor power consumption of the ground source heat pump 10 (whose evaporating side temperature has been raised by the ASHP at this time) and the circulating water pump A.

[0201] Step S304: Compare the first system energy consumption with the second system energy consumption.

[0202] Step S305: If the first system energy consumption > the second system energy consumption, it proves that starting the air source heat pump 8 to provide low-temperature heat source (secondary heating) for the ground source heat pump 10 at this time is more energy-saving than relying on the ground heat exchanger 9 alone.

[0203] Step S306: The control device performs switching, closes the ground heat exchanger outlet (closes the valves 3, 15), starts the air source heat pump 8 (opens the valves 2, 5, starts the water pump A), and the system enters the second heating mode (Table 4 mode) operation.

[0204] Step S307: If the first system energy consumption ≤ the second system energy consumption, it remains in the first heating mode (Table 3 mode) operation.

[0205] This control based on energy consumption comparison rather than temperature threshold makes the system dynamically adapt to electricity price fluctuations and equipment efficiency decay, always choosing the operation mode with the lowest economic cost, which is one of the key innovations of the present application.

[0206] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A multi-energy complementary interseasonal thermal storage and heating system, characterized in that, include: Air source heat pump; A ground source heat pump, wherein the ground source heat pump has an evaporation side and a condensation side; buried pipe; An energy station is connected to the condenser side of the ground source heat pump for heat exchange, and is used to supply heat to the heat exchange station; The first heating loop is used to connect the outlet of the buried pipe to the evaporation side of the ground source heat pump. The second heating loop is used to connect the outlet of the air source heat pump to the evaporation side of the ground source heat pump. as well as A control device configured to control switching between at least the first heating loop and the second heating loop to selectively provide a low-grade heat source to the evaporator side of the ground source heat pump; The control device is configured to: Monitor the energy consumption of the first system when the first heating loop is in operation; Monitor the energy consumption of the second system when the second heating loop is in operation; Based on the comparison between the energy consumption of the first system and the energy consumption of the second system, a switch is executed between the first heating loop and the second heating loop.

2. The system as described in claim 1, characterized in that, The system also includes a third heat source loop, which connects the outlet of the air source heat pump to the energy station. The control device is also configured to: control the third heat source loop to be turned on during periods of high outdoor temperature in the transitional season or heating season, so that the air source heat pump can directly supply heat to the energy station.

3. The system as described in claim 1, characterized in that, The system also includes a heat storage loop, which connects the outlet of the air source heat pump to the inlet of the buried pipe. The control device is also configured to: control the heat storage loop to be activated during the non-heating season so that the air source heat pump can store heat in the soil around the buried pipe.

4. The system as described in claim 1, characterized in that, The system also includes: Solar photovoltaic panels; The solar photovoltaic panel is electrically connected to the electrical equipment of the system and is used to power the system using solar energy.

5. The system as described in claim 4, characterized in that, The solar photovoltaic panel is a photovoltaic-thermal integrated PVT module; The system also includes: Water tank; And the first plate heat exchanger; The water tank is heat exchanged with the PVT module and is used to store the heat collected by the PVT module; the water tank is heat exchanged with the first heating loop or the second heating loop via the first plate heat exchanger and is used to preheat the low-grade heat source entering the evaporator side of the ground source heat pump during the heating season.

6. The system as described in claim 1, characterized in that, The system also includes: Industrial waste heat source; Second plate heat exchanger; The industrial waste heat source is connected to the first heating loop or the second heating loop via the second plate heat exchanger for heat exchange, and is used to preheat the low-grade heat source entering the evaporation side of the ground source heat pump.

7. The system as described in claim 6, characterized in that, The control device is also configured to: When the temperature of the industrial waste heat source is greater than the preset high temperature threshold, the industrial waste heat source is controlled to be connected in parallel with the condenser side of the ground source heat pump to jointly supply heat to the energy station. The waste heat return water from the energy station is controlled to enter the second plate heat exchanger to preheat the low-grade heat source.

8. The system as described in claim 1, characterized in that, The system also includes: A steam heat pump having an evaporation side and a condensation side; The evaporator side of the steam heat pump is connected to the energy station for heat exchange. The condenser side of the steam heat pump is used to supply steam to steam users.

9. The system of claim 1, wherein the control device is configured to execute a hierarchical control strategy, the strategy comprising a supervisory controller and a field controller; The supervisory controller is configured as follows: Based on long-term timescales and dynamic soil thermal models, the cross-seasonal thermal balance of the buried pipe is optimized. Based on the optimization results, a long-term operating constraint for the first heating loop is determined and output to the field controller; The field controller is configured as follows: Obtain predicted electricity price data and predicted heat load data within the short-term forecast time domain; Under the premise of strictly meeting the long-term operating constraints output by the supervisory controller; Optimize a short-run multi-objective cost function using a model predictive control algorithm. This function should include at least the following: Future operating electricity costs calculated based on the predicted electricity price data; and A preset depreciation cost associated with the switching action between the first heating loop and the second heating loop; Based on the results of the short-term multi-objective cost function optimization, a switch is performed between the first heating loop and the second heating loop to minimize the short-term multi-objective cost function.

10. A control method for a multi-energy complementary interseasonal thermal storage and heating system, the system comprising a ground source heat pump, a first heating loop and a second heating loop connected to the evaporator side of the ground source heat pump, wherein the first heating loop utilizes buried pipes for heating, and the second heating loop utilizes an air source heat pump for heating, the method being characterized in that... include: In the first heating mode, heat is supplied to the evaporator side of the ground source heat pump through the first heating loop; Monitor the energy consumption of the first system under the first heating mode; In the second heating mode, heat is supplied to the evaporator side of the ground source heat pump through the second heating loop; Monitor the energy consumption of the second system under the second heating mode; Based on the comparison results of the energy consumption of the first system and the energy consumption of the second system, the system switches between the first heating mode and the second heating mode.

Images