Distributed compressed air energy storage system based on abandoned mine and control method

Abstract

The invention provides a distributed compressed air energy storage system based on an abandoned mine. The distributed compressed air energy storage system comprises an air storage and heat management unit, a compression and expansion power unit, an intelligent detection and control unit and a multi-scene energy interface unit. The air storage and heat management unit is used for achieving efficient storage of air and recovery, storage and dynamic distribution of heat energy in the compression or expansion process. The compression and expansion power unit is used for completing air compression energy storage and expansion energy release, and conversion between mechanical energy and electric energy is achieved. The intelligent detection and control unit is used for collecting system parameters in real time, constructing a digital twin model and generating a control strategy; and the multi-scene energy interface unit is used for dynamically adjusting output pressure, flow, voltage grade and temperature according to different scene requirements. The system provided by the invention solves the problem that an existing system cannot realize adaptive load change intelligent regulation and control. The invention further provides a method applied to the system.

Description

1.1 Technical Field This invention relates to the field of novel energy storage technology, and in particular to a distributed compressed air energy storage system and control method based on abandoned mines. 1.2 Background Technology Distributed compressed air energy storage systems (CASS) are an important branch of emerging energy storage technology. Their core lies in integrating air compression, energy storage, energy release, and energy conversion functions through modular design. In terms of development, CASS technology, with its advantages of large capacity, long lifespan, and environmental friendliness, has become one of the key technologies for addressing the large-scale grid connection of renewable energy. However, existing distributed CASS systems are still largely in a stage of development characterized by "large scale but lack of precision," with significant room for improvement in energy efficiency and control accuracy.

[0003] Existing distributed compressed air energy storage systems suffer from significant shortcomings in intelligent control, making them ill-suited to the dynamic demands of distributed energy scenarios. Current technologies largely employ traditional threshold-triggered control strategies, relying solely on the compliance of single or a few parameters (such as sludge concentration, oxygen concentration, and fixed thresholds for gas storage pressure) to initiate start-up, shutdown, or parameter adjustments. This approach lacks comprehensive consideration of the system's highly coupled multi-parameter characteristics and the dynamic changes in external loads. For instance, in grid peak-shaving scenarios, when loads fluctuate rapidly, traditional control systems cannot adjust core parameters such as compressor speed and expander nozzle angle in real time, resulting in a low match between system output power and load demand. In industrial aeration scenarios, fixed pressure and flow control modes cannot flexibly adjust to dynamic changes in dissolved oxygen concentration in the aeration tank, leading not only to energy waste but also potentially impacting wastewater treatment efficiency.

[0004] To address the problem that distributed compressed air energy storage systems cannot achieve intelligent control based on adaptive load changes, this invention provides a distributed compressed air energy storage system and control method based on abandoned mines. 1.3 Summary of the Invention To address the problem that distributed compressed air energy storage systems cannot achieve intelligent control that adapts to load changes, this invention provides a distributed compressed air energy storage system based on abandoned mines, including an air storage and heat management unit, a compression and expansion power unit, an intelligent detection and control unit, and a multi-scenario energy interface unit.

[0006] The gas storage and heat management unit is used to achieve efficient air storage and the recovery, storage, and dynamic distribution of heat energy during compression or expansion. The compression and expansion power unit is used to complete air compression energy storage and expansion energy release, realizing the conversion of mechanical energy into electrical energy. The intelligent detection and control unit is connected to the gas storage and heat management unit, the compression and expansion power unit, and the multi-scenario energy interface unit, respectively, and is used to collect system parameters in real time, build a digital twin model, and generate control strategies. The intelligent detection and control unit monitors the energy storage system through a distributed fiber optic sensor network, a micro-sensor array, and a laser system analysis system, and controls the system through a digital twin system and an execution and decision-making system. The multi-scenario energy interface unit is used to dynamically adjust the output pressure, flow rate, voltage level, and temperature according to different scenario requirements.

[0007] The system provided by this invention addresses the core problem of existing systems' inability to adapt to load changes. It uses an intelligent detection and control unit as its core, linking other units to construct a closed loop of perception, decision-making, and execution. First, load-related data is acquired through a distributed fiber optic sensor network, a micro-sensor array, and a laser gas analysis system within the intelligent detection and control unit. Then, relying on a digital twin system, combined with LSTM neural networks, deep reinforcement learning algorithms, and model predictive control, decisions are generated. Finally, the gas storage and heat management unit, and the compression and expansion power unit are simultaneously adjusted to match load demand. The system provided by this invention breaks through the limitations of traditional passive threshold control, predicting load fluctuations in advance; it solves the problem of multi-parameter coupling, achieving multi-unit collaborative control; and it supports adaptive switching across five scenarios, completely solving the challenge of intelligent control adapting to load changes. Optionally, the gas storage and heat management unit includes a gas storage unit and a heat management unit. The gas storage unit includes a composite gas storage tank body, a sealing and connection system, and a pressure regulating device. The composite gas storage tank body is provided with an outer rigid support, a middle insulation layer, and an inner flexible buffer bladder from the outside to the inside.

[0008] Optionally, the inner flexible buffer bladder is an nitrile rubber-aramid fiber composite membrane with a thickness of 5 mm, wherein the nitrile rubber layer has a thickness of 3 mm and the aramid braided layer has a thickness of 2 mm.

[0009] Optionally, the thermal management unit includes a thermal storage subsystem, a cold storage subsystem, and an intelligent heat distribution system. The thermal storage subsystem, the cold storage subsystem, and the intelligent heat distribution system are interconnected. The intelligent heat distribution system is used to detect and dynamically adjust the temperature of the thermal storage subsystem and the cold storage subsystem. The thermal storage subsystem includes a phase change thermal storage material, which is a sodium nitrate-potassium nitrate mixed salt, wherein the mass ratio of sodium nitrate to potassium nitrate in the mixed salt is 60:40.

[0010] Optionally, the cold storage subsystem includes a cold storage medium, which is a 40% ethylene glycol aqueous solution.

[0011] Optionally, the compression and expansion power unit includes a compressor and an expander unit, the compressor and the expander unit are connected, the compressor includes a magnetic levitation compressor, and the expander unit includes a turbine expander.

[0012] Optionally, the distributed optical fiber sensing network includes armored optical fibers, which are spirally wound along the outer wall of the gas storage tank and arranged axially along the compressor and expander unit axis; the micro-sensor array includes a pressure sensor array, a flow sensor array, and a temperature sensor array; the laser gas analysis system includes a tunable semiconductor laser absorption spectrometer, which is installed in the gas storage tank outlet and expander inlet pipe.

[0013] Optionally, the multi-scenario energy interface unit may include a pneumatic interface subsystem, an electric interface subsystem, a waste heat recovery and utilization interface subsystem, and a scenario collaborative control subsystem, depending on the different scenario requirements.

[0014] The present invention also provides a control method applicable to the distributed compressed air energy storage system based on abandoned mines as described in any of the above claims.

[0015] Optionally, the method further includes: collecting key system operating parameters through a distributed fiber optic sensor network, a micro-sensor array, and a laser gas analysis system in the intelligent detection and control unit to obtain real-time operating data covering the gas storage and thermal management unit, the compression and expansion power unit, and the multi-scenario energy interface unit; constructing a digital twin in the intelligent detection and control unit based on the real-time operating data, integrating physical models and data-driven algorithms to generate control strategies adapted to multiple scenarios; and synchronously adjusting the energy distribution status of the gas storage and thermal management unit and the operating parameters of the compression and expansion power unit according to the control strategy.

[0016] This invention provides a distributed compressed air energy storage system based on abandoned mines, comprising an air storage and heat management unit, a compression and expansion power unit, an intelligent detection and control unit, and a multi-scenario energy interface unit. The air storage and heat management unit is used to achieve efficient air storage and the recovery, storage, and dynamic distribution of heat energy during compression or expansion. The compression and expansion power unit is used to complete air compression energy storage and expansion energy release, realizing the conversion of mechanical energy into electrical energy. The intelligent detection and control unit is used to collect system parameters in real time, construct a digital twin model, and generate control strategies. The multi-scenario energy interface unit is used to dynamically adjust the output pressure, flow rate, voltage level, and temperature according to different scenario requirements. The system provided by this invention solves the problem that existing systems cannot achieve intelligent control based on adaptive load changes. This invention also provides a method applied to the above system. 1.4 Description of the attached figures To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of the control method for a distributed compressed air energy storage system based on abandoned mines. 1.5 Detailed Implementation The embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described below do not represent all embodiments consistent with the present invention. They are merely examples of systems and methods consistent with some aspects of the invention as detailed in the claims.

[0020] To address the problem that distributed compressed air energy storage systems cannot achieve intelligent control that adapts to load changes, this invention provides a distributed compressed air energy storage system based on abandoned mines, including an air storage and heat management unit, a compression and expansion power unit, an intelligent detection and control unit, and a multi-scenario energy interface unit.

[0021] The gas storage and heat management unit is used to achieve efficient air storage and the recovery, storage, and dynamic distribution of heat energy during compression or expansion. The compression and expansion power unit is used to complete air compression energy storage and expansion energy release, realizing the conversion of mechanical energy into electrical energy. The intelligent detection and control unit is connected to the gas storage and heat management unit, the compression and expansion power unit, and the multi-scenario energy interface unit, respectively, and is used to collect system parameters in real time, build a digital twin model, and generate control strategies. The intelligent detection and control unit monitors the energy storage system through a distributed fiber optic sensor network, a micro-sensor array, and a laser system analysis system, and controls the system through a digital twin system and an execution and decision-making system. The multi-scenario energy interface unit is used to dynamically adjust the output pressure, flow rate, voltage level, and temperature according to different scenario requirements.

[0022] This system is a modular compressed air energy storage and multi-scenario energy coupling system. Its core is a four-level linkage of "compression-heat storage-energy release-intelligent regulation," achieving deep integration with scenarios such as industrial aeration, grid peak shaving, and underwater energy replenishment. The overall architecture comprises four core units: a gas storage and heat management unit, a compression and expansion power unit, an intelligent detection and control unit, and a multi-scenario energy interface unit. These units are interconnected through pressure-resistant pipelines, data buses, and thermal circulation pipelines, forming a closed-loop system. The system provided by this invention addresses the core problem of existing systems' inability to adapt to load changes. It uses the intelligent detection and control unit as its core, linking other units to construct a closed loop of perception, decision-making, and execution. First, load-related data is acquired through the distributed fiber optic sensor network, micro-sensor array, and laser gas analysis system of the intelligent detection and control unit. Then, relying on a digital twin system, combined with LSTM neural networks, deep reinforcement learning algorithms, and model predictive control, decisions are generated. Finally, the gas storage and heat management unit and the compression and expansion power unit are simultaneously adjusted to match load demands. The system provided by this invention breaks through the traditional passive threshold control, predicts load fluctuations in advance, solves the problem of multi-parameter coupling, realizes multi-unit collaborative regulation, and supports adaptive switching of 5 scenarios, completely solving the problem of intelligent regulation of adaptive load changes.

[0023] In some embodiments, the gas storage and heat management unit includes a gas storage unit and a heat management unit. The gas storage unit includes a composite gas storage tank body, a sealing and connection system, and a pressure regulating device. The composite gas storage tank body is provided with an outer rigid support, a middle heat insulation layer, and an inner flexible buffer bladder from the outside to the inside.

[0024] In some embodiments, the inner flexible buffer bladder is an nitrile rubber-aramid fiber composite membrane with a thickness of 5 mm, wherein the nitrile rubber layer has a thickness of 3 mm and the aramid braided layer has a thickness of 2 mm.

[0025] In some embodiments, the gas storage unit achieves efficient air storage within a pressure range of 0.5-3MPa, while also possessing pressure buffering, corrosion protection, and dynamic volume adjustment capabilities, adapting to the needs of multiple scenarios such as industrial aeration (low pressure) and power grid peak shaving (high pressure).

[0026] The gas storage unit consists of a composite gas storage tank body, a sealing and connection system, and a pressure regulating device. The composite gas storage tank body, from the outside in, consists of an outer rigid support, a middle insulation layer, and an inner flexible buffer layer. The outer rigid support is forged from 316L stainless steel (yield strength ≥205MPa, elongation ≥40%), with wall thickness designed according to pressure gradients (8mm for 0.5MPa, 15mm for 1-2MPa, and 25mm for 2-3MPa). The inner wall is electrolytically polished (Ra≤0.8μm) to reduce gas adsorption loss. The middle insulation layer is filled with nano-aerogel felt (thermal conductivity ≤0.018W / (m²)). The outer layer is made of nitrile rubber and aramid fiber composite film (5mm thick, including a 3mm rubber layer and a 2mm aramid braided layer), with a 50mm thick outer layer and an aluminum foil reflective layer (reflectivity ≥95%) to reduce heat exchange with the outside environment. The inner flexible buffer bladder is made of nitrile rubber-aramid fiber composite film (5mm thick, including a 3mm rubber layer and a 2mm aramid braided layer), with a tensile strength ≥25MPa and an ozone aging resistance (100pphm, 40℃) ≥1000 hours, and can achieve ±30% dynamic volume adjustment.

[0027] The sealing and connection system is specifically designed with tongue and groove flanges at the tank opening, and a fluororubber O-ring (hardness 70 Shore A, operating temperature -20~200℃) embedded in the sealing surface. This, combined with a metal spiral wound gasket (material 316L + graphite), achieves a double seal with a leakage rate ≤1×10⁻⁶. -7 Pa m 3 / s. The intake / exhaust pipe adopts a double compression fitting (material Hastelloy C276), with a pipe diameter of Φ150mm and a design flow velocity of 15-20m / s to avoid turbulent noise (≤85dB).

[0028] The pressure regulating device includes a main pressure sensor and a safety relief valve. The main pressure sensor is a silicon piezoresistive transmitter (model PX409-3MPaG, accuracy ±0.1%FS, response time ≤1ms), located in the gas phase zone at the bottom of the tank. The safety relief valve has two levels of protection: the first is a pilot-operated safety valve (opening pressure 3.2MPa, reseating pressure 2.9MPa), and the second is a rupture disc (burst pressure 3.5MPa, nickel alloy material), ensuring relief within 100ms in case of overpressure. The pressure is released through a nitrogen storage tank (1m³). 3 The pressure (2MPa) is linked with an electromagnetic regulating valve (Φ10mm diameter, ±0.01MPa adjustment accuracy) to maintain the pressure between the flexible bladder and the rigid body at a stable 0.1±0.02MPa. The nitrogen storage tank is equipped with a variable frequency gas supply pump (flow rate 0-2m³ / h). 3 ( / h), the pressure sensor provides real-time air replenishment / exhaust to ensure stable pressure in the buffer chamber.

[0029] The operation of the gas storage and heat management unit includes three stages: energy storage, energy release, and dynamic regulation. Energy storage stage: High-pressure air, after drying and purification (dew point ≤ -40℃), enters the flexible buffer bladder. The bladder expands, compressing nitrogen in the buffer chamber. Pressure sensors provide real-time feedback. When the set pressure (e.g., 3MPa) is reached, the inlet valve (pneumatic ball valve, Φ150mm diameter, opening / closing time ≤ 2s) closes. Energy release stage: Based on downstream demand (e.g., 0.2MPa for industrial aeration), the exhaust rate is controlled by a frequency converter valve (adjustment range 0-100%). The flexible bladder contracts, and the nitrogen pressure in the buffer chamber decreases synchronously, preventing the rigid inner wall from bearing negative pressure. Dynamic regulation: When a pressure fluctuation exceeding ±0.05MPa is detected, an auxiliary air supply / exhaust pump (flow rate 5m³ / h) is activated. Combined with buffer chamber pressure compensation, the instantaneous pressure fluctuation is controlled within ±0.02MPa.

[0030] In some embodiments, the thermal management unit includes a thermal storage subsystem, a cold storage subsystem, and an intelligent heat distribution system. The thermal storage subsystem, the cold storage subsystem, and the intelligent heat distribution system are interconnected. The intelligent heat distribution system is used to detect and dynamically adjust the temperature of the thermal storage subsystem and the cold storage subsystem. The thermal storage subsystem includes a phase change thermal storage material, which is a sodium nitrate-potassium nitrate mixed salt, wherein the mass ratio of sodium nitrate to potassium nitrate in the mixed salt is 60:40.

[0031] In some embodiments, the cold storage subsystem includes a cold storage medium, which is a 40% aqueous solution of ethylene glycol.

[0032] In some embodiments, the core function of the thermal management unit is to efficiently recover the sensible and latent heat (150-250°C) of the compression process, while capturing the cold energy (-5 to 10°C) of the expansion process, and to achieve long-term energy storage through phase change materials, with a heat loss rate of ≤2% / day.

[0033] The thermal storage subsystem includes a phase change material (PCM) and a heat exchanger. The PCM is a sodium nitrate-potassium nitrate mixed salt (mass ratio 60:40), with a phase change temperature of 220±5℃ and a latent heat of phase change of 200±10 kJ / kg. The addition of 5% expanded graphite (50 μm particle size) increases the thermal conductivity to 2.5 W / (m²). K) (The thermal conductivity of pure salt is 0.6 W / (m) The encapsulation uses an aluminum alloy honeycomb structure (cell size 50×50mm, wall thickness 0.5mm), each cell is filled with PCM and sealed, and the whole assembly is placed in a 316L stainless steel storage tank (volume 50m³). 3 (Design pressure 1.0 MPa).

[0034] The heat exchange unit employs a three-stage shell-and-tube heat exchanger, with an inner tube of Φ50mm (316L, 3mm wall thickness) and an outer tube of Φ80mm (304 stainless steel). Each stage is 6m long, with a total heat exchange area of ​​45m². The heat transfer oil (L-QC320) flows at a velocity of 1.5m / s, with the inlet and outlet temperature difference controlled within 30℃. The single-stage heat exchange efficiency is ≥85%. A four-way reversing valve (working pressure 1.6MPa, switching time ≤5s) is used to ensure that the heat transfer oil flows from the compressor to the PCM storage tank during heat storage and flows in the opposite direction during heat release, preheating the air at the expander inlet. An overpressure protection valve (opening pressure 0.8MPa) is added between the heat exchanger and the storage tank to prevent pressure from entering the heat storage system.

[0035] The cold storage subsystem uses a 40% ethylene glycol aqueous solution as the cold storage medium (freezing point -25℃, specific heat capacity 3.5kJ / (kg)). K), stored in a polyethylene-lined storage tank (30m³). 3 The exterior is encased in a 100mm thick polyurethane insulation layer. Cold energy recovery takes place in a spiral plate heat exchanger with 304 stainless steel plates (1mm thick), 5mm channel spacing, and a total heat exchange area of ​​20m². 2 The expander exhaust (-5 to 10℃) exchanges heat with the ethylene glycol solution in a countercurrent manner, and the cold energy recovery rate is ≥70%.

[0036] The intelligent heat distribution system achieves precise control of heat flow through temperature monitoring and dynamic adjustment logic. In the temperature monitoring stage, 16 platinum resistance thermometers (Pt100, accuracy ±0.1℃) are installed inside the PCM storage tank to monitor the temperature field distribution of the phase change material in real time, thereby accurately determining the current heat storage or release state of the system. In the dynamic adjustment logic, when the compression heat temperature exceeds 250℃, the bypass cooler (water-cooled, 50kW heat exchange capacity) is immediately activated to prevent the phase change heat storage material (PCM) from decomposing due to overheating. After entering the energy release stage, based on the set temperature requirement of 150℃ at the expander inlet, the flow rate of the heat transfer oil is adjusted via a variable frequency pump (0-50m³ / h) to stabilize the temperature control accuracy within ±3℃. Simultaneously, the cold energy captured by the system is preferentially supplied to the compressor cooling system (which can reduce the compressor inlet air temperature by 10℃, corresponding to a 3% increase in compression efficiency). If there is surplus cold energy, it is used to cool the industrial workshop through a plate heat exchanger.

[0037] In some embodiments, the compression and expansion power unit includes a compressor and an expander unit, the compressor and the expander unit being connected, the compressor including a magnetic levitation compressor, and the expander unit including a turbine expander.

[0038] The compression and expansion power unit breaks through the bottlenecks of traditional compressed air energy storage, such as "large mechanical friction loss, low efficiency under varying operating conditions, and weak energy coupling". Through magnetic levitation contactless transmission, transcritical energy coupling, and adaptive power regulation technology, it achieves a compression efficiency of ≥88% and an expansion efficiency of ≥90% (rated operating conditions), with an efficiency fluctuation of ≤5% within the 30%-110% load range.

[0039] The compressor employs a three-stage magnetic levitation centrifugal compressor in series and supercritical CO2 waste heat recompression technology to replace the traditional screw compressor, eliminating mechanical friction losses (reducing the friction coefficient to below 0.0001) and simultaneously achieving cascade recovery of compression heat. Its main body uses a single-stage centrifugal (300mm impeller diameter) three-stage series design with a total pressure ratio of 27 (first stage pressure ratio 3, second stage pressure ratio 3, final stage pressure ratio 3) to meet gas storage requirements ranging from 0.1MPa to 3MPa. The radial bearing in the magnetic levitation bearing system adopts active magnetic levitation control (8-pole electromagnet, control accuracy ±5μm), with a load capacity of 2000N and a response frequency of 1kHz. It can compensate for impeller dynamic balance deviation (≤0.02mm / s) in real time. The axial bearing thrust disk has a diameter of 150mm, a magnetic gap of 0.5mm, and a maximum axial force of 5000N. It maintains axial position stability through PID closed-loop control. The impeller is made of TC4 titanium alloy integral forging (density 4.5g / cm³, tensile strength ≥900MPa), formed by five-axis machining, and the blade profile is a three-dimensional flow design (inlet angle 18°, outlet angle 45°, aerodynamic efficiency ≥92%). The casing is a 316L stainless steel welded structure, and the inner wall is sprayed with a ceramic wear-resistant layer (thickness 0.2mm, hardness HV1200), which can withstand slight wear from air dust content ≤1mg / m³.

[0040] In the compressor's intermediate cooling and heat recovery system, a microchannel heat exchanger is connected in series at the compressor outlet of each stage. Its core is an aluminum alloy flat tube with a cross-section of 20mm × 2mm and an internal fin spacing of 1mm, achieving a single-stage heat exchange area of ​​20m². 2 The cooling medium is a 40% ethylene glycol aqueous solution from the cold storage unit, which can reduce the compressed air temperature from 180-250℃ to 40±5℃, thereby reducing the power consumption of the next stage of compression and achieving an energy saving effect of 15%-20% and interstage cooling. Simultaneously, a channel spacing of 2mm and a heat exchange area of ​​50m² are provided at the outlet of the final stage compressor. 2 The plate-fin heat exchanger can transfer the heat of compression at 250°C to supercritical CO2 at a working pressure of 8MPa, raising the CO2 temperature from 100°C to 220°C to form a recompression loop. The high-temperature CO2 can also drive an auxiliary turbine with a power of 500kW, which drives a booster compressor with a pressure ratio of 1.5 to boost some of the low-pressure CO2 and send it back to the heat source side, ultimately improving the system thermal efficiency by 8%-10%.

[0041] In terms of variable frequency drive and control, a permanent magnet synchronous motor with a rated power of 10MW, a speed of 15000r / min, and an efficiency of 98.5% is adopted, paired with a three-level frequency converter with a switching frequency of 5kHz. This enables continuous speed adjustment from 10% to 100%, with a response time of ≤50ms to meet the rapid peak-shaving requirements of the power grid. Anti-surge control is achieved through the linkage between the inlet guide vane (adjustment angle -30° to +15°) and the outlet bypass valve. When the flow rate is detected to be 10% lower than the surge line, the bypass valve can be opened within 100ms (opening degree 0-50%) to prevent the unit vibration intensity from exceeding 2.8mm / s.

[0042] The expander unit adopts an axial turbine expander and phase change thermal storage and reheat technology to replace the traditional axial flow expander. Through variable nozzle and blade coating technology, it achieves high-efficiency output under a wide range of operating conditions. The turbine expander body is a two-stage axial turbine series structure. The first stage has an inlet pressure of 3MPa and an outlet pressure of 1MPa, and the second stage has an inlet pressure of 0.9MPa and an outlet pressure of 0.12MPa, with a total expansion ratio of 25. The nozzle ring is made of Inconel 718 alloy with a temperature resistance of 650℃. The nozzle blades are adjustable with an angle range of 10°-30°. The flow area is adjusted in real time by a servo motor with a positioning accuracy of ±0.1° to adapt to flow rate changes of 30%-110% of the rated value. The turbine blades are made of single-crystal high-temperature alloy CMSX-4, with a 50μm thick MCrAlY coating that resists oxidation at 1100℃. The blade tip clearance is controlled to ≤0.1mm by magnetorheological sealing to reduce air leakage. The bearing system uses an air bearing with a supply pressure of 0.6MPa, a friction power consumption of ≤0.5kW, and allows axial displacement of ±0.2mm to adapt to thermal deformation under varying operating conditions.

[0043] In terms of the reheat and preheating system, a phase change heat exchanger is installed between the two turbine stages. The core is a Φ25mm×2mm 316L spiral coil, wrapped with 2000kg of NaNO3-KNO3 molten salt phase change material. It can heat the expanded air from 120℃ to 200℃, increasing the secondary expansion work by 20%. The expander inlet pipe is wrapped with a heat pipe preheating jacket. The heat pipe working fluid is potassium, and the working temperature is 300℃. It can extract heat from the heat storage unit to increase the inlet air temperature from the ambient temperature of 25℃ to 180℃. This alone can increase the expansion efficiency by 12%.

[0044] The expander's reheat and preheating systems achieve efficient utilization and precise control of thermal energy through a staged design. The first-stage outlet reheat stage is located between the two turbine stages, employing a phase change heat exchanger with 316L stainless steel spiral coils as its core. The coils are wrapped with a 2000kg mass-to-weight ratio (60:40) sodium nitrate-potassium nitrate (NaNO3-KNO3) mixed salt phase change heat exchanger, which can heat the expanded air (temperature reduced to 120℃) to 200℃, increasing the secondary expansion efficiency by 20%. The inlet preheating stage involves wrapping a heat pipe preheating jacket around the expander inlet pipe. The heat pipe working fluid is potassium with an operating temperature of 300℃, extracting heat from the heat storage unit to raise the inlet air temperature from ambient temperature (25℃) to 180℃. This single step increases the expansion efficiency by 12%.

[0045] In terms of power output and regulation, the turbine shaft of the expander unit transmits power to a synchronous generator via a speed-increasing gearbox with a transmission ratio of 5:1. This synchronous generator has a rated power of 12MW and a power factor of 0.95, which can efficiently convert the mechanical energy output by the turbine into electrical energy. After the electrical energy is output, it is first compensated for by an SVG static var generator before being connected to the grid, effectively controlling the harmonic distortion rate to ≤5%, meeting the grid's requirements for power quality, and maintaining a stable output voltage of 10kV. To adapt to the load demand in different scenarios, power regulation is achieved through a dual approach: on the one hand, the nozzle angle is changed by a servo motor, with a response time of ≤200ms, which can quickly adapt to changes in flow rate; on the other hand, the outlet pressure of the gas storage tank is adjusted in conjunction with a frequency converter valve, forming a linkage control between pressure and flow rate, ultimately achieving a continuous power output of 3MW-13.2MW (corresponding to a load range of 30%-110%), accurately matching the dynamic needs of grid peak shaving, industrial aeration, and other scenarios, and meeting the grid's AGC (automatic generation control) requirements.

[0046] In some embodiments, the distributed optical fiber sensing network includes armored optical fibers, which are spirally wound along the outer wall of the gas storage tank and arranged axially along the compressor and expander unit axis; the micro-sensor array includes a pressure sensor array, a flow sensor array, and a temperature sensor array; the laser gas analysis system includes a tunable semiconductor laser absorption spectrometer, which is installed in the gas storage tank outlet and expander inlet pipes.

[0047] Addressing the pain points of compressed air energy storage systems, such as "strong coupling of multiple parameters, dynamic and varied operating conditions, and difficulty in multi-scenario coordination," the intelligent detection and control unit achieves refined management and control through a four-level architecture of "full-domain perception, digital twin, intelligent decision-making, and precise execution." It can achieve detection accuracy of ±0.01MPa for pressure, ±0.1℃ for temperature, ±0.5% for flow rate, and ±0.05% for gas concentration. It can also achieve rapid feedback from parameter acquisition to execution response within ≤100ms, and maintain a system efficiency of ≥95% under load fluctuations of 30%-110%.

[0048] Among them, the global perception system constructs a multi-dimensional detection system consisting of "fiber optic sensor network, micro sensor array, and laser spectral analysis".

[0049] The distributed optical fiber sensing network uses Φ0.9mm armored optical fiber (single-mode G.652D), spirally wound along the outer wall of the gas storage tank (50mm spacing), and arranged axially along the compressor unit and expander unit (one measuring point every 100mm), with a total length of 1500m. It covers the temperature and strain fields of key equipment. Based on Brillouin optical time domain reflectometry (BOTDR) technology, it simultaneously acquires temperature and strain data by analyzing the frequency shift changes of Brillouin scattered light in the optical fiber. The temperature range is -50℃ to 300℃, with a spatial resolution of 1m and an accuracy of ±0.5℃. The strain range is -2000με to 2000με and an accuracy of ±5με. It can identify minute deformations of ≥5με on the gas storage tank wall caused by pressure fluctuations and provide early warning of structural fatigue risks.

[0050] The miniature sensor array encompasses three monitoring modules: pressure, flow, and temperature, including pressure sensor arrays, flow sensor arrays, and temperature sensor arrays. For pressure monitoring, silicon piezoresistive pressure transmitters (model PX8000, range 0-4MPa, accuracy ±0.1%) are installed at the top, bottom, and inside / outside the flexible bladder of the gas storage unit. The system uses a FS (Flat Change Flow) sensor with a sampling frequency of 100Hz. It calculates the pressure uniformity within the tank (deviation ≤ 0.05MPa) through multi-point data fusion. Simultaneously, high-frequency dynamic pressure sensors (model KuliteXCQ-062-100A, response frequency 10kHz) are installed at key pipeline nodes such as the compressor outlet and expander inlet to capture transient pressure pulsations such as surge precursors. For flow monitoring, an electromagnetic flowmeter (DN200, range 0-500m³ / h, accuracy ±0.5%) is used on the main pipeline, and a Coriolis mass flowmeter (DN50, accuracy ±0.1%) is used on the bypass pipeline to achieve synchronous measurement of total and branch flow. For temperature monitoring, 16 platinum resistance thermometers (Pt1000, accuracy ±0.1℃) are arranged inside the phase change thermal storage tank, and thermocouples (K-type, accuracy ±1℃) are installed in the compressor oil circuit and expander cooling water circuit to construct a three-dimensional temperature field model.

[0051] The core equipment of the laser gas analysis system is a tunable semiconductor laser absorption spectroscopy (TDLAS) analyzer (model SICKGM700), installed at the outlet of the gas storage tank and the inlet pipe of the expander. It can detect the concentration of O2, CO2, and H2O, with a response time ≤1s, resolution 0.01%, and drift rate ≤0.1% / month. It can monitor the air dryness (dew point ≤-40℃) and the oxygen concentration in industrial aeration scenarios in real time. It is also equipped with dual optical path compensation (measurement light plus reference light) to eliminate the influence of dust and vibration on laser transmission, and maintains detection accuracy even when the dust content in the pipeline is ≤10mg / m³.

[0052] As an intelligent control layer, the digital twin decision system constructs a 1:1 accurate digital twin of the system, integrates physical models and data-driven algorithms, and realizes closed-loop control of "prediction-optimization-execution-feedback", which is different from traditional experience-based control strategies.

[0053] In the digital twin modeling stage, geometric modeling uses 3D laser scanning technology to obtain the physical dimensions of the equipment (error ≤ 0.1mm). A virtual model containing 128 components, including the gas tank, compressor, and expander, is built on the Unity platform, supporting real-time rendering at 1000fps. In terms of physical modeling, a high-pressure air state equation is established based on the Redlich-Kwong equation to simulate the coupling relationship between pressure, temperature, and density (calculation error ≤ 2%). The enthalpy tracking method is introduced to simulate the phase transition process of molten salt from solid to liquid to solid (phase transition latent heat calculation deviation ≤ 5%). The dynamic stiffness of the magnetic levitation bearing is simulated through a multibody dynamics model (ADAMS software) (error ≤ 3%). Data fusion uses a Kalman filter algorithm to fuse the physical model output with sensor measured data, updating the twin state every 10ms to ensure consistency between the virtual image and the physical entity (deviation ≤ 1%).

[0054] The intelligent algorithm system comprises three main modules: a load forecasting module, an optimization decision-making module, and a real-time control module. The load forecasting module employs an attention-based LSTM neural network. The input layer contains 24-dimensional features (historical load, meteorological data, electricity price signals, etc.), and the hidden layer has three layers with 128 neurons each. The output layer predicts the net load for the next 24 hours (time resolution 15 minutes). Short-term (within 1 hour) prediction errors are ≤3%, and medium-term (within 24 hours) prediction errors are ≤5%, providing forward-looking decisions for energy storage and release scheduling. The optimization decision-making module is based on deep reinforcement learning (DRL), with the dual objectives of "maximizing system efficiency and minimizing lifetime loss" (weights are dynamically adjustable). It constructs a state space containing 28 system parameters and an action space covering 12 control variables, such as compressor speed and expander nozzle angle. Simultaneously, a composite algorithm of "cascaded PID and fuzzy logic" is designed to achieve a gas storage pressure steady-state error ≤ ±0.02 MPa, expander inlet temperature fluctuation ≤ ±2℃, and grid-connected power control tracking accuracy ≤ ±1% of the rated value.

[0055] The system also has five preset operating modes, including grid peak shaving mode, industrial aeration mode, emergency power supply mode, maintenance standby mode, and energy storage priority mode. It automatically switches to the optimal control strategy through digital twin simulation and evaluation. In grid peak shaving mode, it prioritizes tracking load forecasts, with a power response speed ≥10% of rated value / second. In industrial aeration mode, it locks the oxygen concentration within the range of 21±0.5% and stabilizes the pressure at 0.2±0.01MPa. In emergency power supply mode, it sacrifices some efficiency (allowing a reduction to 65%) to ensure continuous power output (uninterrupted switching). In maintenance standby mode, it automatically reduces the system pressure to 0.3MPa, maintaining minimal operation of core components (energy consumption ≤5% of rated value). In energy storage priority mode, when the renewable energy curtailment rate is >10%, it forces full-load energy storage, ignoring short-term efficiency fluctuations.

[0056] The execution and communication system serves as the implementation platform for the control strategy, equipped with high-precision actuators and an industrial-grade communication network. The high-precision actuators comprise valve control and unit regulation. For valve control, the main pipeline valves utilize linear-stroke electric regulating valves (DN150, adjustment range 0-100%, positioning accuracy ±0.5%, full stroke time ≤2s), equipped with an absolute encoder to provide feedback on opening. Emergency shut-off valves employ pneumatic ball valves (DN200, response time ≤50ms), featuring a redundant design (dual solenoid valve control) that automatically closes upon power failure. For unit regulation, the compressor achieves continuous speed adjustment from 0-15000 r / min via a permanent magnet synchronous motor frequency converter (capacity 12MW, carrier frequency 10kHz). The expander nozzle angle is driven by a servo motor (power 2.2kW, positioning accuracy ±0.05°), with an adjustment range of 10°-30° and a response time ≤100ms. The industrial-grade communication network adopts a hybrid architecture of "fiber optic ring network plus wireless mesh". Key nodes (controller-sensor-actuator) are connected through a gigabit fiber optic ring network. This ring network adopts a redundant design, with a self-healing time of ≤50ms after fiber breakage and a transmission latency of ≤1ms. Distributed sensors rely on LoRa wireless mesh network communication, with a communication distance of 2km, a node capacity of 500, a data refresh rate of 1Hz, and a power consumption of ≤10mA. At the same time, it uses the national cryptographic SM4 algorithm for encrypted transmission, updates the session key every 10 minutes, and sets up a data verification mechanism (CRC32 + timestamp) to prevent tampering and replay attacks and ensure data security.

[0057] To ensure long-term stable operation of the system, the intelligent detection and control unit has also constructed a "three-level defense + self-healing control" safety and fault tolerance system. Through multi-dimensional data fusion, it achieves early warning, precise location, and automatic handling of faults, increasing the system's mean time between failures (MTBF) to over 10,000 hours. The first-level early warning is based on anomaly detection using a digital twin. When the deviation between the virtual model and measured data exceeds 3% (e.g., a deviation of over 0.06 MPa in the predicted pressure of the gas storage tank), an early warning is triggered, identifying potential faults 30 seconds to 5 minutes in advance. The second-level diagnosis uses a random forest algorithm, inputting 28-dimensional features to analyze the sensors... The system achieves an accuracy rate of ≥98% in identifying 23 typical fault types, including drift, abnormal bearing noise, and valve jamming, with positioning precision down to the specific component level. A three-tiered response system employs differentiated measures for different fault severity levels: automatic adjustment of control parameters for minor faults (e.g., activating a data compensation algorithm when a sensor drifts); and triggering safety interlocks for severe faults (e.g., opening the bypass valve within 100ms when a compressor surges), while simultaneously activating a backup circuit (e.g., in a dual-redundant expander unit). The self-healing control function allows the system to automatically return to its optimal operating state after fault resolution via a "memory-learning-recovery" mechanism, requiring no manual intervention (recovery time ≤5min).

[0058] In some embodiments, the multi-scenario energy interface unit includes a pneumatic interface subsystem, an electric interface subsystem, a waste heat recovery and utilization interface subsystem, and a scenario collaborative control subsystem, depending on different scenario requirements.

[0059] The multi-scenario energy interface unit serves as the "energy conversion hub" between the compressed air energy storage system and external application scenarios. Through modular design, it enables the output of multiple forms of energy, including high-pressure air, electricity, and heat. It is adaptable to diverse scenarios such as industrial aeration, power grid peak shaving, and underwater operations. Its core functions include converting the high-pressure air energy stored in the system into directly usable low-pressure air energy, electricity, and waste heat. It dynamically adjusts the output pressure (0.1-3MPa), flow rate (0-500m³ / h), voltage level (0.4kV-10kV), and temperature (room temperature-150℃) according to different scenario requirements. It also links with the intelligent detection and control unit to achieve real-time interaction between scenario demand signals and system operating status, ensuring the stability and safety of energy output.

[0060] This system provides pneumatic interface subsystems for compressed air at different pressure levels for industrial aeration and underwater power equipment, achieving precise control "from high-pressure air to suitable pressure air". Its three-stage series pressure reducing device is made of 316L stainless steel. The first stage inlet pressure is 3MPa, the second stage reduces it to 1MPa, and the final stage can be adjusted within the range of 0.1-0.5MPa according to the scenario requirements. Each pressure reducing valve is equipped with a pressure feedback sensor with an accuracy of ±0.01MPa. A PID algorithm is used to achieve pressure overshoot-free control with an adjustment time ≤2s. For industrial aeration adaptation, the final stage outlet connects to a 50mm diameter Venturi mixer, mixing 0.2MPa compressed air with ambient air at a 1:3 ratio. This mixture is then injected into the aeration tank through a microporous aeration disc with a 0.5mm orifice and an oxygen utilization rate ≥25%, and is linked to the dissolved oxygen sensor of the detection device to dynamically adjust the mixing ratio. This subsystem also employs a combination of electric ball valves and vortex flow meters (range 0-200m³ / h, accuracy ±1%) to support remote stepless adjustment with a response time ≤500ms. A precision filter with a filtration accuracy of 0.01μm and a dryer with a dew point ≤-40℃ are connected in series after the pressure reducing module to ensure that the oil content of the output air is ≤0.01mg / m³, meeting the cleanliness requirements of underwater equipment (such as submersible pneumatic motors). Its underwater-specific interface uses a deep-sea pressure-resistant quick connector made of titanium alloy TC4, with a working pressure ≤10MPa and a insertion / removal life ≥1000 times. It is equipped with seawater-resistant fluororubber O-rings, suitable for operating environments with a water depth ≤300m. It also integrates an overpressure protection valve with a starting pressure of 1.2 times the rated value and an underwater cable sealing chamber to prevent seawater backflow and electrical short circuits.

[0061] The power interface subsystem is responsible for converting the electrical energy generated by the expander unit into a form of electricity that meets the needs of the power grid or users. It supports grid-connected, off-grid, and distributed energy collaborative operation. Its grid-connected interface module's power distribution equipment includes a 10kV / 0.4kV, 15MVA prefabricated substation, an SVG static var generator with a compensation capacity of ±5Mvar, and a harmonic filter with a total harmonic distortion rate ≤5%, meeting the GB / T19963-2011 grid connection standard. It also receives grid dispatch instructions via a remote terminal unit (RTU) and supports AGC (Automatic Generation Control) and AVC (Automatic Voltage Control), with a power regulation rate ≥10%. Rated voltage per minute can meet peak shaving requirements. For off-grid and microgrid interfaces, a 5MW bidirectional DC / AC converter with an efficiency ≥97% is configured, linked with a 2MWh lithium iron phosphate battery pack. In off-grid mode, it maintains output voltage fluctuations ≤±2%, achieving energy storage and load balance. It also features a dedicated DC800V photovoltaic / wind power interface, supporting maximum power point tracking. The energy management system (EMS) prioritizes renewable energy consumption, with any shortfall supplemented by compressed air energy storage. The emergency power supply interface is equipped with an ATS dual-power automatic transfer switch with a switching time ≤50ms, automatically switching to compressed air energy storage power supply in the event of a grid failure, ensuring uninterrupted operation of critical loads such as hospitals and data centers. Simultaneously, a 2MVA dry-type transformer provides multiple output levels, including 0.4kV and 6kV, to meet the voltage requirements of different emergency loads.

[0062] The waste heat recovery and utilization interface subsystem functions to convert waste heat generated during system operation (such as waste heat from expander exhaust and surplus heat from the heat storage unit) into usable thermal energy, thereby improving the overall energy utilization rate. The industrial heat user interface uses a 316L plate heat exchanger with a heat exchange area of ​​50m², exchanging heat between the 150℃ heat transfer oil from the heat storage unit and industrial water, outputting 60-90℃ hot water for workshop heating or process heating. It also uses a temperature sensor linked to an electric regulating valve to dynamically adjust the hot water flow rate (0-100m³ / h) according to the heat user's needs (e.g., a set water temperature of 60℃), with a control accuracy of ±2℃. The domestic heat utilization interface provides heating to surrounding buildings through a floor radiant heating system with a designed supply water temperature of 50℃. A supporting heating station enables zoned temperature control of 18±1℃ indoor temperatures. It also integrates a 5m³ volumetric heat exchanger to provide 45℃ domestic hot water, with a daily supply of ≥100 tons, meeting the needs of residents or industrial plants.

[0063] The scenario collaborative control subsystem, acting as the "brain" of the interface unit, enables intelligent identification, prioritization, and resource scheduling of multi-scenario needs, and is deeply integrated with the system's digital twin decision-making system. In scenario identification and signal interaction, it receives demand signals from various scenarios (such as dissolved oxygen concentration in aeration tanks, power grid load commands, and underwater equipment operating status) via industrial Ethernet at a sampling frequency of 10Hz. Simultaneously, it constructs simplified models of each scenario in a virtual environment (such as an oxygen transfer model for aeration tanks and a power grid load curve model), simulating changes in scenario energy demand through real-time data-driven simulation. Its priority scheduling strategy has four levels of rules: Level 1 is for emergency power supply (such as hospitals and communication base stations), Level 2 is for grid peak shaving with a response rate of ≥95%, Level 3 is for industrial aeration and underwater operations, and Level 4 is for waste heat utilization. When multiple scenarios have conflicting demands (such as grid peak shaving and industrial aeration requiring energy at the same time), the optimal allocation scheme is calculated through reinforcement learning algorithms to ensure the maximum overall benefits of the system (such as prioritizing scenarios with high energy value per kilowatt-hour). In terms of safety interlocking and protection, when an anomaly is detected in a certain scenario (such as a sudden drop in pressure at the underwater interface), the energy supply to that scenario will be cut off immediately, and the operation and maintenance personnel will be notified through audible and visual alarm devices and remote notification systems (SMS, APP push). Furthermore, key control modules (such as CPU and communication interfaces) adopt 1:1 hot backup, and the fault switching time is ≤100ms to ensure the continuity of scenario coordination.

[0064] The present invention also provides a control method, such as Figure 1 As shown, this is applied to the distributed compressed air energy storage system based on abandoned mines described in any of the above-mentioned embodiments. First, parameters are collected through a distributed fiber optic sensor network, a micro-sensor array, and a laser gas analysis system in the intelligent detection and control unit to obtain real-time system operating data. Next, a digital twin is constructed based on the data, and a virtual image of the system is generated. Based on this, an operating scheme adapted to multiple scenarios is formulated to form a control strategy. Then, adjustments are executed according to the control strategy, synchronously adjusting the operating status of core units such as the gas storage and heat management unit and the compression and expansion power unit. Finally, the adjustment effect is monitored through closed-loop feedback, and the data is transmitted back to the intelligent detection and control unit to achieve cyclical optimization of the control process and solve the problem of adaptive load change regulation of the system.

[0065] In some embodiments, the method further includes: collecting key system operating parameters through a distributed fiber optic sensor network, a micro-sensor array, and a laser gas analysis system in the intelligent detection and control unit to obtain real-time operating data covering the gas storage and thermal management unit, the compression and expansion power unit, and the multi-scenario energy interface unit; constructing a digital twin in the intelligent detection and control unit based on the real-time operating data, integrating physical models and data-driven algorithms to generate control strategies adapted to multiple scenarios; and synchronously adjusting the energy distribution status of the gas storage and thermal management unit and the operating parameters of the compression and expansion power unit according to the control strategies.

[0066] This invention provides a distributed compressed air energy storage system based on abandoned mines, comprising an air storage and heat management unit, a compression and expansion power unit, an intelligent detection and control unit, and a multi-scenario energy interface unit. The air storage and heat management unit is used to achieve efficient air storage and the recovery, storage, and dynamic distribution of heat energy during compression or expansion. The compression and expansion power unit is used to complete air compression energy storage and expansion energy release, realizing the conversion of mechanical energy into electrical energy. The intelligent detection and control unit is used to collect system parameters in real time, construct a digital twin model, and generate control strategies. The multi-scenario energy interface unit is used to dynamically adjust the output pressure, flow rate, voltage level, and temperature according to different scenario requirements. The system provided by this invention solves the problem that existing systems cannot achieve intelligent control based on adaptive load changes. This invention also provides a method applied to the above system.

[0067] The effectiveness of this invention system is reflected in multiple dimensions, including technical performance, economy, reliability, and scenario adaptability. Specifically, in terms of technical performance, the system's round-trip efficiency (electric-electric) can reach 65%-70%, which is significantly improved compared to traditional compressed air energy storage without heat storage (approximately 50%-55%). This improvement mainly relies on the magnetic levitation compressor / expander unit to reduce mechanical friction (compression efficiency increased to 85%-88%, expansion efficiency increased to 88%-90%, both under rated operating conditions), phase change heat storage to recover 50%-60% of the compression heat and increase expansion work by 15%-20% through reheat technology, and digital twin control to reduce losses under varying operating conditions (efficiency retention rate under 30%-110% load fluctuation). Approximately 85%-90%, compared to only 70%-75% for traditional units; meanwhile, the system has good dynamic response capabilities, with a power response speed of 5%-8% of the rated value per second during grid peak shaving, which can meet the peak shaving needs of general regional power grids. The switching time for multiple scenarios is approximately 1-3 seconds (such as switching from industrial aeration to emergency power supply), basically meeting non-extreme and sudden demand. The parameter control accuracy can also ensure stable operation, with a gas storage pressure steady-state error of ±0.03MPa, expander inlet temperature fluctuation of ±3-5℃, and grid-connected power tracking error of ±2%-3%, which can ensure stable system operation while having a certain control margin.

[0068] In terms of economics, the unit construction cost of the system is approximately RMB 2000-2500 / kWh (including land, equipment, and installation), lower than traditional pumped storage (RMB 2500-4000 / kWh, subject to geographical limitations). The cost breakdown is as follows: magnetic levitation turbines account for 30%-35% (compared to approximately 20%-25% for traditional mechanical turbines), and composite air storage tanks and phase change thermal energy storage systems account for 25%-30%. Modular design can further reduce costs by 10%-15%. Operation and maintenance costs are also advantageous, with annual operation and maintenance costs of approximately RMB 0.03-0.05 / kWh (compared to approximately RMB 0.06-0.08 / kWh for traditional compressed air storage). The frictionless design of magnetic levitation reduces maintenance work such as bearing replacement, resulting in a levelized cost of electricity (LCOE) of approximately RMB 0.45-0.6 / kWh, making it competitive in the long-term energy storage (≥4 hours) field.

[0069] In terms of reliability and scenario adaptability, the system is designed with a mean time between failures (MTBF) of approximately 8,000-9,000 hours (an average of 40-50 hours of downtime per year), mainly relying on the dual backup design of key components (such as compressors and sensors) and a three-level safety interlock mechanism for pressure, temperature, and vibration to reduce the risk of major accidents. The system demonstrates outstanding scenario adaptability. In industrial aeration scenarios, oxygen utilization is increased to 20%-22% (compared to approximately 15%-18% for traditional aeration), and aeration energy consumption in wastewater treatment plants is reduced by 8%-12%. In grid peak shaving scenarios, a single 100MW system can smooth out a daily peak-valley difference of 50-80MW, reducing wind and solar power curtailment by approximately 50-80 million kWh annually. In waste heat recovery scenarios, it can provide approximately 3-4 million kWh of industrial or residential heat energy annually, equivalent to saving 1,000-1,500 tons of standard coal per year.

[0070] Similar parts between the embodiments provided in this invention can be referred to mutually. The specific embodiments provided above are merely examples under the overall concept of this invention and do not constitute a limitation on the scope of protection of this application. For those skilled in the art, any other embodiments extended from the solution of this invention without creative effort are within the scope of protection of this invention.

Claims

1. A distributed compressed air energy storage system based on abandoned mines, characterized in that, It includes a gas storage and thermal management unit, a compression and expansion power unit, an intelligent detection and control unit, and a multi-scenario energy interface unit. These units are interconnected via pressure-resistant pipelines, a data bus, and thermal circulation pipelines. The gas storage and heat management unit is used to realize efficient air storage and the recovery, storage and dynamic distribution of heat energy during compression or expansion processes; The compression and expansion power unit is used to complete the compression and energy storage of air and the expansion and energy release, realizing the conversion of mechanical energy and electrical energy; The intelligent detection and control unit is connected to the gas storage and heat management unit, the compression and expansion power unit and the multi-scenario energy interface unit respectively. It is used to collect system parameters in real time, build a digital twin model and generate control strategies. The intelligent detection and control unit monitors the energy storage system through a distributed optical fiber sensor network, a micro sensor array and a laser system analysis system, and controls the system through a digital twin system and an execution and decision system. The multi-scenario energy interface unit is used to dynamically adjust the output pressure, flow rate, voltage level and temperature according to different scenario requirements.

2. The system according to claim 1, characterized in that, The gas storage and heat management unit includes a gas storage unit and a heat management unit. The gas storage unit includes a composite gas storage tank body, a sealing and connection system, and a pressure regulating device. The composite gas storage tank body is provided with an outer rigid support, a middle insulation layer, and an inner flexible buffer bladder from the outside to the inside.

3. The system according to claim 2, characterized in that, The inner flexible buffer bladder is a nitrile rubber-aramid fiber composite membrane with a thickness of 5 mm, wherein the nitrile rubber layer has a thickness of 3 mm and the aramid braided layer has a thickness of 2 mm.

4. The system according to claim 3, characterized in that, The thermal management unit includes a thermal storage subsystem, a cold storage subsystem, and an intelligent heat distribution system. The thermal storage subsystem, cold storage subsystem, and intelligent heat distribution system are interconnected. The intelligent heat distribution system is used to detect and dynamically adjust the temperature of the thermal storage subsystem and the cold storage subsystem. The thermal storage subsystem includes a phase change thermal storage material, which is a sodium nitrate-potassium nitrate mixed salt with a mass ratio of sodium nitrate to potassium nitrate of 60:

40.

5. The system according to claim 4, characterized in that, The cold storage subsystem includes a cold storage medium, which is a 40% ethylene glycol aqueous solution.

6. The system according to claim 1, characterized in that, The compression and expansion power unit includes a compressor and an expander unit, the compressor and the expander unit are connected, the compressor includes a magnetic levitation compressor, and the expander unit includes a turbine expander.

7. The system according to claim 1, characterized in that, The distributed optical fiber sensing network includes armored optical fibers, which are spirally wound along the outer wall of the gas storage tank and arranged axially along the compressor and expander unit axis; the micro sensor array includes a pressure sensor array, a flow sensor array, and a temperature sensor array; the laser gas analysis system includes a tunable semiconductor laser absorption spectrometer, which is installed in the gas storage tank outlet and expander inlet pipe.

8. The system according to claim 1, characterized in that, The multi-scenario energy interface unit includes a pneumatic interface subsystem, an electric interface subsystem, a waste heat recovery and utilization interface subsystem, and a scenario collaborative control subsystem, depending on the different scenario requirements.

9. A control method, characterized in that, Applied to the distributed compressed air energy storage system based on abandoned mines as described in any one of claims 1 to 8.

10. The method according to claim 9, characterized in that, include: By using the distributed fiber optic sensing network, micro-sensor array and laser gas analysis system in the intelligent detection and control unit, key operating parameters of the system are collected to obtain real-time operating data covering the gas storage and thermal management unit, the compression and expansion power unit and the multi-scenario energy interface unit. Based on real-time operational data, a digital twin is constructed in the intelligent detection and control unit, integrating physical models and data-driven algorithms to generate control strategies adapted to multiple scenarios; According to the control strategy, the energy distribution status of the gas storage and heat management unit and the operating parameters of the compression and expansion power unit are adjusted synchronously.

Images