Distributed multi-energy supply system and method based on integration of internal combustion engine heat pump

Abstract

The application provides a distributed multi-energy supply system and method based on an integrated heat pump of an internal combustion engine. It belongs to the field of multi-energy supply. The system comprises an integrated core energy supply subsystem, a photovoltaic energy storage subsystem, an intelligent control subsystem, a cold and warm energy supply subsystem and a waste heat recovery subsystem. The integrated core energy supply subsystem comprises a natural gas internal combustion engine generator integrated machine and a gas storage and supply system. The photovoltaic energy storage subsystem comprises photovoltaic components, an inverter, energy storage batteries and a BMS management system. The intelligent control subsystem comprises an edge intelligent controller and a cloud AI platform. The cold and warm energy supply subsystem comprises a heat pump heat exchanger, a circulating water pump and terminal energy supply equipment. The waste heat recovery subsystem comprises a cylinder sleeve water heat exchanger, a flue gas waste heat heat exchanger and a pressure-bearing buffer water tank. Through integrated arrangement, the energy conversion loss is reduced on the basis of reducing the installation space.

Description

Technical Field

[0001] This application relates to the field of multi-energy supply, specifically to a distributed multi-energy supply system and method based on the integration of internal combustion engine and heat pump. Background Technology

[0002] Traditional distributed multi-energy supply systems typically configure CHP units, photovoltaic systems, energy storage systems, and heat pump systems independently. There is a lack of deep integration and synergistic optimization between these devices, resulting in the following core technical defects: low equipment integration, leading to large system size, complex pipelines, and high maintenance costs; multiple energy conversion stages, where the internal combustion engine generates electricity and then needs to be converted into electricity to drive the heat pump, resulting in additional energy loss, and the overall energy utilization rate is generally less than 85%. Summary of the Invention

[0003] The purpose of this application is to provide a distributed multi-energy supply system and method based on the integration of internal combustion engine and heat pump, which reduces energy conversion losses by reducing installation space through integrated design.

[0004] This application is implemented as follows: In a first aspect, this application provides a distributed multi-energy supply system based on an integrated internal combustion engine and heat pump, comprising: The integrated core energy supply subsystem includes a natural gas internal combustion engine generator and a gas storage and supply system. The generator integrates a natural gas internal combustion engine, a generator, a drive motor, a heat pump compressor, a motor switching device, and a mechanical coupling device. The mechanical coupling device adopts a shared spindle design, allowing the generator and drive motor to share the spindle of the natural gas internal combustion engine. The motor switching device is used to control the connection and disconnection of the generator and drive motor with the spindle, thereby achieving seamless switching between power generation mode, drive mode, and cooperative mode. The photovoltaic energy storage subsystem includes photovoltaic modules, inverters, energy storage batteries, and a BMS management system. The photovoltaic modules are electrically connected to the load and the energy storage batteries through the inverters. The BMS management system is connected to the energy storage batteries to monitor and manage the battery status. The photovoltaic energy storage subsystem is used to provide peak shaving supplement and backup power support. The intelligent control subsystem includes an edge intelligent controller and a cloud AI platform. The edge intelligent controller is connected to the integrated generator, inverter, energy storage battery, heat pump compressor and waste heat recovery subsystem to collect operating data and perform local scheduling. The cloud AI platform is connected to the edge intelligent controller to realize AI predictive analysis and global optimization decision-making. The heating and cooling energy supply subsystem includes a heat pump heat exchanger, a circulating water pump, and terminal energy supply equipment; the heat pump compressor is connected to the heat pump heat exchanger through a refrigerant pipeline, and the heat pump heat exchanger is connected to the terminal energy supply equipment through a circulating water pump to realize the building's heating and cooling energy supply. The waste heat recovery subsystem includes a cylinder liner water heat exchanger, a flue gas waste heat heat exchanger, and a pressure buffer tank. The cylinder liner water heat exchanger and the flue gas waste heat heat exchanger are respectively connected to the cylinder liner water outlet and the flue gas outlet of the natural gas internal combustion engine through pipelines. The outlets of both are connected to the pressure buffer tank for recovering engine waste heat.

[0005] Based on the first aspect, the operating modes of the integrated natural gas internal combustion engine generator include power generation mode, drive mode, and coordinated mode. In power generation mode, the natural gas internal combustion engine drives the generator through a mechanical coupling device to output electrical energy. In drive mode, the natural gas internal combustion engine switches power output through a mechanical coupling device to drive the drive motor, which directly drives the heat pump compressor. In coordinated mode, the natural gas internal combustion engine drives both the generator and the drive motor simultaneously, achieving coordinated operation of power generation and heat pump drive.

[0006] Based on the first aspect, the mechanical coupling device includes a common spindle, an electromagnetic clutch assembly, and a transmission gear set; the common spindle is used to transmit power from the natural gas internal combustion engine; the electromagnetic clutch assembly includes a first clutch for controlling the connection of the generator and a second clutch for controlling the connection of the drive motor; the transmission gear set includes a first gear set from the spindle to the generator and a second gear set from the spindle to the drive motor.

[0007] Based on the first aspect, the intelligent control subsystem adopts a three-layer architecture of cloud-edge collaboration, including the edge layer, cloud layer, and power grid interaction layer. The edge layer collects equipment operation data in real time through edge intelligent controllers and executes local intelligent scheduling logic. The cloud layer realizes AI predictive analysis and global optimization decision-making through the cloud AI platform. The power grid interaction layer realizes encrypted two-way communication with the State Grid intelligent dispatch system through two-way interactive terminals.

[0008] Based on the first aspect, the cloud-based AI platform adopts an improved multi-timescale prediction algorithm, which includes: photovoltaic power prediction based on a temporal convolutional network and introducing a meteorological feature attention mechanism; load demand prediction based on a bidirectional long short-term memory network and incorporating the coupling features of electric and thermal loads; the scheduling algorithm of the intelligent control subsystem adopts a combination of deep reinforcement learning and model predictive control, wherein the optimization result of model predictive control is used as the state input of the deep reinforcement learning algorithm, and the deep reinforcement learning strategy is used as the initial solution of model predictive control.

[0009] Based on the first aspect, the waste heat recovery subsystem adopts a two-stage recovery method: the waste heat of the cylinder liner water is recovered through the cylinder liner water heat exchanger, and the waste heat of the flue gas is recovered through the flue gas waste heat heat exchanger.

[0010] Based on the first aspect, the intelligent control subsystem supports three operating modes, including normal grid-connected mode, grid emergency power supply mode, and off-grid operation mode. In normal grid-connected mode, photovoltaic power output is prioritized to supply the system's auxiliary loads. In grid emergency power supply mode, after receiving a power shortage or peak-shaving dispatch instruction from the grid, the integrated generator switches to collaborative mode. In off-grid operation mode, the system automatically switches to off-grid mode when there is a grid fault and power outage.

[0011] Secondly, this application provides a distributed multi-energy supply method based on the integration of internal combustion engine and heat pump, comprising the following steps: S1: Real-time acquisition of system operation data, including generator power, drive motor power, photovoltaic output, energy storage SOC, electrical load, heat load, electricity price and meteorological data; S2: Based on an improved multi-timescale prediction algorithm, it predicts photovoltaic power, load demand and electricity price for the next 24 hours; the improvements include the fusion of meteorological attention mechanism of temporal convolutional network and electrothermal coupling feature of bidirectional long short-term memory network; S3: Based on the prediction results, an algorithm combining deep reinforcement learning and model predictive control is used to generate the optimal scheduling strategy for the next 24 hours. S4: The edge intelligent controller receives the optimal scheduling strategy and executes the control command for the first time step; S5: Repeat steps S1 to S4 every 5 minutes to perform rolling optimization.

[0012] Based on the second aspect, step S3, generating the optimal scheduling strategy, further includes: establishing a multi-objective optimization function that includes economy, reliability, and environmental protection; setting energy balance constraints, energy storage constraints, equipment power constraints, motor switching constraints, and comfort constraints; training a deep reinforcement learning agent using the Soft Actor-Critic algorithm and designing a multi-objective reward function; using model predictive control for rolling optimization every 5 minutes; and using the optimization results of model predictive control as the state input of the deep reinforcement learning algorithm, with the deep reinforcement learning strategy serving as the initial solution for model predictive control.

[0013] Based on the second aspect, the execution of control commands in step S4 further includes: determining the operating mode of the integrated generator according to the scheduling strategy; switching modes through a motor switching device; adjusting the power distribution between the generator and the drive motor through a mechanical coupling device; and adjusting the photovoltaic output distribution, energy storage charging and discharging power, and grid power purchase and sale.

[0014] Compared with the prior art, this application has at least the following advantages or beneficial effects: This invention provides a distributed multi-energy supply system and method based on an integrated internal combustion engine and heat pump. Through the integrated design of the internal combustion engine, generator, and drive motor, it achieves "dual-purpose use of one machine," eliminating the need for a separate heat pump drive motor, reducing energy conversion steps, and lowering energy conversion losses by approximately 15%-20%. Through the collaboration of energy storage batteries and intelligent dispatching, it prioritizes local consumption of photovoltaic power, with excess energy stored for backup or fed back to the grid. It supports three operation modes: normal grid connection, grid emergency power supply, and off-grid operation, and can participate in peak shaving in response to grid dispatching commands. The integrated generator design eliminates the need for a separate heat pump drive motor, reducing equipment volume by 30%, simplifying pipelines, and minimizing installation space. By efficiently utilizing natural gas and clean photovoltaic energy, carbon emissions are reduced; environmental targets are set in the multi-objective reward function, prioritizing low-emission schemes during optimized operation. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of a distributed multi-energy supply system based on an integrated internal combustion engine and heat pump, as described in this application. Figure 2 This is a flowchart of a distributed multi-energy supply method based on the integration of internal combustion engine and heat pump, as described in this application.

[0017] icon: 100. Integrated Core Energy Supply Subsystem; 101. Natural Gas Internal Combustion Engine Generator; 1011. Natural Gas Internal Combustion Engine; 1012. Generator; 1013. Drive Motor; 1014. Heat Pump Compressor; 1015. Motor Switching Device; 1016. Mechanical Coupling Device; 102. Gas Storage and Supply System; 200. Photovoltaic Energy Storage Subsystem; 201. Photovoltaic Module; 202. Inverter; 203. Energy Storage Battery; 204. BMS Management System; 300. Intelligent Control Subsystem; 301. Edge Intelligent Controller; 302. Cloud AI Platform; 400. Heating and Cooling Energy Supply Subsystem; 401. Heat Pump Heat Exchanger; 402. Circulating Water Pump; 403. Terminal Energy Supply Equipment; 500. Waste Heat Recovery Subsystem; 501. Cylinder Liner Water Heat Exchanger; 502. Flue Gas Waste Heat Exchanger; 503. Pressure Buffer Water Tank. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the various embodiments and features described below can be combined with each other. Example

[0020] This application provides a distributed multi-energy supply system and method based on the integration of internal combustion engine and heat pump. Through integrated design, it reduces energy conversion losses while minimizing installation space.

[0021] Please refer to Figure 1 A distributed multi-energy supply system based on an internal combustion engine heat pump integration includes: an integrated core energy supply subsystem 100, a photovoltaic energy storage subsystem 200, an intelligent control subsystem 300, a heating and cooling energy supply subsystem 400, and a waste heat recovery subsystem 500, wherein: The integrated core energy supply subsystem 100 includes a natural gas internal combustion engine generator 101 and a gas storage and supply system 102. The generator 101 integrates a natural gas internal combustion engine 1011, a generator 1012, a drive motor 1013, a heat pump compressor 1014, a motor switching device 1015, and a mechanical coupling device 1016. The mechanical coupling device 1016 adopts a shared spindle design, allowing the generator 1012 and the drive motor 1013 to share the spindle of the natural gas internal combustion engine 1011. The motor switching device 1015 is used to control the connection and disconnection of the generator 1012 and the drive motor 1013 with the spindle, thereby achieving seamless switching between power generation mode, drive mode, and cooperative mode. By integrating the internal combustion engine, generator, and drive motor into a single unit, the independent heat pump drive motor is eliminated, reducing energy conversion steps, lowering equipment size and installation space, and simultaneously achieving the function of "one machine for two purposes".

[0022] The photovoltaic energy storage subsystem 200 includes a photovoltaic module 201, an inverter 202, an energy storage battery 203, and a BMS management system 204. The photovoltaic module 201 is electrically connected to the load and the energy storage battery 203 through the inverter 202. The BMS management system 204 is connected to the energy storage battery 203 and is used to monitor and manage the battery status. The photovoltaic energy storage subsystem 200 is used to provide peak shaving supplement and backup power support. This setup enables self-consumption of photovoltaic power, with excess energy stored in batteries to provide backup power when photovoltaic output is insufficient or when the grid fails, thus improving the system's power supply reliability.

[0023] The intelligent control subsystem 300 includes an edge intelligent controller 301 and a cloud AI platform 302. The edge intelligent controller 301 is communicatively connected to the integrated generator 101, inverter 202, energy storage battery 203, heat pump compressor 1014, and waste heat recovery subsystem 500, respectively, and is used to collect operating data and perform local scheduling. The cloud AI platform 302 is communicatively connected to the edge intelligent controller 301 and is used to realize AI predictive analysis and global optimization decision-making. Specifically, the intelligent control subsystem 300 adopts a three-layer cloud-edge collaborative architecture, including an edge layer, a cloud layer, and a power grid interaction layer. The edge layer collects equipment operation data in real time and executes local intelligent scheduling logic through the edge intelligent controller 301. The cloud layer realizes AI predictive analysis and global optimization decision-making through the cloud AI platform 302. The power grid interaction layer realizes encrypted two-way communication with the State Grid intelligent dispatch system through a two-way interactive terminal.

[0024] The cloud-edge collaborative architecture achieves a unified approach to low-latency local control and global optimization, thereby improving the system's intelligence level and grid interaction capabilities.

[0025] The heating and cooling energy supply subsystem 400 includes a heat pump heat exchanger 401, a circulating water pump 402, and a terminal energy supply device 403; the heat pump compressor 1014 is connected to the heat pump heat exchanger 401 through a refrigerant pipeline, and the heat pump heat exchanger 401 is connected to the terminal energy supply device 403 through the circulating water pump 402, for the purpose of realizing building heating and cooling energy supply. The waste heat recovery subsystem 500 includes a cylinder liner water heat exchanger 501, a flue gas waste heat heat exchanger 502, and a pressure buffer water tank 503. The cylinder liner water heat exchanger 501 and the flue gas waste heat heat exchanger 502 are respectively connected to the cylinder liner water outlet and the flue gas outlet of the natural gas internal combustion engine 1011 through pipelines. The outlets of both are connected to the pressure buffer water tank 503 for recovering engine waste heat.

[0026] With this setup, regardless of the operating mode of the generator set, the cylinder liner water and flue gas waste heat generated by the engine can be recovered and utilized, significantly improving the overall energy utilization rate of the system.

[0027] In some embodiments of the present invention, the operating modes of the natural gas internal combustion engine generator 101 include a power generation mode, a drive mode, and a coordinated mode. In the power generation mode, the natural gas internal combustion engine 1011 drives the generator 1012 through a mechanical coupling device 1016 to output electrical energy. In the drive mode, the natural gas internal combustion engine 1011 switches power output through the mechanical coupling device 1016 to drive the drive motor 1013, and the drive motor 1013 directly drives the heat pump compressor 1014. In the coordinated mode, the natural gas internal combustion engine 1011 simultaneously drives the generator 1012 and the drive motor 1013 to achieve coordinated operation of power generation and heat pump drive.

[0028] By flexibly switching between the three modes, the operating mode can be dynamically adjusted according to real-time load demand, achieving dynamic matching of energy supply and demand and avoiding energy waste.

[0029] In some embodiments of the present invention, the mechanical coupling device 1016 includes a common spindle, an electromagnetic clutch assembly, and a transmission gear set; the common spindle is used to transmit power to the natural gas internal combustion engine 1011; the electromagnetic clutch assembly includes a first clutch for controlling the connection of the generator and a second clutch for controlling the connection of the drive motor; the transmission gear set includes a first gear set from the spindle to the generator and a second gear set from the spindle to the drive motor.

[0030] Specifically, in the power generation mode: the first clutch is engaged, the second clutch is disengaged, and the natural gas internal combustion engine drives the generator to rotate and generate electricity through the main shaft and the first gear set.

[0031] In drive mode: the first clutch is disengaged and the second clutch is engaged. The natural gas internal combustion engine drives the drive motor to rotate through the main shaft and the second gear set, thereby driving the heat pump compressor.

[0032] In cooperative mode: the first clutch and the second clutch engage simultaneously, and the natural gas internal combustion engine drives both the generator and the drive motor.

[0033] By sharing a main shaft and an electromagnetic clutch assembly, a single internal combustion engine can independently or collaboratively drive the generator and drive motor. The structure is compact, the switching is reliable, and the redundant design of using two independent power systems is avoided.

[0034] In some embodiments of the present invention, the cloud AI platform 302 adopts an improved multi-timescale prediction algorithm. The prediction algorithm includes: photovoltaic power prediction based on a temporal convolutional network and introducing a meteorological feature attention mechanism; load demand prediction based on a bidirectional long short-term memory network and incorporating the coupling features of electric and thermal loads; the scheduling algorithm of the intelligent control subsystem 300 adopts a combination of deep reinforcement learning and model predictive control, wherein the optimization result of model predictive control is used as the state input of the deep reinforcement learning algorithm, and the deep reinforcement learning strategy is used as the initial solution of model predictive control.

[0035] Specifically, the cloud-based AI platform 302 employs an improved multi-timescale prediction algorithm: Photovoltaic power prediction is based on a temporal convolutional network (TCN) and incorporates a meteorological feature attention mechanism. This improvement, by adding an attention module after the expanded convolutional layer, adaptively weights meteorological features such as irradiance, temperature, and humidity, effectively capturing the key impact of irradiance variation rate on photovoltaic output.

[0036] Load demand forecasting is based on a bidirectional long short-term memory network (BiLSTM) and incorporates electrothermal load coupling features. By constructing an electrothermal coupling feature vector (including electrical load, thermal load, their product, rate of change, and historical ratio), and employing a dual-stream BiLSTM architecture to achieve deep interaction of electrothermal information, the correlation between electrothermal loads can be explicitly expressed.

[0037] The scheduling algorithm of the intelligent control subsystem 300 adopts a combination of deep reinforcement learning (DRL) and model predictive control (MPC): The deep reinforcement learning module employs the Soft Actor-Critic (SAC) algorithm, which designs a multi-objective reward function to simultaneously optimize economy, reliability, and environmental friendliness.

[0038] The model prediction control module is based on mixed integer linear programming (MILP) and updates the scheduling strategy every 5 minutes.

[0039] DRL and MPC collaborative mechanism: The MPC optimization result serves as the state input of the SAC algorithm, and the SAC strategy serves as the initial solution of MPC, achieving complementary advantages.

[0040] With this setup, the improved TCN meteorological attention mechanism enhances the accuracy of photovoltaic power forecasting; the improved BiLSTM electrothermal coupling feature fusion enhances the accuracy of load forecasting; and the DRL and MPC collaborative mechanism balances the needs of long-term strategy learning and short-term rolling optimization, improving the global optimality and real-time performance of scheduling decisions.

[0041] Furthermore, the waste heat recovery subsystem 500 adopts a two-stage recovery method: the waste heat of the cylinder liner water is recovered through the cylinder liner water heat exchanger 501, and the waste heat of the flue gas is recovered through the flue gas waste heat heat exchanger 502.

[0042] In some embodiments of the present invention, the intelligent control subsystem 300 supports three operating modes, including normal grid-connected mode, grid emergency power supply mode, and off-grid operation mode. In normal grid-connected mode, photovoltaic power output is prioritized to supply the system's auxiliary loads. In grid emergency power supply mode, after receiving a power shortage or peak-shaving dispatch instruction from the grid, the integrated generator switches to collaborative mode. In off-grid operation mode, the system automatically switches to off-grid mode when the grid experiences a power outage.

[0043] With this configuration, the system has the capability for multi-mode adaptive switching, enabling it to operate normally connected to the grid, respond to grid dispatch instructions to participate in peak shaving, and operate independently off-grid during grid faults, thereby enhancing the system's grid interaction capability and power supply reliability.

[0044] Please refer to Figure 2 The present invention also provides a distributed multi-energy supply method based on the integration of internal combustion engine and heat pump, comprising the following steps: S1: Real-time acquisition of system operation data, including generator power, drive motor power, photovoltaic output, energy storage SOC, electrical load, heat load, electricity price and meteorological data; S2: Based on an improved multi-timescale prediction algorithm, it predicts photovoltaic power, load demand and electricity price for the next 24 hours; the improvements include the fusion of meteorological attention mechanism of temporal convolutional network and electrothermal coupling feature of bidirectional long short-term memory network; S3: Based on the prediction results, an algorithm combining deep reinforcement learning and model predictive control is used to generate the optimal scheduling strategy for the next 24 hours. S4: The edge intelligent controller receives the optimal scheduling strategy and executes the control command for the first time step; S5: Repeat steps S1 to S4 every 5 minutes to perform rolling optimization.

[0045] Specifically, through a closed-loop control process of "collection-prediction-optimization-execution-rolling", intelligent scheduling of multiple energy sources is realized, enabling the system to adaptively respond to load changes, electricity price fluctuations and weather conditions, thereby improving operational economy and energy supply reliability.

[0046] Furthermore, step S3, generating the optimal scheduling strategy, further includes: establishing a multi-objective optimization function that includes economy, reliability, and environmental protection; setting energy balance constraints, energy storage constraints, equipment power constraints, motor switching constraints, and comfort constraints; training a deep reinforcement learning agent using the Soft Actor-Critic algorithm and designing a multi-objective reward function; using model predictive control for rolling optimization every 5 minutes; and using the optimization results of model predictive control as the state input of the deep reinforcement learning algorithm, with the deep reinforcement learning strategy serving as the initial solution for model predictive control.

[0047] Furthermore, the execution of control commands in step S4 further includes: determining the operating mode of the integrated generator according to the scheduling strategy; switching modes through a motor switching device; adjusting the power distribution between the generator and the drive motor through a mechanical coupling device; and adjusting the photovoltaic output distribution, energy storage charging and discharging power, and grid power purchase and sale.

[0048] To verify the technical effectiveness of this invention, a comparative experiment was conducted for 62 consecutive days at a research and development center. The experimental equipment was the system of this invention (15kW integrated generator + 12kW photovoltaic + 15kWh energy storage), and the comparative equipment was a traditional discrete energy supply system (10kW CHP unit + 12kW independent heat pump + 10kW photovoltaic + 10kWh energy storage). Please refer to Tables 1-3 for the experimental results: Table 1. Test Results of System Overall Energy Utilization Rate

[0049] The experimental results in Table 1 show that the overall energy utilization rate of the system of the present invention reaches 92.3%, which meets the technical requirement of ≥92%. Compared with the traditional system, the overall energy utilization rate is increased by 8.9%, the photovoltaic self-generation and self-consumption rate is increased by 31.7%, and the annual operating cost is reduced by 17.5%, which verifies the beneficial effects of the present invention.

[0050] Table 2. Verification of the Improved AI Algorithm Effect

[0051] The experimental results in Table 2 show that the improved TCN meteorological attention mechanism improves the photovoltaic power prediction accuracy by 38.7%; the improved BiLSTM electrothermal coupling feature fusion improves the load prediction accuracy by 33.3%; and the DRL and MPC collaborative mechanism reduces the MPC solution time by 62.4%, verifying the effectiveness of the AI ​​algorithm improvement in this invention.

[0052] Table 3. Off-grid mode energy storage SOC maintenance test results

[0053] The experimental results in Table 3 show that in off-grid mode, the system maintains the energy storage SOC within the range of 35%-80% through intelligent scheduling strategy, with a minimum SOC of 35%. The system operates stably and provides continuous energy, verifying the practicality of the off-grid mode of this invention.

[0054] The memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc.

[0055] A processor can be an integrated circuit chip with signal processing capabilities. This processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0056] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this application. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A distributed multi-energy supply system based on an integrated internal combustion engine and heat pump, characterized in that, include: An integrated core energy supply subsystem (100) includes a natural gas internal combustion engine generator (101) and a gas storage and supply system (102). The generator (101) integrates a natural gas internal combustion engine (1011), a generator (1012), a drive motor (1013), a heat pump compressor (1014), a motor switching device (1015), and a mechanical coupling device (1016). The mechanical coupling device (1016) adopts a shared spindle design, allowing the generator (1012) and the drive motor (1013) to share the spindle of the natural gas internal combustion engine (1011). The motor switching device (1015) is used to control the connection and disconnection of the generator (1012) and the drive motor (1013) with the spindle, thereby achieving seamless switching between power generation mode, drive mode, and cooperative mode. A photovoltaic energy storage subsystem (200) includes a photovoltaic module (201), an inverter (202), an energy storage battery (203), and a BMS management system (204). The photovoltaic module (201) is electrically connected to the load and the energy storage battery (203) through the inverter (202). The BMS management system (204) is connected to the energy storage battery (203) and is used to monitor and manage the battery status. The photovoltaic energy storage subsystem (200) is used to provide peak shaving supplement and backup power support. The intelligent control subsystem (300) includes an edge intelligent controller (301) and a cloud AI platform (302); the edge intelligent controller (301) is communicatively connected to the integrated generator (101), the inverter (202), the energy storage battery (203), the heat pump compressor (1014), and the waste heat recovery subsystem (500), and is used to collect operating data and perform local scheduling; the cloud AI platform (302) is communicatively connected to the edge intelligent controller (301) and is used to realize AI predictive analysis and global optimization decision-making; The heating and cooling energy supply subsystem (400) includes a heat pump heat exchanger (401), a circulating water pump (402), and a terminal energy supply device (403); the heat pump compressor (1014) is connected to the heat pump heat exchanger (401) through a refrigerant pipeline, and the heat pump heat exchanger (401) is connected to the terminal energy supply device (403) through the circulating water pump (402) to realize building heating and cooling energy supply; The waste heat recovery subsystem (500) includes a cylinder liner water heat exchanger (501), a flue gas waste heat heat exchanger (502), and a pressure buffer tank (503). The cylinder liner water heat exchanger (501) and the flue gas waste heat heat exchanger (502) are respectively connected to the cylinder liner water outlet and the flue gas outlet of the natural gas internal combustion engine (1011) through pipelines. The outlets of both are connected to the pressure buffer tank (503) for recovering engine waste heat.

2. The distributed multi-energy supply system based on the integration of internal combustion engine and heat pump according to claim 1, characterized in that, The operation modes of the natural gas internal combustion engine generator (101) include a power generation mode, a drive mode, and a coordinated mode. In the power generation mode, the natural gas internal combustion engine (1011) drives the generator (1012) through the mechanical coupling device (1016) to output electrical energy. In the drive mode, the natural gas internal combustion engine (1011) switches power output through the mechanical coupling device (1016) to drive the drive motor (1013), and the drive motor (1013) directly drives the heat pump compressor (1014). In the coordinated mode, the natural gas internal combustion engine (1011) simultaneously drives the generator (1012) and the drive motor (1013) to achieve coordinated operation of power generation and heat pump drive.

3. A distributed multi-energy supply system based on an integrated internal combustion engine and heat pump as described in claim 1, characterized in that, The mechanical coupling device (1016) includes a common spindle, an electromagnetic clutch assembly, and a transmission gear set; the common spindle is used to transmit power from the natural gas internal combustion engine (1011); the electromagnetic clutch assembly includes a first clutch for controlling the connection of the generator and a second clutch for controlling the connection of the drive motor; the transmission gear set includes a first gear set from the spindle to the generator and a second gear set from the spindle to the drive motor.

4. A distributed multi-energy supply system based on an integrated internal combustion engine and heat pump as described in claim 1, characterized in that, The intelligent control subsystem (300) adopts a cloud-edge collaborative three-layer architecture, including an edge layer, a cloud layer, and a power grid interaction layer. The edge layer collects equipment operation data in real time and executes local intelligent scheduling logic through the edge intelligent controller (301). The cloud layer realizes AI predictive analysis and global optimization decision-making through the cloud AI platform (302). The power grid interaction layer realizes encrypted two-way communication with the State Grid intelligent dispatch system through a two-way interactive terminal.

5. A distributed multi-energy supply system based on an integrated internal combustion engine and heat pump as described in claim 1, characterized in that, The cloud-based AI platform (302) employs an improved multi-timescale prediction algorithm, which includes: photovoltaic power prediction based on a temporal convolutional network and incorporating a meteorological feature attention mechanism; load demand prediction based on a bidirectional long short-term memory network and integrating electrical and thermal load coupling features; the scheduling algorithm of the intelligent control subsystem (300) adopts a combination of deep reinforcement learning and model predictive control, wherein the optimization result of model predictive control is used as the state input of the deep reinforcement learning algorithm, and the deep reinforcement learning strategy is used as the initial solution of model predictive control.

6. A distributed multi-energy supply system based on an integrated internal combustion engine and heat pump as described in claim 1, characterized in that, The waste heat recovery subsystem (500) adopts a two-stage recovery method: the waste heat of the cylinder liner water is recovered through the cylinder liner water heat exchanger (501), and the waste heat of the flue gas is recovered through the flue gas waste heat heat exchanger (502).

7. A distributed multi-energy supply system based on an integrated internal combustion engine and heat pump as described in claim 1, characterized in that, The intelligent control subsystem (300) supports three operating modes, including normal grid-connected mode, grid emergency power supply mode, and off-grid operation mode. In the normal grid-connected mode, the photovoltaic power output is prioritized to supply the system's auxiliary loads. In the grid emergency power supply mode, after receiving a power shortage or peak-shaving dispatch instruction from the grid, the generator switches to the collaborative mode. In the off-grid operation mode, the system automatically switches to the off-grid mode when the grid fails and loses power.

8. A distributed multi-energy supply method based on the integration of internal combustion engine and heat pump, characterized in that, Includes the following steps: S1: Real-time acquisition of system operation data, including generator power, drive motor power, photovoltaic output, energy storage SOC, electrical load, heat load, electricity price and meteorological data; S2: Based on an improved multi-timescale prediction algorithm, the photovoltaic power, load demand, and electricity price for the next 24 hours are predicted; the improvement includes the fusion of the meteorological attention mechanism of the temporal convolutional network and the electrothermal coupling feature of the bidirectional long short-term memory network; S3: Based on the prediction results, an algorithm combining deep reinforcement learning and model predictive control is used to generate the optimal scheduling strategy for the next 24 hours. S4: The edge intelligent controller receives the optimal scheduling strategy and executes the control command for the first time step; S5: Repeat steps S1 to S4 every 5 minutes to perform rolling optimization.

9. A distributed multi-energy supply method based on an integrated internal combustion engine heat pump according to claim 8, characterized in that, Step S3, generating the optimal scheduling strategy, further includes: establishing a multi-objective optimization function that includes economy, reliability, and environmental protection; setting energy balance constraints, energy storage constraints, equipment power constraints, motor switching constraints, and comfort constraints; training a deep reinforcement learning agent using the Soft Actor-Critic algorithm and designing a multi-objective reward function; using model predictive control for rolling optimization every 5 minutes; and using the optimization results of model predictive control as the state input of the deep reinforcement learning algorithm, with the deep reinforcement learning strategy serving as the initial solution for model predictive control.

10. A distributed multi-energy supply method based on an integrated internal combustion engine heat pump according to claim 8, characterized in that, Step S4, executing control commands, further includes: determining the operating mode of the integrated generator according to the scheduling strategy; switching modes through a motor switching device; adjusting the power distribution between the generator and the drive motor through a mechanical coupling device; and adjusting the photovoltaic output distribution, energy storage charging and discharging power, and grid power purchase and sale.

Images