Distributed energy system based on gas-fired boiler and operation method

Abstract

The invention relates to the technical field of gas-fired boilers, in particular to a distributed energy system based on a gas-fired boiler and an operation method. A three-layer convolutional neural network model is adopted by a load prediction module of the intelligent control platform, a cooling, heating and power load prediction value is output through convolutional layer feature extraction and a full connection layer, and an energy scheduling module is combined based on a model prediction control algorithm, and the lowest system operation cost is taken as a target. State variables such as steam pressure and the opening degree of a fuel regulating valve are optimized in a rolling mode, the load prediction accuracy and the multivariable collaborative optimization capacity are improved, and steam emptying is reduced. A steam exhaust port of a low-pressure steam turbine of a double-shaft back pressure type steam turbine set is divided into two paths which are connected with a steam-water heat exchanger and an absorption refrigerator respectively, steam exhaust flow distribution is adjusted through an electric adjusting valve, a water heat storage tank and a lithium battery energy storage system are combined, cooperative control is conducted through an energy management gateway, heat is stored and released during heat load fluctuation, and charging and discharging are conducted in the valley-peak section of a power grid. The heat-to-electricity ratio adjusting range is widened, and insufficient heat supply is reduced.

Description

Technical Field

[0001] This invention relates to the field of gas-fired boiler technology, and in particular to a distributed energy system and operation method based on a gas-fired boiler. Background Technology

[0002] Gas-fired boilers are core equipment in distributed energy systems. They convert chemical energy into thermal energy by burning fuels such as natural gas or biomass gas to heat boiler feedwater and generate steam. The steam drives a steam turbine to achieve combined heat and power (CHP). They can also provide heat or cold energy through waste heat recovery devices.

[0003] However, existing technologies have the following significant drawbacks: load forecasting relies on empirical formulas and cannot accurately capture the dynamic changes in cooling, heating, and power loads; energy dispatch strategies lack multi-variable collaborative optimization capabilities; the range of heat-to-power ratio adjustment is limited; when the proportion of user heat load is too high, insufficient heating or excess steam venting is likely to occur; and the capacity to absorb renewable energy sources such as wind and solar power is insufficient, making it difficult to meet the needs of low-carbon energy transition. Summary of the Invention

[0004] Therefore, the technical problem to be solved by the present invention is to overcome the problems in the prior art, such as load forecasting relying on empirical formulas, failing to accurately capture the dynamic changes of cooling, heating and power loads, lacking multi-variable collaborative optimization capabilities in energy dispatch strategies, and having a limited range of heat-to-power ratio adjustment.

[0005] To address the aforementioned technical problems, this invention provides a distributed energy system based on a gas-fired boiler, comprising:

[0006] The gas-fired boiler module includes a main furnace and an auxiliary furnace. The outlet of the main furnace is connected to the inlet of the auxiliary furnace via a flue gas duct. The auxiliary furnace has a built-in spiral tube sheet heat exchanger. The main furnace has a fuel inlet and a primary air inlet on its left side wall, a secondary air inlet on its right side wall, and a slag discharge port at its bottom. The auxiliary furnace has a chimney at the top and a feedwater pipe at the bottom connected to the deaerator outlet. A preheater and a main feedwater pump are connected in series on the feedwater pipe. The preheater is used to recover waste heat from the flue gas in the auxiliary furnace to heat the boiler feedwater, and the main feedwater pump is used to pressurize and deliver the deaerated feedwater to the gas-fired boiler.

[0007] A dual-shaft back-pressure steam turbine unit, connected to the gas-fired boiler module, includes a high-pressure steam turbine and a low-pressure steam turbine, used to regulate the distribution of exhaust steam flow;

[0008] The intelligent control module, connected to the gas boiler module and the dual-shaft back-pressure turbine assembly, includes an edge computing controller, an industrial switch, and a distributed sensor network, used to acquire data and perform calculations.

[0009] Preferably, it further includes:

[0010] The fuel supply module, connected to the gas boiler module, includes a natural gas supply branch and a biomass gasification branch, with the two fuel lines connected to the fuel inlet via a three-way reversing valve.

[0011] Preferably, the high-pressure turbine inlet of the dual-shaft back-pressure steam turbine unit is connected to the main steam outlet of the gas-fired boiler via the main steam pipeline, the exhaust outlet is connected to the reheater inlet via the reheat pipeline, and the reheater outlet is connected to the low-pressure turbine inlet.

[0012] The low-pressure steam turbine exhaust port is divided into two paths. The first exhaust pipe is connected to the inlet of the steam-water heat exchanger, and the second exhaust pipe is connected to the inlet of the absorption chiller. Both pipes are equipped with electric regulating valves to regulate the exhaust flow distribution.

[0013] Preferably, the edge computing controller includes a load prediction module and an energy scheduling module. The load prediction module adopts a three-layer convolutional neural network model, and the model training uses the mean squared error loss function.

[0014] The energy dispatch module is built based on a model predictive control algorithm. The state variables include steam pressure, fuel flow rate, and energy storage device state of charge. The control variables are the opening degree of the fuel regulating valve, the opening degree of the turbine bypass valve, and the charging and discharging power of the energy storage device. The objective function is to minimize the system operating cost, which includes fuel cost and grid purchase and sale cost. The solution is obtained through rolling optimization.

[0015] Preferably, the edge computing controller further includes a safety protection module, which is set in the hard-wired protection circuit of the edge computing controller. The turbine shaft vibration monitoring adopts an eddy current sensor. When the vibration amplitude exceeds the preset amplitude value, the trip relay is triggered. The boiler drum water level monitoring adopts a three-out-of-two redundancy configuration. When the detection values ​​of any two liquid level sensors do not meet the preset liquid level, the fuel supply solenoid valve is directly cut off through hard wiring.

[0016] Preferably, it further includes: a composite energy storage module, comprising a water thermal storage tank and a lithium battery energy storage system; the water thermal storage tank adopts a vertical cylindrical structure, with a spiral coil heat exchanger installed inside the tank, the inlet connected to the return water pipe of the steam-water heat exchanger, the outlet connected to the water supply pipe of the heat user, an exhaust valve installed on the top of the tank, and a drain valve installed at the bottom of the tank; the lithium battery energy storage system consists of multiple battery clusters, connected to the AC bus through a bidirectional converter, and has the function of switching between PQ control and V / f control modes.

[0017] Preferably, the thermal storage tank and the energy storage system are controlled collaboratively through an energy management gateway. The gateway receives instructions from the intelligent control platform to adjust the flow rate of the thermal storage tank's circulating pump and the charging and discharging current of the energy storage system.

[0018] This invention provides a distributed energy operation method based on a gas-fired boiler, comprising:

[0019] Obtain the outlet steam pressure of the gas-fired boiler;

[0020] The steam pressure at the outlet of the gas boiler is extracted using a three-layer convolutional neural network model to obtain a load prediction curve. If the predicted heat load fluctuation is greater than the preset value, the water storage tank circulation pump is started in advance.

[0021] Fuel switching is based on real-time natural gas prices, and energy dispatch is based on energy storage management strategies.

[0022] Preferably, the energy dispatching based on the energy storage management strategy includes:

[0023] A multivariable state-space model is established, with steam pressure, fuel flow rate, and energy storage device state of charge as input variables, and electrical load, thermal load, and cooling load as output variables.

[0024] The rolling optimization problem is solved based on the model predictive control algorithm, and the optimization results are updated periodically.

[0025] When the steam pressure required by the heat user meets the preset conditions, it is controlled by the electric regulating valve and pressure matching device in conjunction with the exhaust pipe of the low-pressure steam turbine.

[0026] Preferably, the system also includes a variable load adaptive control mode. When the load change rate is less than or equal to the preset change rate, the PI control algorithm is used to adjust the fuel quantity; otherwise, the system switches to a fuzzy control algorithm. The input variables are the steam pressure deviation and its change rate, and the output variable is the fuel regulating valve opening increment.

[0027] The technical solution of the present invention has the following advantages over the prior art:

[0028] The distributed energy system and operation method based on a gas-fired boiler described in this invention employs a three-layer convolutional neural network model in the load prediction module of the intelligent control platform. It takes a 24×1 time-series vector composed of historical load data from 24 time points as input, performs feature extraction in the convolutional layers and a fully connected layer, and outputs predicted cooling, heating, and power load values ​​for the next 12 hours. Combined with an energy dispatch module based on a model predictive control algorithm, it optimizes state variables such as steam pressure and control variables such as fuel regulating valve opening, aiming to minimize system operating costs. This improves the accuracy of load prediction and the ability to perform multi-variable collaborative optimization, reducing steam venting. The low-pressure turbine exhaust port of the dual-shaft back-pressure turbine unit is divided into two paths, connected to a steam-water heat exchanger and an absorption chiller, respectively. The exhaust flow distribution is regulated by an electric regulating valve. Combined with the water storage tank and lithium battery energy storage system of the composite energy storage system, and coordinated control via an energy management gateway, it stores and releases heat during load fluctuations and charges and discharges during peak and off-peak periods of the power grid, widening the heat-to-power ratio adjustment range and reducing insufficient heating. Attached Figure Description

[0029] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein:

[0030] Figure 1 The present invention provides a structural diagram of a distributed energy system based on a gas-fired boiler.

[0031] Figure 2 This is a schematic diagram of the gas boiler module architecture provided by the present invention;

[0032] Figure 3 A schematic diagram of the dual-shaft back-pressure steam turbine unit architecture provided by the present invention;

[0033] Figure 4 This is a schematic diagram of the intelligent control platform architecture provided by the present invention. Detailed Implementation

[0034] The core of this invention is to provide a distributed energy system and operation method based on a gas-fired boiler, which effectively improves the accuracy of load forecasting and the ability of multi-variable collaborative optimization, and reduces insufficient heating.

[0035] To enable those skilled in the art to better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] Please refer to Figure 1 , Figure 1 The present invention provides a structural diagram of a distributed energy system based on a gas-fired boiler; the specific operation steps are as follows:

[0037] like Figure 2 As shown, the gas-fired boiler module includes a main furnace and an auxiliary furnace. The outlet of the main furnace is connected to the inlet of the auxiliary furnace via a flue gas duct. The auxiliary furnace has a built-in spiral tube sheet heat exchanger. The main furnace has a fuel inlet and a primary air inlet on its left side wall, a secondary air inlet on its right side wall, and a slag discharge port at its bottom. The auxiliary furnace has a chimney at the top and a feedwater pipe at the bottom connected to the deaerator outlet. A preheater and a main feedwater pump are connected in series on the feedwater pipe. The preheater is used to recover the waste heat from the flue gas in the auxiliary furnace to heat the boiler feedwater, and the main feedwater pump is used to pressurize and deliver the deaerated feedwater to the gas-fired boiler.

[0038] like Figure 3As shown, the dual-shaft back-pressure steam turbine unit includes a high-pressure steam turbine and a low-pressure steam turbine. The steam inlet of the high-pressure steam turbine is connected to the main steam outlet of the gas-fired boiler via a main steam pipeline, and the steam outlet is connected to the reheater inlet via a reheat pipeline. The reheater outlet is connected to the steam inlet of the low-pressure steam turbine. The steam outlet of the low-pressure steam turbine is divided into two paths: the first exhaust pipeline is connected to the inlet of the steam-water heat exchanger, and the second exhaust pipeline is connected to the inlet of the absorption chiller. Both pipelines are equipped with electric regulating valves to regulate the distribution of exhaust steam flow.

[0039] In one embodiment, the high-pressure turbine and the low-pressure turbine adopt a dual-shaft independent support structure. The high-pressure shaft system is equipped with tilting pad bearings, with 5 to 7 pads, and the bearing oil temperature is controlled at 40 to 45°C. The low-pressure shaft system is equipped with elliptical pad bearings, with a length-to-diameter ratio of 0.8 to 1.0 and a top clearance of 0.25 to 0.35 mm. A temperature sensor is installed at the exhaust port of the high-pressure turbine, with a range of 0 to 550°C and an accuracy of ±2°C. The signal is connected to an edge computing controller to calculate the heating power required by the reheater.

[0040] like Figure 4 As shown, the intelligent control platform includes an edge computing controller, an industrial switch, and a distributed sensor network. The sensor network includes a pressure transmitter installed on the main steam pipeline, with a measurement range of 0 to 16 MPa and an accuracy of ±0.1% FS, and a mass flow meter installed at the fuel inlet, with a range of 0 to 5000 m³ / h. 3 / h, with an accuracy of ±0.2%, and a flue gas temperature sensor installed in the auxiliary furnace, with a range of 0 to 600℃ and an accuracy of ±1℃. Each sensor is connected to an industrial switch via an RS485 bus and then transmitted to the edge computing controller via the Modbus protocol.

[0041] The edge computing controller includes:

[0042] The load forecasting module adopts a three-layer convolutional neural network model. The input layer receives a 24×1 time-series vector composed of historical load data from 24 time points. The first convolutional layer uses five 5×1 convolutional kernels for feature extraction with a stride of 1, and outputs a 5×20 feature vector after passing through the ReLU activation function. The second convolutional layer uses three 3×1 convolutional kernels with a stride of 1, and outputs a 3×18 feature vector. The third convolutional layer reduces the dimensionality through a 1×1 convolutional kernel. Finally, the fully connected layer outputs the predicted values ​​of cooling, heating and power loads for the next 12 hours. The model training uses the mean squared error loss function, and the prediction error is controlled within 4%.

[0043] The energy dispatch module is built based on the model predictive control algorithm. The prediction time domain is set to 4 hours, the control time domain is 1 hour, and the sampling period is 15 minutes. The state variables include steam pressure, fuel flow rate, and energy storage device state of charge. The control variables are fuel regulating valve opening, turbine bypass valve opening, and energy storage device charging and discharging power. The objective function is to minimize the system operating cost, including fuel cost and grid power purchase and sale cost. Through rolling optimization, the fuel regulating valve opening adjustment accuracy is ±0.5%, the turbine bypass valve opening adjustment accuracy is ±0.2%, and the energy storage device charging and discharging power resolution is 1kW.

[0044] The safety protection module is equipped with a hard-wired protection circuit independent of the edge computing controller. The turbine shaft vibration monitoring uses an eddy current sensor. When the vibration amplitude exceeds 120μm, it triggers the trip relay. The boiler drum water level monitoring adopts a three-out-of-two redundancy configuration. When the detection values ​​of any two liquid level sensors are lower than -100mm or higher than +80mm, the fuel supply solenoid valve is directly cut off through hard wiring.

[0045] A distributed energy system based on a gas-fired boiler also includes a fuel supply system comprising:

[0046] Natural gas supply branch: It consists of an LNG storage tank, a vaporizer, a pressure regulating valve group, and an electromagnetic shut-off valve. The vaporizer has a vaporization capacity of 500 to 1000 kg / h, the pressure regulating valve group has a pressure regulating range of 0.1 to 0.8 MPa, and the electromagnetic shut-off valve has a response time of ≤80 ms. It is connected to the fuel inlet of the main furnace through a stainless steel pipeline.

[0047] The biomass gasification branch includes a fluidized bed gasifier, a cyclone separator, and a wet desulfurization tower. The fluidized bed gasifier has an inner diameter of 1.8 to 2.2 m and a height of 6 to 8 m. The cyclone separator has a separation efficiency of ≥99%. The packing layer height of the wet desulfurization tower is 3 to 5 m. The gas outlet of the gasifier is connected to the inlet of the cyclone separator via a high-temperature pipeline. The outlet of the separator is connected in sequence to the desulfurization tower and the gas buffer tank. The buffer tank is connected to the fuel input port of the main furnace through a flashback preventer.

[0048] Two fuel lines are connected to the fuel inlet via a three-way reversing valve. The inlet of the three-way valve is connected to the natural gas pipeline and the biomass gas pipeline respectively, and the outlet is connected to the fuel inlet. The switching time of the three-way reversing valve is ≤25s.

[0049] The composite energy storage system includes:

[0050] Water storage tanks with a volume of 200 to 300 m³ 3 It adopts a vertical cylindrical structure, with a spiral coil heat exchanger installed inside the tank, and a heat exchange area of ​​150 to 200 m². 2 The inlet is connected to the return water pipe of the steam-water heat exchanger, and the outlet is connected to the water supply pipe of the heat user. An air vent valve is installed on the top of the tank, and a drain valve is installed at the bottom of the tank.

[0051] The lithium battery energy storage system consists of 20 to 30 battery clusters, with a capacity of 50kWh per cluster. It is connected to a 380V AC bus via a bidirectional converter. The bidirectional converter has a rated power of 200 to 300kW and a conversion efficiency of ≥96%. It has PQ control and V / f control mode switching functions.

[0052] The thermal storage tank and energy storage system are controlled collaboratively through an energy management gateway. The gateway receives commands from the intelligent control platform and adjusts the flow rate of the thermal storage tank's circulation pump, with a flow rate range of 0 to 100 m³ / h. 3 / h, and the charging and discharging current of the energy storage system, with a current range of 0 to 500A.

[0053] This embodiment provides a distributed energy system based on a gas-fired boiler. The load prediction module of the intelligent control platform employs a three-layer convolutional neural network model. It takes a 24×1 time-series vector composed of historical load data from 24 time points as input, performs feature extraction in the convolutional layers and a fully connected layer, and outputs predicted cooling, heating, and power load values ​​for the next 12 hours. Combined with an energy dispatch module based on a model predictive control algorithm, it optimizes state variables such as steam pressure and control variables such as fuel regulating valve opening, aiming to minimize system operating costs. This improves the accuracy of load prediction and the ability to perform multi-variable collaborative optimization, reducing steam venting. A dual-shaft back-pressure turbine unit uses a low-pressure turbine exhaust port with two paths, connected to a steam-water heat exchanger and an absorption chiller, respectively. The exhaust flow distribution is regulated by an electric regulating valve. Combined with a water storage tank and a lithium battery energy storage system in a composite energy storage system, and coordinated control via an energy management gateway, it stores and releases heat during load fluctuations and charges and discharges during peak and off-peak periods of the power grid, widening the heat-to-power ratio adjustment range and reducing insufficient heating.

[0054] Based on the above embodiments, this embodiment describes a distributed energy operation method based on a gas-fired boiler, as follows:

[0055] S1. Data Acquisition Stage: The gas boiler outlet steam pressure is acquired in real time through a distributed sensor network with a sampling period of 1 second, the biomass gasification furnace bed temperature with a sampling period of 5 seconds, and the heat user return water temperature with a sampling period of 30 seconds. The data is then uploaded to the intelligent control platform via an industrial switch.

[0056] S2. Load forecasting stage: The three-layer convolutional neural network model is used to extract features from the historical 72-hour load data to generate the load forecast curve for the next 12 hours. If the predicted heat load fluctuation is greater than 15%, the water storage tank circulation pump will be started 30 minutes in advance.

[0057] S3, Fuel Switching Control: When the natural gas price is > 3 yuan / m³ 3When the biomass fuel supply is sufficient, the intelligent control platform sends a switching command to the three-way reversing valve, simultaneously adjusting the gasifier feed rate (range 50 to 200 kg / h) and the primary air volume (range 1000 to 3000 m³ / h). 3 / h, to ensure that the main furnace combustion temperature is maintained at 1050 to 1150℃;

[0058] S4. Energy Storage Management Strategy: During the off-peak period of grid electricity prices from 23:00 to 7:00, if the lithium battery's state of charge is <70% and the system's active power is > user demand, charge it at 200kW to the state of charge = 90%. During the peak period from 10:00 to 15:00, if the system's active power is < user demand, discharge it at 150kW until the state of charge = 30%.

[0059] This includes the energy scheduling process, specifically:

[0060] A multivariable state-space model is established, with steam pressure, fuel flow rate, and energy storage device state of charge as input variables, and electrical load, thermal load, and cooling load as output variables.

[0061] The rolling optimization problem is solved based on the model predictive control algorithm. The optimization time domain is 4 hours and the control time domain is 1 hour. The optimization results are updated every 15 minutes.

[0062] When the steam pressure required by the heat user is 0.4 to 0.6 MPa, the electric regulating valve and the pressure matching device of the low-pressure steam turbine exhaust pipeline are used for joint control. The electric regulating valve has an adjustment range of 20% to 80%, and the pressure matching device has a pressure boost ratio of 1.2 to 1.5 to ensure that the steam supply pressure fluctuation is ≤ ±0.03 MPa.

[0063] The system is set to variable load adaptive control mode:

[0064] When the load change rate is ≤5% / min, the fuel quantity is adjusted using a PI control algorithm with a proportional coefficient of 0.5 to 1.0 and an integral time of 10 to 20 seconds.

[0065] When the load change rate is greater than 5% / min, the system switches to a fuzzy control algorithm. The input variables are the steam pressure deviation and its rate of change. The steam pressure deviation ranges from -0.3 to +0.3 MPa, and the rate of change ranges from -0.1 to +0.1 MPa / s. The output variable is the fuel regulating valve opening increment, with an increment range of -10% to +10%. The fuzzy rule base contains 49 control rules.

[0066] During load regulation, the exhaust temperature of the high-pressure steam turbine is maintained 50 to 80°C higher than the saturation temperature to prevent water from entering the blades.

[0067] Based on the above embodiments, this embodiment provides a specific solution to describe a distributed energy system based on a gas-fired boiler, as follows:

[0068] Example 1:

[0069] The gas-fired boiler module includes a main furnace and an auxiliary furnace. The main furnace has a fuel inlet and a primary air inlet on the left side wall, a secondary air inlet on the right side wall, and a slag discharge port at the bottom. The main furnace outlet is connected to the auxiliary furnace inlet via a flue gas duct. The auxiliary furnace has a built-in spiral tube sheet heat exchanger. A chimney is installed at the top of the auxiliary furnace, and the bottom is connected to the deaerator outlet via a feedwater pipe. A preheater and a main feedwater pump are connected in series on the feedwater pipe. The preheater is used to recover the waste heat of the flue gas in the auxiliary furnace to heat the boiler feedwater, and the main feedwater pump is used to pressurize and deliver the deaerated feedwater to the gas-fired boiler.

[0070] The main furnace adopts a specific structural design, and the layout of the fuel inlet, primary air inlet, and secondary air inlet ensures complete combustion of fuel; the ash discharge port is connected to the ash discharge equipment, which facilitates the cleaning of the residue after combustion; the design of the flue gas duct ensures that the flue gas flows smoothly to the auxiliary furnace, and the spiral tube plate heat exchanger realizes heat exchange in the auxiliary furnace, improving energy utilization.

[0071] The chimney of the auxiliary furnace is used to discharge the treated flue gas, and the connection of the feedwater pipeline ensures the recycling of boiler feedwater; the configuration of the preheater and the main feedwater pump effectively recovers the waste heat of the flue gas and provides sufficient pressure for the feedwater, thereby improving the overall thermal efficiency of the system.

[0072] Example 2:

[0073] The dual-shaft back-pressure steam turbine unit includes a high-pressure steam turbine and a low-pressure steam turbine. The steam inlet of the high-pressure steam turbine is connected to the main steam outlet of the gas-fired boiler through the main steam pipeline, and the steam outlet is connected to the inlet of the reheater through the reheat pipeline. The reheater outlet is connected to the steam inlet of the low-pressure steam turbine. The steam outlet of the low-pressure steam turbine is divided into two paths. The first steam outlet is connected to the inlet of the steam-water heat exchanger, and the second steam outlet is connected to the inlet of the absorption chiller. Both paths are equipped with electric regulating valves to regulate the steam flow distribution.

[0074] The high-pressure steam turbine and the low-pressure steam turbine adopt a dual-shaft independent support structure, and are equipped with different types of bearings to ensure stable operation of the shaft system; the bearing oil temperature is controlled within a specific range to ensure normal operation of the steam turbine; the connection between the main steam pipeline and the reheat pipeline must meet the steam pressure and temperature requirements to ensure effective transfer of steam energy.

[0075] The two pipelines at the exhaust port of the low-pressure steam turbine achieve dynamic flow distribution through electric regulating valves. Based on the demand for heat and cold loads, the exhaust steam flow direction is intelligently adjusted to improve the overall energy utilization rate of the system. The high-precision regulation capability of the electric regulating valves ensures accurate exhaust steam flow control and meets the energy demand under different operating conditions.

[0076] Example 3:

[0077] The intelligent control platform comprises an edge computing controller, an industrial switch, and a distributed sensor network. The sensor network includes a pressure transmitter installed on the main steam pipeline, with a measurement range of 0 to 16 MPa and an accuracy of ±0.1% FS, and a mass flow meter installed at the fuel inlet, with a range of 0 to 5000 m³ / h. 3 / h, with an accuracy of ±0.2%, and a flue gas temperature sensor installed in the auxiliary furnace, with a range of 0 to 600℃ and an accuracy of ±1℃. Each sensor is connected to an industrial switch via an RS485 bus, and then transmitted to the edge computing controller via the Modbus protocol.

[0078] The deployment of the sensor network must ensure accurate data acquisition from each measuring point. Pressure transmitters, mass flow meters, and flue gas temperature sensors are installed in key locations to monitor system operating parameters in real time. The configuration of RS485 bus and industrial switch ensures stable data transmission to the edge computing controller, providing real-time data support for system control.

[0079] The control algorithm includes a load forecasting module, an energy dispatching module, and a safety protection module. The load forecasting module uses a three-layer convolutional neural network model, extracts features from historical load data, and outputs future load forecasts. The energy dispatching module is based on a model predictive control algorithm, aiming to minimize system operating costs and achieve multi-variable collaborative optimization. The safety protection module is equipped with an independent hard-wired protection circuit to ensure rapid system response in abnormal situations and protect equipment safety.

[0080] Example 4:

[0081] The fuel supply system includes a natural gas supply branch and a biomass gasification branch. Both fuel lines are connected to the fuel inlet via a three-way reversing valve. The natural gas supply branch consists of an LNG storage tank, a vaporizer, a pressure regulating valve group, and an electromagnetic shut-off valve. The vaporizer has a vaporization capacity of 500 to 1000 kg / h, the pressure regulating valve group has a pressure regulation range of 0.1 to 0.8 MPa, and the electromagnetic shut-off valve has a response time of ≤80 ms. It is connected to the fuel inlet of the main furnace via a stainless steel pipeline.

[0082] The biomass gasification branch includes a fluidized bed gasifier, a cyclone separator, and a wet desulfurization tower. The fluidized bed gasifier has an inner diameter of 1.8 to 2.2 m and a height of 6 to 8 m. The cyclone separator has a separation efficiency of ≥99%. The packing layer height of the wet desulfurization tower is 3 to 5 m. The gas outlet of the gasifier is connected to the inlet of the cyclone separator via a high-temperature pipeline. The outlet of the separator is connected in sequence to the desulfurization tower and the gas buffer tank. The buffer tank is connected to the fuel input port of the main furnace through a flashback preventer.

[0083] The switching time of the three-way reversing valve is ≤25s, enabling rapid switching between natural gas and biomass gas. During the switching process, the gasifier feed rate and primary air volume need to be adjusted synchronously to ensure that the combustion temperature of the main furnace is maintained within a stable range. Through the dual-fuel supply design, the system can flexibly switch according to fuel prices and supply conditions, reducing operating costs and improving fuel adaptability.

[0084] Example 5:

[0085] The integrated energy storage system comprises a water-based thermal storage tank and a lithium-ion battery energy storage system, which are coordinated and controlled through an energy management gateway; the water-based thermal storage tank has a volume of 200 to 300 m³. 3 It adopts a vertical cylindrical structure, with a spiral coil heat exchanger installed inside the tank, and a heat exchange area of ​​150 to 200 m². 2 The inlet is connected to the return water pipe of the steam-water heat exchanger, and the outlet is connected to the water supply pipe of the heat user. An air vent valve is installed on the top of the tank, and a drain valve is installed at the bottom of the tank.

[0086] The lithium battery energy storage system consists of 20 to 30 battery clusters, each with a capacity of 50kWh. It is connected to a 380V AC bus via a bidirectional converter with a rated power of 200 to 300kW and a conversion efficiency of ≥96%. It also features PQ control and V / f control mode switching.

[0087] The energy management gateway receives instructions from the intelligent control platform to adjust the flow rate of the circulating pump in the thermal storage tank and the charging and discharging current of the energy storage system. During off-peak hours, the lithium battery energy storage system charges according to the state of charge and the system's active power. During peak hours, it discharges according to the system's active power demand. The water thermal storage tank starts the circulating pump in advance to regulate the heat according to the predicted fluctuations in heat load, thereby realizing the storage and release of heat energy and improving the system's energy storage and peak-shaving capabilities.

[0088] Example 6:

[0089] The system operation method includes steps such as data acquisition, load forecasting, fuel switching control, and energy storage management strategy. In the data acquisition stage, data such as the steam pressure at the outlet of the gas boiler, the bed temperature of the biomass gasification furnace, and the return water temperature of the heat user are acquired in real time through a distributed sensor network. The data is uploaded to the intelligent control platform via an industrial switch. The sampling period is set to 1s, 5s, and 30s according to different parameter requirements.

[0090] During the load forecasting phase, a three-layer convolutional neural network model is used to extract features from the historical 72-hour load data to generate a load forecast curve for the next 12 hours. If the predicted heat load fluctuation is greater than 15%, the water storage tank circulation pump is started 30 minutes in advance to pre-adjust the heat load fluctuation.

[0091] Fuel switching control is triggered based on natural gas prices and biomass fuel supply. When the natural gas price is greater than 3 yuan / m³... 3 When the biomass fuel supply is sufficient, the intelligent control platform sends a switching command to the three-way reversing valve, and at the same time adjusts the gasifier feeding speed and primary air volume to ensure stable combustion temperature in the main furnace.

[0092] The energy storage management strategy controls the charging and discharging power of the lithium battery energy storage system based on the division of the grid electricity price into valley and peak periods. It charges to a specific state of charge during valley periods and discharges during peak periods to meet user needs, thereby optimizing the management of the system's active power and reducing electricity costs.

[0093] Example 7:

[0094] The energy scheduling process of the intelligent control platform includes establishing a multivariable state-space model, with input variables being steam pressure, fuel flow rate, and energy storage device state of charge, and output variables being electrical load, thermal load, and cooling load; solving the rolling optimization problem based on the model predictive control algorithm, with an optimization time domain of 4 hours and a control time domain of 1 hour, and updating the optimization results every 15 minutes.

[0095] When the steam pressure required by the heat user is 0.4 to 0.6 MPa, the electric regulating valve and the pressure matching device of the low-pressure steam turbine exhaust pipeline are used for joint control. The electric regulating valve has an adjustment range of 20% to 80%, and the pressure matching device has a pressure boost ratio of 1.2 to 1.5 to ensure that the steam supply pressure fluctuation is ≤ ±0.03 MPa.

[0096] The system is set to a variable load adaptive control mode. When the load change rate is ≤5% / min, the PI control algorithm is used to adjust the fuel quantity, and the proportional coefficient and integral time are set within a specific range. When the load change rate is >5% / min, the system switches to a fuzzy control algorithm. Based on the steam pressure deviation and its rate of change, the system outputs the fuel regulating valve opening increment. During load regulation, the high-pressure turbine exhaust temperature is maintained 50 to 80°C higher than the saturation temperature to prevent water from entering the blades and ensure stable operation of the system under different load change rates.

[0097] In summary, this distributed energy system and its operation method based on a gas-fired boiler employs a three-layer convolutional neural network model in the load prediction module of the intelligent control platform. It takes a 24×1 time-series vector composed of historical load data from 24 time points as input, performs feature extraction in the convolutional layers and a fully connected layer, and outputs predicted cooling, heating, and power load values ​​for the next 12 hours. Combined with the energy dispatch module's model-based predictive control algorithm, and with the goal of minimizing system operating costs, it performs rolling optimization on state variables such as steam pressure and control variables such as fuel regulating valve opening, thereby improving load prediction accuracy and multi-variable collaborative optimization capabilities, and reducing steam venting.

[0098] Furthermore, this distributed energy system and its operation method based on a gas-fired boiler, where the low-pressure turbine exhaust port of the dual-shaft back-pressure turbine unit is split into two paths, connected to a steam-water heat exchanger and an absorption chiller respectively, and the exhaust flow distribution is regulated by an electric regulating valve. Combined with the water storage tank and lithium battery energy storage system of the composite energy storage system, and coordinated control by the energy management gateway, it stores and releases heat during heat load fluctuations and charges and discharges during peak and off-peak periods of the power grid, thus widening the heat-to-power ratio adjustment range and reducing insufficient heating. This solves the problems of existing technologies that rely on empirical formulas for load forecasting, cannot accurately capture the dynamic changes of cooling, heating, and power loads, lack multi-variable collaborative optimization capabilities in energy dispatch strategies, have limited heat-to-power ratio adjustment range, are prone to insufficient heating or excess steam venting when the user's heat load is too high, and have insufficient capacity to absorb renewable energy sources such as wind and solar, making it difficult to meet the needs of low-carbon energy transition.

[0099] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0100] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0101] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A distributed energy system based on a gas-fired boiler, characterized in that, include: The gas-fired boiler module includes a main furnace and an auxiliary furnace. The outlet of the main furnace is connected to the inlet of the auxiliary furnace via a flue gas duct. The auxiliary furnace has a built-in spiral tube sheet heat exchanger. The main furnace has a fuel inlet and a primary air inlet on its left side wall, a secondary air inlet on its right side wall, and a slag discharge port at its bottom. The auxiliary furnace has a chimney at the top and a feedwater pipe at the bottom connected to the deaerator outlet. A preheater and a main feedwater pump are connected in series on the feedwater pipe. The preheater is used to recover waste heat from the flue gas in the auxiliary furnace to heat the boiler feedwater, and the main feedwater pump is used to pressurize and deliver the deaerated feedwater to the gas-fired boiler. A dual-shaft back-pressure steam turbine unit, connected to the gas-fired boiler module, includes a high-pressure steam turbine and a low-pressure steam turbine, used to regulate the distribution of exhaust steam flow; The intelligent control module, connected to the gas boiler module and the dual-shaft back-pressure turbine assembly, includes an edge computing controller, an industrial switch, and a distributed sensor network, used to acquire data and perform calculations.

2. The distributed energy system based on a gas-fired boiler according to claim 1, characterized in that, Also includes: The fuel supply module, connected to the gas boiler module, includes a natural gas supply branch and a biomass gasification branch, with the two fuel lines connected to the fuel inlet via a three-way reversing valve.

3. The distributed energy system based on a gas-fired boiler according to claim 1, characterized in that, The high-pressure turbine inlet of the dual-shaft back-pressure steam turbine unit is connected to the main steam outlet of the gas-fired boiler via the main steam pipeline, the exhaust outlet is connected to the reheater inlet via the reheat pipeline, and the reheater outlet is connected to the low-pressure turbine inlet. The low-pressure steam turbine exhaust port is divided into two paths. The first exhaust pipe is connected to the inlet of the steam-water heat exchanger, and the second exhaust pipe is connected to the inlet of the absorption chiller. Both pipes are equipped with electric regulating valves to regulate the exhaust flow distribution.

4. The distributed energy system based on a gas-fired boiler according to claim 1, characterized in that, The edge computing controller includes a load prediction module and an energy scheduling module. The load prediction module adopts a three-layer convolutional neural network model, and the model training uses the mean squared error loss function. The energy dispatch module is built based on a model predictive control algorithm. The state variables include steam pressure, fuel flow rate, and energy storage device state of charge. The control variables are the opening degree of the fuel regulating valve, the opening degree of the turbine bypass valve, and the charging and discharging power of the energy storage device. The objective function is to minimize the system operating cost, which includes fuel cost and grid purchase and sale cost. The solution is obtained through rolling optimization.

5. The distributed energy system based on a gas-fired boiler according to claim 4, characterized in that, The edge computing controller also includes a safety protection module, which is set in the hard-wired protection circuit of the edge computing controller. The turbine shaft vibration monitoring adopts an eddy current sensor. When the vibration amplitude exceeds the preset amplitude value, the trip relay is triggered. The boiler drum water level monitoring adopts a three-out-of-two redundancy configuration. When the detection values ​​of any two liquid level sensors do not meet the preset liquid level, the fuel supply solenoid valve is directly cut off through hard wiring.

6. The distributed energy system based on a gas-fired boiler according to claim 1, characterized in that, Also includes: The composite energy storage module includes a water thermal storage tank and a lithium battery energy storage system. The water thermal storage tank adopts a vertical cylindrical structure, with a spiral coil heat exchanger installed inside. The inlet is connected to the return water pipe of the steam-water heat exchanger, and the outlet is connected to the water supply pipe of the heat user. An exhaust valve is installed on the top of the tank, and a drain valve is installed at the bottom of the tank. The lithium battery energy storage system consists of multiple battery clusters and is connected to the AC bus through a bidirectional converter. It has the function of switching between PQ control and V / f control modes.

7. The distributed energy system based on a gas-fired boiler according to claim 6, characterized in that, The thermal storage tank and the energy storage system are controlled collaboratively through an energy management gateway. The gateway receives instructions from the intelligent control platform to adjust the flow rate of the thermal storage tank's circulating pump and the charging and discharging current of the energy storage system.

8. A distributed energy operation method based on a gas-fired boiler, characterized in that, include: Obtain the outlet steam pressure of the gas-fired boiler; The steam pressure at the outlet of the gas boiler is extracted using a three-layer convolutional neural network model to obtain a load prediction curve. If the predicted heat load fluctuation is greater than the preset value, the water storage tank circulation pump is started in advance. Fuel switching is based on real-time natural gas prices, and energy dispatch is based on energy storage management strategies.

9. The distributed energy operation method based on a gas-fired boiler according to claim 8, characterized in that, The energy dispatching based on the energy storage management strategy includes: A multivariable state-space model is established, with steam pressure, fuel flow rate, and energy storage device state of charge as input variables, and electrical load, thermal load, and cooling load as output variables. The rolling optimization problem is solved based on the model predictive control algorithm, and the optimization results are updated periodically. When the steam pressure required by the heat user meets the preset conditions, it is controlled by the electric regulating valve and pressure matching device in conjunction with the exhaust pipe of the low-pressure steam turbine.

10. The distributed energy operation method based on a gas-fired boiler according to claim 8, characterized in that, It also includes a system setting for a variable load adaptive control mode. When the load change rate is less than or equal to the preset change rate, the PI control algorithm is used to adjust the fuel quantity; otherwise, it switches to a fuzzy control algorithm. The input variables are the steam pressure deviation and its change rate, and the output variable is the fuel regulating valve opening increment.

Images