Distributed fuel cell power generation system based on solid hydrogen storage

Abstract

The invention belongs to the technical field of fuel cells, and particularly relates to a distributed fuel cell power generation system based on solid hydrogen storage, which comprises a box body, a fuel cell module, a gas supply module, an output module and a control module, the invention provides a novel solid hydrogen storage bottle structure which is composed of a heat insulation layer, a composite phase change material filling area and a hydrogen storage area, bionic heat conduction fins are designed between the composite phase change material filling area and the hydrogen storage area, and a high-efficiency hydrogen release angle of a solid hydrogen storage bottle is designed. The control module predicts required power according to a deep neural network through collected data, obtains the relation between the required power and the hydrogen and oxygen flow through building a model, and achieves precise control over the gas flow through sliding mode control and PID control so as to ensure that the power can be matched in real time.

Description

Technical Field

[0001] The present invention belongs to the field of fuel cells, and in particular relates to a distributed fuel cell power generation system based on solid-state hydrogen storage. Background Art

[0002] As a clean secondary energy carrier, hydrogen can be easily converted into electricity and heat with high conversion efficiency and multiple sources. Using renewable energy to achieve large-scale hydrogen production can provide a hydrogen source for fuel cells. Fuel cells convert hydrogen and oxygen into water through electrochemical reactions and directly convert the chemical energy of the fuel into electrical energy. Fuel cells have obvious advantages in energy conversion efficiency. Traditional fossil energy conversion efficiency is low, and fuel cells can achieve an efficiency of more than 50% in energy conversion, which is much higher than traditional thermal power generation. This means that fuel cells can provide more electricity with the same energy input, effectively improve energy utilization efficiency, and reduce energy waste. This is of great significance at a time when global resources are scarce. Fuel cells have the characteristics of zero emissions. Fuel cells only produce water when working and do not emit polluting gases. In addition, fuel cells have low noise pollution and long service life.

[0003] Compared with other fuel cells, proton exchange membrane fuel cells have the advantages of high energy conversion efficiency, low operating temperature, air as oxidant, non-corrosive electrolyte, fast dynamic response speed, environmentally friendly by-products, noiseless operation, and recyclable energy. Therefore, they are the most widely used. They are currently used in the power systems of vehicles such as automobiles and ships, and have been widely used in fixed and portable power sources.

[0004] At present, the fuel cell power generation system based on solid-state hydrogen storage does not fully consider the structure of the solid-state hydrogen storage bottle, the placement of the solid-state hydrogen storage bottle in the fuel cell, the control strategy, etc.; for example, Chinese patent document CN 114284536A discloses a portable power source based on solid-state hydrogen storage and fuel cell, wherein the portable power source integrates the fuel cell and the hydrogen storage alloy in contact with each other in a shell, and in order to reduce the volume and realize gas-liquid separation, the hydrogen storage alloy is embedded with a gas-liquid separation pipeline connected to the tail exhaust port of the fuel cell, and the water vapor discharged from the tail of the fuel cell is liquefied and stored in the drainage pipeline due to the heat absorption of the hydrogen storage alloy, and the air discharged from the tail of the fuel cell is discharged through the heat exchange pipeline embedded in the hydrogen storage alloy, and a fan is used to provide air for the fuel cell and power for the heat exchange cycle in the shell, and the heat in the tail exhaust air and the heat of the fuel cell body are The hydrogen storage alloy absorbs and releases hydrogen and plays a role in cooling the fuel cell, while the hydrogen released by the hydrogen storage alloy is used for power generation of the fuel cell, thus achieving the purpose of integrating the fuel cell and the hydrogen storage alloy into a portable power source; the above-mentioned fuel cell power generation system based on solid-state hydrogen storage does not use the prediction of demand power to achieve the purpose of early prediction and timely response to the demand power, the gas supply system of the fuel cell is not fully considered, and the structure of the solid-state hydrogen storage bottle is not described in detail; therefore, technical personnel in this field are committed to developing a fuel cell power generation system with stable gas supply, excellent control strategy, stable and reliable solid-state hydrogen storage bottle, high energy conversion efficiency, and strong endurance. Summary of the invention

[0005] The patent of this invention provides a portable fuel cell power generation system based on solid-state hydrogen storage. The invention combines the fuel cell with solid-state hydrogen storage and control strategy to achieve the goals of stable gas supply, excellent control strategy, stable and reliable solid-state hydrogen storage bottle, high energy conversion efficiency and strong endurance.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solutions: a distributed fuel cell power generation system based on solid-state hydrogen storage, including a box, a fuel cell module, a gas supply module, an output module and a control module.

[0007] Furthermore, the gas supply module is connected to the fuel cell module to provide hydrogen and oxygen to the fuel cell module; the output module is connected to the fuel cell module to output the electric energy generated by the fuel cell in the form of 24V direct current and 220V or 380V alternating current.

[0008] Furthermore, the control module is connected to the gas supply module and the output module. The control module collects current and voltage data of the output module, predicts future power requirements through a deep neural network, and accurately controls the hydrogen and oxygen supply of the gas supply module in advance, thereby solving the problem of slow response of the fuel cell and being able to match the power requirements in real time.

[0009] Furthermore, the gas supply module includes a hydrogen gas supply module and an oxygen gas supply module. The hydrogen gas supply module includes a solid hydrogen storage bottle, a mass flow meter, a flow valve and a tail exhaust valve. The oxygen gas supply module includes an air compressor, a mass flow meter, a cooling humidifier and a throttle.

[0010] Furthermore, the output module includes a DC-DC and an inverter. The DC-DC is used to convert the electric energy output by the fuel cell into 24V direct current through rectification and voltage reduction. The inverter is used to convert the electric energy output by the fuel cell into 220V or 380V alternating current.

[0011] Furthermore, the air compressor is connected to a cooling and humidifying device, which is connected to the cathode inlet of the fuel cell. The cooling and humidifying device cools and humidifies the air delivered by the air compressor to reduce the temperature of the fuel cell, ensure that the fuel cell operates at a suitable temperature, prevent heat accumulation inside the fuel cell from damaging the proton exchange membrane and the catalyst layer, and keep the proton exchange membrane at an appropriate humidity to reduce the ohmic polarization loss of the fuel cell; the cathode outlet of the fuel cell is connected to a throttle to adjust the cathode pressure level.

[0012] Furthermore, the solid-state hydrogen storage bottle has a hydrogen storage pressure of 2 MPa, a maximum tolerable pressure of 4.5 MPa, and can store 200 to 250 g of hydrogen.

[0013] Furthermore, there can be 8 to 16 bottles of the solid-state hydrogen storage bottles in one box; the solid-state hydrogen storage bottles are connected in parallel, and hydrogen can be supplied by a single hydrogen bottle or by multiple hydrogen bottles at the same time. Each solid-state hydrogen storage bottle is connected to a flow valve at the output interface, which is used to freely control the size of the hydrogen flow rate output by each solid-state hydrogen storage bottle. When providing hydrogen for the fuel cell, the control module can freely control the number of hydrogen supply bottles and the size of the output hydrogen flow rate of each bottle by controlling the opening of the flow valve according to the predicted required power through the functional relationship between the hydrogen flow rate and the power; the flow valve is connected to a mass flow meter for real-time measurement of the actual output flow rate of each solid-state hydrogen storage bottle, and the mass flow meter is connected to the anode inlet of the fuel cell for transmitting the measured hydrogen to the fuel cell; the anode outlet of the fuel cell is connected to the tail exhaust valve for draining water and discharging nitrogen that penetrates from the cathode to the anode.

[0014] Furthermore, when the solid-state hydrogen storage bottle is placed in the box and is in a hydrogen release state, the axial direction of the output interface of the solid-state hydrogen storage bottle should be θ

[0015]

[0016] In the formula, Q is the hydrogen flow rate collected by the mass flowmeter, T is the time interval between each collection, and S is the total mass of hydrogen stored in the solid hydrogen storage bottle.

[0017] Furthermore, the solid-state hydrogen storage bottle is composed of three components: a heat insulation layer, a composite phase change material filling area, and a hydrogen storage area.

[0018] Furthermore, the heat insulating layer serves to isolate heat loss and fix the phase change material and the lithium battery.

[0019] Furthermore, the hydrogen storage area is approximately a cylinder with a cross-sectional radius of R, and is filled with a low-temperature solid hydrogen storage alloy LaNi5.

[0020] Furthermore, the solid-state hydrogen storage bottle absorbs heat during the dehydrogenation process and releases heat during the hydrogenation process, stores the heat released during the hydrogenation process through the composite phase change material and releases the heat during the dehydrogenation process; the composite phase change material is composed of lauric acid, oleic acid and capric acid as the original phase change material, and the three are mixed in proportion according to the lowest eutectic principle to obtain the composite phase change material; when the dehydrogenation temperature of the solid-state hydrogen storage bottle reaches 28-34°C, the maximum dehydrogenation flow rate can be obtained. Continuously release hydrogen; therefore, when the mass ratio of lauric acid, oleic acid and capric acid is 31.07-37.68:50.61-54.29:11.71-14.64, the solidification point of the composite material can be guaranteed to be lower than 28°C; the hydrogen stored in the solid hydrogen storage bottle is released at the maximum hydrogen flow rate The mass of the composite phase change material required to be filled when all the heat generated by the composite phase change material is completely absorbed is 5.24kg to 6.47kg.

[0021] Furthermore, a bionic heat-conducting fin based on a parallel vein structure is designed on the contact surface between the composite phase change material filling area and the hydrogen storage area to improve the heat exchange efficiency between the composite phase change material filling area and the hydrogen storage area.

[0022] Furthermore, a bionic heat-conducting fin based on the jade lotus leaf vein structure is designed on the wall between the composite phase change material filling area and the hydrogen storage area to improve the heat exchange efficiency between the composite phase change material filling area and the hydrogen storage area and maintain the thermal balance between the two; the bionic heat-conducting fins are evenly distributed around the wall between the composite phase change material filling area and the hydrogen storage area, and the leaf vein length in the heat-conducting fin structure is L k , leaf vein width is W k , the branch angle is α k, where k represents the number of leaf veins, L1 and W1 are the initial length and width of the first main leaf vein and the first branching angle α2 of the leaf vein structure; let the ratio of the length of the k+1th leaf vein to the length of the kth leaf vein be the length ratio L θ , Let the ratio of the width of the k+1th leaf vein to the width of the kth leaf vein be the length ratio W θ , Let the ratio of the k+1th level leaf vein branch angle to the kth level leaf vein branch angle be the length ratio α θ , Inside the hydrogen storage area, the main leaf veins are extended into the hydrogen storage area with an integrated wavy heat-conducting fin to increase the contact area of ​​the solid hydrogen storage alloy. The length is L0 and the width is W0. The function expression is: When L1=20mm, W1=2mm, α2=60°, and L θ ∈(0.68,0.74),W θ ∈(0.42,0.45),α θ When ∈(0.54,0.59), the thermal fins have better thermal conductivity.

[0023] Furthermore, the control module comprises a data acquisition module, a data analysis module and a control execution module based on the control system;

[0024] S1: The data acquisition module collects various data used by the distributed power generation system in real time, and transmits the collected data to the data analysis module;

[0025] S2: The data analysis module processes the collected data and extracts features, builds a deep neural network model to predict the required power of the power generation system and transmits the results to the control execution module;

[0026] S3: The control execution module automatically executes control measures according to the predicted demand power, and controls the output power of the fuel cell to match the changing demand power in advance.

[0027] Furthermore, in step S1, the data acquisition module acquires the current signal I and the current differential signal dI as well as the voltage signal V and the voltage differential signal dV of the controlled fuel cell power generation system; the current signal I and the voltage signal V are obtained by collecting and differentially extracting the output current of the power generation system after the Hall current sensor and the voltage sensor perform noise filtering; the current differential signal dI and the voltage differential signal dV are obtained by performing differential signal extraction on the extracted current signal I and voltage signal V by the tracking differentiator; the acquisition frequency f is set, and the data acquisition module collects the above data in real time according to the set frequency and transmits it to the data analysis module.

[0028] Furthermore, in step S2, the incoming data is normalized and features are extracted; the voltage signal V, the current signal I, the voltage differential signal dV, and the current differential signal dI are used as input vectors, the input vectors are input into the deep neural network, a deep neural network model is obtained, and the predicted demand power P is output. N .

[0029] Furthermore, the deep neural network model includes an input layer, a hidden layer and an output layer, and each layer is fully connected, wherein the output strategy formula for the jth neuron in the lth layer is:

[0030]

[0031] In the formula, is the output of the jth neuron in the lth layer, n l-1 is the number of neurons in the l-1th layer, is the weight from the kth neuron in the l-1th layer to the jth neuron in the lth layer, The output value of the kth neuron in the l-1th layer, is the bias of the jth neuron in the lth layer, σ is the activation function, and in this deep neural network model, the input layer and hidden layer use the Relu function, and the output layer uses the Sigmoid function.

[0032] Furthermore, the original data set of 1,000 sets of working condition data obtained was divided into a training data set and a test data set at a ratio of 8:2; the training data set was trained 8,000 to 10,000 times, hoping that the deep neural network could accurately learn and predict the changes in demand power. After the model training is completed, the remaining test data set will be used to test the deep neural network to verify its accuracy in predicting the changes in demand power.

[0033] Furthermore, in step S3, a model is established to analyze the relationship between the required power and the gas mass flow rate of the positive and negative electrodes of the fuel cell, and the flow rates of the two electrodes are precisely controlled through a control algorithm.

[0034] Further, S31, the execution module establishes an output power model of the fuel cell;

[0035] In this power generation system, the number of fuel cells is 1000, so the function expression of the output current and the output power of the fuel cells of this power generation system is approximately:

[0036]

[0037] Where P N is the required power predicted by the deep neural network in step S2, I N is the current value corresponding to the required power;

[0038] Through the functional relationship between the oxygen mass flow rate and current required by the fuel cell cathode and the air mass flow rate input to the cathode by the air compressor in the power generation system, it can be obtained that the power generation system at power P N When the mass flow rate of air input to the air compressor is Q a,N (kg / s) is:

[0039]

[0040] Through the relationship between the oxygen mass flow rate and current required by the fuel cell cathode, it can be obtained that the power generation system has a power of P N When the mass flow rate of air input to the air compressor is Q H,N (kg / s) is:

[0041]

[0042] Further, S32, according to S31, the anode required hydrogen mass flow Q is calculated H,N Dynamically control the opening V of the flow valve of the solid hydrogen storage bottle and the number of hydrogen bottles N H ; Each solid-state hydrogen storage bottle sets the maximum hydrogen discharge flow rate through a rectifier valve So when the predicted power is P N hour:

[0043] S321, if but The flow valve opening of each hydrogen bottle is the maximum V max , the hydrogen discharge flow rate is

[0044] S322, if but Top N H - The flow valve opening of one hydrogen bottle is the maximum V max , the hydrogen discharge flow rate is Then the Nth H The hydrogen flow rate that the hydrogen tank needs to output

[0045] The execution module establishes a sliding mode control model and combines the real-time feedback of hydrogen flow to calculate the hydrogen flow deviation e H To control the Nth H The opening V of the flow valve of the first hydrogen bottle is controlled to achieve H Accurate control of the hydrogen flow rate of each hydrogen bottle; according to the hydrogen flow deviation e H Construct the switching function s. The expression of the switching function is:

[0046]

[0047] Then the derivative of the switching function s is:

[0048]

[0049] In order to make the system approach stability, the system approach rate is set to exponential approach:

[0050]

[0051] Among them, sat(s) is a continuous function, which is used to prevent the vibration of the opening V of the hydrogen bottle flow valve and make it converge smoothly. The expression of sat(s) is:

[0052]

[0053] Further, S33, according to the cathode required air mass flow Q calculated in S31 a,N Dynamically control the speed n of the air compressor; the execution module establishes a PID model to control the speed n of the air compressor by calculating the air flow deviation Δe in combination with the real-time feedback of the air flow, thereby achieving precise control of the air flow; use the PID algorithm for feedback control to convert the air mass flow measurement value Q a,a and the target value Q a,N The proportion (P), integral (I) and differential (D) of the deviation Δe are linearly combined to form the control quantity; the expression of the PID algorithm feedback control flow valve opening is:

[0054]

[0055] The beneficial effects of the present invention are:

[0056] 1. The present invention provides a dynamic hydrogen release angle that changes with the total amount of hydrogen released when the solid-state hydrogen storage bottle is in operation, thereby improving the hydrogen release efficiency of the solid-state hydrogen storage bottle;

[0057] 2. The present invention provides a sandwich-type solid-state hydrogen storage bottle structure, in which a phase change material is installed in the interlayer for absorbing the heat released by the hydrogen storage material when the solid-state hydrogen storage bottle is hydrogenated, and the heat is retransmitted to the hydrogen storage material when the solid-state hydrogen storage bottle is working, thereby improving energy utilization; and a simulated heat dissipation fin based on the jade lotus leaf veins is designed on the wall surface between the hydrogen storage area and the phase change material area to improve the thermal conductivity efficiency.

[0058] 3. The present invention provides a predictive control method for a distributed power generation system. By reading the current, voltage and their differential signals, the required power is predicted through a deep neural network model. The required amount of hydrogen and oxygen is then obtained using a mathematical model of the required power and hydrogen and oxygen. The output of hydrogen and oxygen is controlled by a control algorithm so that the output power of the fuel cell meets the required power in real time, thereby ensuring the operating stability of the load. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a structural schematic diagram of a distributed fuel cell power generation system based on solid-state hydrogen storage according to the present invention;

[0060] Figure 2 This is a schematic diagram of the control module flow of the distributed fuel cell power generation system based on solid-state hydrogen storage according to the present invention;

[0061] Figure 3 and Figure 4 This is a schematic diagram of the structure of a solid-state hydrogen storage bottle of a portable fuel cell power generation system based on solid-state hydrogen storage according to the present invention. DETAILED DESCRIPTION

[0062] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] like Figure 1 As shown, a distributed fuel cell power generation system based on solid-state hydrogen storage includes a box, a fuel cell module, a gas supply module, an output module and a control module;

[0064] The gas supply module is connected to the fuel cell module, the output module is connected to the fuel cell module, the control module is connected to the gas supply module and the output module, and the control module collects current and voltage data of the output module.

[0065] The gas supply module includes a hydrogen gas supply module and an oxygen gas supply module. The hydrogen gas supply module includes a solid hydrogen storage bottle, a mass flow meter, a flow valve and a tail exhaust valve. The oxygen gas supply module includes an air compressor, a mass flow meter, a cooling humidifier and a throttle.

[0066] The output module includes a DC-DC and an inverter. The DC-DC outputs 24V direct current, and the inverter outputs 220V or 380V alternating current. The air compressor is connected to a cooling humidifier, which is connected to the cathode inlet of the fuel cell, and the cathode outlet of the fuel cell is connected to a throttle.

[0067] The solid hydrogen storage bottle has a hydrogen storage pressure of 2Mpa, a maximum pressure of 4.5Mpa, and can store 200 to 250g of hydrogen. There can be 8 to 16 bottles of the solid hydrogen storage bottle in one box.

[0068] The solid-state hydrogen storage bottles are connected in parallel, and each solid-state hydrogen storage bottle is connected to a flow valve at the output interface; the flow valve is connected to a mass flow meter, the mass flow meter is connected to the anode inlet of the fuel cell, and the anode outlet of the fuel cell is connected to the tail exhaust valve.

[0069] When the solid-state hydrogen storage bottle is placed in the box and is in a hydrogen release state, the axial direction of the output interface of the solid-state hydrogen storage bottle should be θ with the horizontal direction.

[0070] like Figure 3 As shown, the solid-state hydrogen storage bottle is composed of a heat insulation layer, a composite phase change material filling area and a hydrogen storage area. The heat insulation layer is used to isolate heat loss and fix the phase change material and the lithium battery.

[0071] The hydrogen storage area is approximately a cylinder with a cross-sectional radius of R, and is filled with a low-temperature solid hydrogen storage alloy LaNi5.

[0072] The composite phase change material is composed of lauric acid, elaidic acid and decanoic acid as the original phase change materials, and the three are mixed in proportion according to the lowest eutectic principle to obtain the composite phase change material; when the mass ratio of lauric acid, elaidic acid and decanoic acid is 31.07-37.68:50.61-54.29:11.71-14.64, the solidification point of the composite material can be guaranteed to be lower than 28°C; the hydrogen stored in the solid hydrogen storage bottle is discharged at a maximum hydrogen discharge flow rate The mass of the composite phase change material required to be filled when all the heat generated by the composite phase change material is completely absorbed is 5.24kg to 6.47kg.

[0073] The wall surface between the composite phase change material filling area and the hydrogen storage area is designed with bionic heat-conducting fins based on the jade lotus leaf vein structure to improve the heat exchange efficiency between the composite phase change material filling area and the hydrogen storage area and maintain the thermal balance between the two; the bionic heat-conducting fins are evenly distributed around the wall surface between the composite phase change material filling area and the hydrogen storage area, and the leaf vein length in the heat-conducting fin structure is L k , leaf vein width is W k , the branch angle is α k , where k represents the number of leaf veins, L1 and W1 are the initial length and width of the first main leaf vein and the first branching angle α2 of the leaf vein structure; let the ratio of the length of the k+1th leaf vein to the length of the kth leaf vein be the length ratio L θ , Let the ratio of the width of the k+1th leaf vein to the width of the kth leaf vein be the length ratio W θ , Let the ratio of the k+1th level leaf vein branch angle to the kth level leaf vein branch angle be the length ratio α θ , Inside the hydrogen storage area, the main leaf veins are extended into the hydrogen storage area with an integrated wavy heat-conducting fin to increase the contact area of ​​the solid hydrogen storage alloy. The length is L0 and the width is W0. The function expression is: When L1=20mm, W1=2mm, α2=60°, and L θ ∈(0.68,0.74),W θ ∈(0.42,0.45),α θ When ∈(0.54,0.59), the thermal fins have better thermal conductivity.

[0074] like Figure 2 As shown, the control module is based on the control system and includes: a data acquisition module, a data analysis module and a control execution module;

[0075] S1: The data acquisition module collects various data used by the distributed power generation system in real time, and transmits the collected data to the data analysis module;

[0076] S2: The data analysis module processes the collected data and extracts features, builds a deep neural network model to predict the required power of the power generation system and transmits the results to the control execution module;

[0077] S3: The control execution module automatically executes control measures according to the predicted demand power, and controls the output power of the fuel cell to match the changing demand power in advance.

[0078] In step S1, the data acquisition module acquires the current signal I and the current differential signal dI as well as the voltage signal V and the voltage differential signal dV of the controlled fuel cell power generation system; the current signal I and the voltage signal V are obtained by collecting and differentially extracting the output current of the power generation system after the Hall current sensor and the voltage sensor perform noise filtering; the current differential signal dI and the voltage differential signal dV are obtained by performing differential signal extraction on the extracted current signal I and voltage signal V by the tracking differentiator; the acquisition frequency f is set, and the data acquisition module collects the above data in real time according to the set frequency and transmits it to the data analysis module.

[0079] In step S2, the incoming data is normalized and features are extracted; the voltage signal V, the current signal I, the voltage differential signal dV, and the current differential signal dI are used as input vectors, the input vectors are input into the deep neural network, the deep neural network model is obtained, and the predicted demand power P is output. N .

[0080] The deep neural network model includes an input layer, a hidden layer and an output layer, and each layer is fully connected. The output strategy formula for the jth neuron in the lth layer is:

[0081]

[0082] In the formula, is the output of the jth neuron in the lth layer, n l-1 is the number of neurons in the l-1th layer, is the weight from the kth neuron in the l-1th layer to the jth neuron in the lth layer, The output value of the kth neuron in the l-1th layer, is the bias of the jth neuron in the lth layer, σ is the activation function, and in this deep neural network model, the input layer and hidden layer use the Relu function, and the output layer uses the Sigmoid function.

[0083] The original data set of 1,000 sets of working condition data obtained will be divided into training data set and test data set in a ratio of 8:2; the training data set will be trained 8,000 to 10,000 times, hoping that the deep neural network can accurately learn and predict the changes in demand power. After the model training is completed, the remaining test data set will be used to test the deep neural network to verify its accuracy in predicting the changes in demand power.

[0084] In step S3, a model is established to analyze the relationship between the required power and the gas mass flow rate of the positive and negative electrodes of the fuel cell, and the flow rates of the two electrodes are precisely controlled through a control algorithm.

[0085] S31, the execution module establishes an output power model of the fuel cell;

[0086] In this power generation system, the number of fuel cells is 1000, so the function expression of the output current and the output power of the fuel cells of this power generation system is approximately:

[0087]

[0088] Where P N is the required power predicted by the deep neural network in step S2, I N is the current value corresponding to the required power;

[0089] Through the functional relationship between the oxygen mass flow rate and current required by the fuel cell cathode and the air mass flow rate input to the cathode by the air compressor in the power generation system, it can be obtained that the power generation system at power P N When the mass flow rate of air input to the air compressor is Q a,N (kg / s) is:

[0090]

[0091] Through the relationship between the oxygen mass flow rate and current required by the fuel cell cathode, it can be obtained that the power generation system has a power of PN When the mass flow rate of air input to the air compressor is Q H,N (kg / s) is:

[0092]

[0093] S32, calculate the anode required hydrogen mass flow Q according to S31 H,N Dynamically control the opening V of the flow valve of the solid hydrogen storage bottle and the number of hydrogen bottles N H ; Each solid-state hydrogen storage bottle sets the maximum hydrogen discharge flow rate through a rectifier valve So when the predicted power is P N hour:

[0094] S321, if but The flow valve opening of each hydrogen bottle is the maximum V max , the hydrogen discharge flow rate is

[0095] S322, if but Top N H - The flow valve opening of one hydrogen bottle is the maximum V max , the hydrogen discharge flow rate is Then the Nth H The hydrogen flow rate that the hydrogen tank needs to output

[0096] The execution module establishes a sliding mode control model and combines the real-time feedback of hydrogen flow to calculate the hydrogen flow deviation e H To control the Nth H The opening V of the flow valve of the first hydrogen bottle is controlled to achieve H Accurate control of the hydrogen flow rate of each hydrogen bottle; according to the hydrogen flow deviation e H Construct the switching function s. The expression of the switching function is:

[0097]

[0098] Then the derivative of the switching function s is:

[0099]

[0100] In order to make the system approach stability, the system approach rate is set to exponential approach:

[0101]

[0102] Among them, sat(s) is a continuous function, which is used to prevent the vibration of the opening V of the hydrogen bottle flow valve and make it converge smoothly. The expression of sat(s) is:

[0103]

[0104] S33, cathode required air mass flow Q calculated according to S31 a,N Dynamically control the speed n of the air compressor; the execution module establishes a PID model to control the speed n of the air compressor by calculating the air flow deviation Δe in combination with the real-time feedback of the air flow, thereby achieving precise control of the air flow; the PID algorithm is used for feedback control to convert the air mass flow measurement value Q a,a and the target value Q a,N The proportion (P), integral (I) and differential (D) of the deviation Δe are linearly combined to form the control quantity; the expression of the PID algorithm feedback control flow valve opening is:

[0105]

[0106] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A distributed fuel cell power generation system based on solid-state hydrogen storage, characterized in that: It includes a box body, a fuel cell module, an air supply module, an output module and a control module; the air supply module is connected to the fuel cell module to provide reaction gas to the fuel cell module; the output module is connected to the fuel cell module to smoothly output the electric energy generated by the fuel cell; the control module is connected to the air supply module and the output module to control the supply amount of hydrogen and oxygen of the air supply module in advance.

2. A distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 1, characterized in that: The gas supply module includes a hydrogen gas supply module and an oxygen gas supply module. The hydrogen gas supply module includes a solid hydrogen storage bottle, a mass flow meter, a flow valve and a tail exhaust valve, and the oxygen gas supply module includes an air compressor, a mass flow meter, a cooling humidifier and a throttle; the output module includes a DC-DC and an inverter; the air compressor is connected to the cooling humidifier, and the cooling humidifier is connected to the cathode inlet of the fuel cell to reduce the temperature of the fuel cell and keep the proton exchange membrane at an appropriate humidity; the cathode outlet of the fuel cell is connected to the throttle to adjust the cathode pressure level.

3. A distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 2, characterized in that: The solid-state hydrogen storage bottle has a hydrogen storage pressure of 2Mpa, a maximum tolerable pressure of 4.5Mpa, and can store 200 to 250g of hydrogen; there can be 8 to 16 bottles of the solid-state hydrogen storage bottles in a box; the solid-state hydrogen storage bottles are connected in parallel, and each solid-state hydrogen storage bottle is connected to a flow valve at the output interface to control the size of the hydrogen flow rate output by the hydrogen storage bottle. When providing hydrogen for the fuel cell, the control module can freely control the number of hydrogen supply bottles and the size of the output hydrogen flow rate of each bottle by controlling the opening of the flow valve; the flow valve is connected to a mass flow meter to measure the actual output flow of each solid-state hydrogen storage bottle, and the mass flow meter is connected to the anode inlet of the fuel cell to transmit the measured hydrogen to the fuel cell; the anode outlet of the fuel cell is connected to the tail exhaust valve to drain water and discharge nitrogen that penetrates from the cathode to the anode.

4. A distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 2, characterized in that: When the solid-state hydrogen storage bottle is placed in the box and is in a hydrogen release state, the axial direction of the output interface of the solid-state hydrogen storage bottle should be θ with the horizontal direction, so as to maintain a high hydrogen release efficiency of the solid-state hydrogen storage bottle; In the formula, Q is the hydrogen flow rate collected by the mass flowmeter, T is the time interval between each collection, and S is the total mass of hydrogen stored in the solid hydrogen storage bottle.

5. A distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 1, characterized in that: The solid-state hydrogen storage bottle is composed of an insulation layer, a composite phase change material filling area and a hydrogen storage area; the insulation layer serves to isolate heat loss and fix the phase change material; the hydrogen storage area is approximately a cylinder with a cross-sectional radius of R, and is filled with a low-temperature solid hydrogen storage alloy LaNi5; the solid-state hydrogen storage bottle absorbs heat during hydrogenation and releases heat during hydrogenation, and stores the heat released during hydrogenation through the composite phase change material and releases the heat during hydrogenation; the composite phase change material uses lauric acid, oleic acid and hemp acid as the original phase change materials, and the three are mixed in proportion according to the principle of lowest eutectic to obtain a composite phase change material; when the hydrogenation temperature of the solid-state hydrogen storage bottle reaches 28-34°C, it can be discharged at a maximum hydrogen discharge flow rate Continuously release hydrogen; therefore, when the mass ratio of lauric acid, oleic acid and capric acid is 31.07-37.68:50.61-54.29:11.71-14.64, the solidification point of the composite material can be guaranteed to be lower than 28°C; the hydrogen stored in the solid hydrogen storage bottle is released at the maximum hydrogen flow rate The mass of the composite phase change material required to be filled when all the heat generated by the composite phase change material is completely absorbed is 5.24kg to 6.47kg.

6. A distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 1, characterized in that: The wall surface between the composite phase change material filling area and the hydrogen storage area is designed with bionic heat-conducting fins based on the jade lotus leaf vein type structure to improve the heat exchange efficiency between the composite phase change material filling area and the hydrogen storage area and maintain the thermal balance between the two; the bionic heat-conducting fins are evenly distributed around the wall surface between the composite phase change material filling area and the hydrogen storage area, and the leaf vein length in the heat-conducting fin structure is L k , leaf vein width is W k , the branch angle is α k , where k represents the order of the leaf veins of the jade lotus, L1 and W1 are the initial length and width of the first main leaf vein and the first branching angle α2 of the leaf vein structure; let the ratio of the length of the k+1th level leaf vein to the length of the kth level leaf vein be the length ratio L θ , Let the ratio of the width of the k+1th leaf vein to the width of the kth leaf vein be the length ratio W θ , Let the ratio of the k+1th level leaf vein branch angle to the kth level leaf vein branch angle be the length ratio α θ , Inside the hydrogen storage area, the main leaf veins extend into the hydrogen storage area with an integrated wavy heat-conducting fin to increase the contact area with the solid hydrogen storage alloy. The width is W0, and its function expression is: When L1=20mm, W1=2mm, α2=60°, and L θ ∈(0.68,0.74),W θ ∈(0.42,0.45),α θ When ∈(0.54,0.59), the thermal fins have better thermal conductivity.

7. A distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 1, characterized in that: The control module comprises a data acquisition module, a data analysis module and a control execution module based on the control system; S1: The data acquisition module collects various data of the distributed power generation system in real time, and transmits the collected data to the data analysis module through the bus; S2: The data analysis module processes the collected data and extracts features, builds a deep neural network model to predict the required power of the power generation system and transmits the results to the control execution module; S3: The control execution module automatically executes control measures according to the predicted demand power, and controls the output power of the fuel cell to match the changing demand power in advance.

8. A distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 6, characterized in that: In step S1, the data acquisition module acquires the current signal I and the current differential signal dI as well as the voltage signal V and the voltage differential signal dV of the controlled fuel cell power generation system; the current signal I and the voltage signal V are obtained by collecting and differentially extracting the output current of the power generation system after the Hall current sensor and the voltage sensor perform noise filtering; the current differential signal dI and the voltage differential signal dV are obtained by performing differential signal extraction on the extracted current signal I and voltage signal V by the tracking differentiator; the acquisition frequency f is set, and the data acquisition module collects the above data in real time according to the set frequency and transmits it to the data analysis module through the bus.

9. A distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 6, characterized in that: In step S2, the incoming data is normalized and features are extracted; the voltage signal V, the current signal I, the voltage differential signal dV, and the current differential signal dI are used as input vectors, the input vectors are input into the deep neural network, the deep neural network model is obtained, and the predicted demand power P is output. N ; The deep neural network model includes an input layer, a hidden layer and an output layer, and each layer is fully connected. The output strategy formula for the jth neuron in the lth layer is: In the formula, is the output of the jth neuron in the lth layer, n l-1 is the number of neurons in the l-1th layer, is the weight from the kth neuron in the l-1th layer to the jth neuron in the lth layer, The output value of the kth neuron in the l-1th layer, is the bias of the jth neuron in the lth layer, σ is the activation function. In this deep neural network model, the input layer and hidden layer use the Relu function, and the output layer uses the Sigmoid function; The 1,000 sets of original data sets of operating condition data obtained are divided into training data sets and test data sets in a ratio of 8:2; the training data sets are used for 8,000 to 10,000 trainings, and it is expected that the deep neural network can accurately learn and predict the changes in demand power; after the model training is completed, the remaining test data sets will be used to test the deep neural network to verify its accuracy in predicting the changes in demand power.

10. The distributed fuel cell power generation system based on solid-state hydrogen storage according to claim 6, characterized in that: In step S3, a model is established to analyze the relationship between the required power and the gas mass flow rate of the positive and negative electrodes of the fuel cell, and the flow rates of the two electrodes are precisely controlled through a control algorithm; S31, the execution module establishes an output power model of the fuel cell; In this power generation system, the number of fuel cells is 1000, so the function expression of the output current and the output power of the fuel cells of this power generation system is approximately: Where P N is the required power predicted by the deep neural network in step S2, I N is the current value corresponding to the required power; Through the functional relationship between the oxygen mass flow rate and current required by the fuel cell cathode and the air mass flow rate input to the cathode by the air compressor in the power generation system, it can be obtained that the power generation system at power P N When the mass flow rate of air input to the air compressor is Q a,N (kg / s) is: Through the relationship between the oxygen mass flow rate and current required by the fuel cell cathode, it can be obtained that the power generation system has a power of P N When the mass flow rate of air input to the air compressor is Q H,N (kg / s) is: S32, calculate the anode required hydrogen mass flow Q according to S31 H,N Dynamically control the opening V of the flow valve of the solid hydrogen storage bottle and the number of hydrogen bottles N H ; Each solid-state hydrogen storage bottle sets the maximum hydrogen discharge flow rate through a rectifier valve So when the predicted power is P N hour: S321, if but The flow valve opening of each hydrogen bottle is the maximum V max , the hydrogen discharge flow rate is S322, if but Top N H - The flow valve opening of one hydrogen bottle is the maximum V max , the hydrogen discharge flow rate is Then the Nth H The hydrogen flow rate that the hydrogen tank needs to output The execution module establishes a sliding mode control model and combines the real-time feedback of hydrogen flow to calculate the hydrogen flow deviation e H To control the Nth H The opening V of the flow valve of the first hydrogen bottle is controlled to achieve H Accurate control of the hydrogen flow rate of each hydrogen bottle; according to the hydrogen flow deviation e H Construct the switching function s. The expression of the switching function is: Then the derivative of the switching function s is: In order to make the system approach stability, the system approach rate is set to exponential approach: Among them, sat(s) is a continuous function, which is used to prevent the vibration of the opening V of the hydrogen bottle flow valve and make it converge smoothly. The expression of sat(s) is: S33, cathode required air mass flow Q calculated according to S31 a,N Dynamically control the speed n of the air compressor; the execution module establishes a PID model to control the speed n of the air compressor by calculating the air flow deviation Δe in combination with the real-time feedback of the air flow, thereby achieving precise control of the air flow; use the PID algorithm for feedback control to convert the air mass flow measurement value Q a,a and the target value Q a,N The proportion (P), integral (I) and differential (D) of the deviation Δe are linearly combined to form the control quantity; the expression of the PID algorithm feedback control flow valve opening is:

Images