Combined heat and power system model and cooling water temperature control method, equipment and medium thereof

Abstract

The invention provides a combined heat and power system model and a cooling water temperature control method, equipment and medium thereof, and belongs to the technical field of fuel cell combined heat and power, and the combined heat and power system model comprises a cell stack model, a cooling water tank model, a radiator model, a heat exchanger model and a heat storage water tank model. The cell stack model is connected with a load; a water outlet of the first water pump model is connected with a cooling water inlet of the cell stack model; a water inlet of the radiator model is connected with a water outlet of the cell stack model through a three-way valve, and a water outlet is connected with a water inlet of the first water pump model; a water inlet of the heat exchanger model is connected with a water outlet of the cell stack model through a three-way valve, and a water outlet is connected with a water inlet of the first water pump model; a water outlet of the second water pump model is connected with a water inlet of the heat exchanger model; a water inlet of the heat storage water tank model is connected with a water outlet of the heat exchanger model, and a water outlet is connected with a water inlet of the second water pump model. The heat utilization rate can be remarkably increased.

Description

Technical Field

[0001] The invention belongs to the technical field of fuel cell cogeneration, and in particular relates to a cogeneration system model and a cooling water temperature control method, equipment and medium thereof. Background Art

[0002] In recent years, energy crisis and environmental pollution have become increasingly severe, and the development of clean energy is in full swing. Among them, proton exchange membrane fuel cells (PEMFC) are considered to be the most promising green energy conversion device to replace traditional fossil energy due to their high energy conversion efficiency, high specific power, fast start-up and no pollution.

[0003] Under normal operating conditions, the power generation efficiency of the PEMFC stack is 40% to 60%, which means that about half of the energy is converted into heat energy, and the optimal operating temperature of the PEMFC is 60 to 80°C. In order to maintain the efficient and stable operation of the stack, the heat generated by the stack must be transferred in time, because excessive temperature will affect the performance of the stack and even cause irreversible losses. However, traditional PEMFC dissipates heat through fans, and the heat is completely transferred to the environment and wasted. Summary of the invention

[0004] In order to overcome the above-mentioned shortcomings of the prior art, the present invention provides a combined heat and power system model, comprising:

[0005] A battery stack model, wherein the battery stack model is connected to a load;

[0006] A first water pump model, whose water outlet is connected to the cooling water inlet of the battery stack model;

[0007] A radiator model, whose water inlet is connected to the water outlet of the battery stack model through a three-way valve, and whose water outlet is connected to the water inlet of the first water pump model; the battery stack model, the first water pump model, the radiator model and the three-way valve constitute a heat dissipation circuit;

[0008] The heat exchanger model has a water inlet connected to the water outlet of the battery stack model through a three-way valve, and a water outlet connected to the water inlet of the first water pump model; the battery stack model, the first water pump model, the heat exchanger model and the three-way valve constitute a heat exchange loop;

[0009] A second water pump model, whose water outlet is connected to the water inlet of the heat exchanger model;

[0010] A water storage tank model, wherein the water inlet is connected to the water outlet of the heat exchanger model, and the water outlet is connected to the water inlet of the second water pump model; the heat exchanger model, the second water pump model and the heat exchanger model constitute a heat storage loop, and the water storage tank model is connected to a load;

[0011] The controller, the first water pump model, the second water pump model and the radiator model are all electrically connected to the controller.

[0012] Preferably, a temperature sensor and a liquid level sensor are provided inside the hot water storage tank model, and both the temperature sensor and the liquid level sensor are electrically connected to the controller.

[0013] Preferably, the battery stack model is:

[0014] V cell =E nerst -V act -V ohmic -V conc

[0015] in,

[0016]

[0017] I=i+i n

[0018] V ohmic =IR int =I(R m +R c )

[0019]

[0020] Where V cell Represents the output voltage, E nerst represents the Nernst voltage, V act represents activation loss, V ohmic represents ohmic loss, V conc represents the concentration loss, i n , b, J max are all coefficients, which can be calibrated by experimental data. ξ1, ξ2, ξ3 and ξ4 are all semi-empirical coefficients. n is the no-load current, R c is the electron transfer equivalent resistance, R m is the equivalent impedance of the proton exchange membrane, l is the thickness of the membrane, ρ m is the resistivity of the membrane, λ is the water content of the membrane, J max is the limiting current density, represents the partial pressure of hydrogen, is the oxygen concentration at the cathode catalytic interface, I is the total current, J is the current density, A is the membrane area, R int is the total resistance, ΔS is the reaction entropy, n is the number of electron transfers in the electrode reaction, F is the charge of a single electron, T is the actual temperature, and T ref is the reference temperature, P iis the partial pressure of component i, R is the ideal gas constant, and i represents the working current;

[0021] The first water pump model (2) and the second water pump model (6) are both:

[0022]

[0023] In the formula, q cl Indicates the coolant mass flow rate (kg / s); C cl Indicates the specific heat capacity of the coolant (KJ / kg·K); Indicates the outlet coolant temperature (K); Indicates the inlet coolant temperature (K);

[0024] The radiator model (3) is:

[0025] Q cl =A1K(T rad -T atm )

[0026] In the formula, A1 represents the heat exchange area of ​​the radiator (m 2 ); K represents the heat transfer coefficient of the radiator (W / m 2 ·K); T rad Indicates the heat exchange temperature of the radiator (℃); T atm Indicates ambient temperature (℃);

[0027] The heat exchanger model (5) is:

[0028]

[0029] Q=kA2(T h -T c )

[0030] In the formula, C w，h is the specific heat capacity of hot water (kJ / (kg·K)), T h is the hot water temperature in the heat exchanger (K), q h is the hot water flow rate (kg / s), T h，in is the inlet hot water temperature of the heat exchanger (K), Q is the heat transfer capacity (kW), C w，c is the specific heat capacity of cold water (kJ / (kg·K)), q c is the cold water flow rate (kg / s), T c is the cold water temperature in the heat exchanger (K), T c，in is the inlet cold water temperature of the heat exchanger (K), k is the heat transfer coefficient (kW / (m 2 ·K)), A2 is the effective heat exchange area of ​​the heat exchanger (m 2 );

[0031] The hot water storage tank model is:

[0032]

[0033] Where: tank is the density of water in the hot water tank (kg / m 3 ), V tank is the volume of the hot water tank (m 3 ), C w is the specific heat capacity of water in the hot water storage tank (kJ / (kg·K)), T tank is the temperature of the water in the hot water tank (K), Q is the heat exchange capacity of the heat exchanger (kW), W use The mass flow rate of water used by the user (kg / s), T amb is the ambient temperature (K).

[0034] The present invention also provides a method for controlling the cooling water temperature of a battery stack model based on the combined heat and power system model, comprising the following steps:

[0035] Adjusting the output power of the battery stack model according to the power demand of the load;

[0036] According to the output power, the parameters of the controller are optimized using a radial basis function neural network RBF to obtain an RBF-PID controller;

[0037] Collect the temperature of the hot water storage tank model; determine whether the temperature of the hot water storage tank model is greater than the upper limit of the temperature threshold; if so, close the three-way valve, and use the RBF-PID controller to adjust the speed of the radiator model and the first water pump model to control the inlet temperature and inlet and outlet temperature difference of the cooling water of the battery stack model; if not, determine whether the temperature of the hot water storage tank model is less than the lower limit of the temperature threshold; if so, open the three-way valve, and use the RBF-PID controller to adjust the speed of the first water pump model and the second water pump model to control the inlet temperature and inlet and outlet temperature difference of the cooling water of the battery stack model.

[0038] Preferably, the radial basis function neural network RBF comprises an input layer, a hidden layer and an output layer, the number of nodes in the input layer is 3, the number of nodes in the hidden layer is 4, and the number of nodes in the output layer is 1.

[0039] Preferably, the RBF-PID controller is:

[0040]

[0041] In the formula, K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient, and e(t) is the error function, that is, the difference between the actual speed and the target speed of the water pump and cooling fan.

[0042] The present invention also provides a computer device, including a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the battery stack model cooling water temperature control method.

[0043] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor to execute the battery stack model cooling water temperature control method.

[0044] The combined heat and power system model provided by the present invention has the following beneficial effects:

[0045] The present invention sets a heat exchange circuit and a heat storage circuit, and can transfer the heat generated by the battery stack model to the hot water storage tank through the heat exchange circuit. When the output power of the battery stack model gives priority to meeting the user's electricity demand, the heat generated by the battery stack model can be supplied to the user's heat demand. When the heat supply is greater than the user's demand, the excess heat can be stored in the hot water tank model in the form of hot water through the heat storage circuit. When the heat supply is less than the user's demand, the stored heat can be used preferentially. If the heat supply is still less than the user's heat load, the system output power is adjusted according to the specific user's electricity demand, and the generated heat is supplied to the user to meet the user's heat demand. The present invention can both maintain a suitable operating temperature and recycle waste heat, thereby significantly improving the energy utilization rate of the system.

[0046] The present invention adjusts the output power of the battery stack model according to the power demand of the load; optimizes the parameters of the controller according to the output power and uses a radial basis function neural network RBF to obtain an RBF-PID controller; the RBF-PID controller can be used to adjust the rotation speed of the first water pump model and the second water pump model to control the inlet temperature and the inlet and outlet temperature difference of the cooling water of the battery stack model, thereby being able to accurately and stably control the operating temperature and effectively utilize the heat carried in the cooling water in the battery stack model. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiment of the present invention and its design scheme, the following briefly introduces the drawings required for this embodiment. The drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 PEMFC cogeneration system structure diagram;

[0049] Figure 2 Flow chart of the control method of the combined heat and power system;

[0050] Figure 3 RBF-PID structure diagram;

[0051] Figure 4 Typical working conditions;

[0052] Figure 5 Heat exchange loop RBF-PID vs. PID control effect (inlet and outlet temperature difference);

[0053] Figure 6 Heat exchange loop RBF-PID vs. PID control effect (inlet temperature);

[0054] Figure 7 RBF identification diagram of inlet and outlet cooling water temperature difference (with relative error);

[0055] Figure 8 RBF identification diagram of inlet cooling water temperature (with relative error).

[0056] Description of reference numerals:

[0057] 1-battery stack model, 2-first water pump model, 3-radiator model, 4-three-way valve, 5-heat exchanger model, 6-second water pump model, 7-hot water storage tank model, 8-cooling water tank model. DETAILED DESCRIPTION

[0058] In order to enable those skilled in the art to better understand the technical solution of the present invention and implement it, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the scope of protection of the present invention.

[0059] In the description of the present invention, it is to be understood that the terms “center”, “longitudinal”, “lateral”, “length”, “width”, “thickness”, “up”, “down”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, “axial”, “radial”, “circumferential”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the technical solutions of the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0060] In addition, the terms "first", "second", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance. In the description of the present invention, it should be noted that, unless otherwise clearly specified or limited, the terms "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. In the description of the present invention, unless otherwise specified, "plurality" means two or more, which will not be described in detail here.

[0061] Example

[0062] The present invention provides a combined heat and power system model, specifically as follows Figure 1 As shown, it includes a battery stack model 1, a first water pump model 2, a radiator model 3, a heat exchanger model 5, a second water pump model 6, a hot water storage tank model 7 and a controller. The battery stack model 1 is connected to a load; the water outlet of the first water pump model 2 is connected to the cooling water inlet of the battery stack model 1; the water inlet of the radiator model 3 is connected to the water outlet of the battery stack model 1 through a three-way valve 4, and the water outlet is connected to the water inlet of the first water pump model 2; the water inlet of the heat exchanger model 5 is connected to the water outlet of the battery stack model 1 through a three-way valve 4, and the water outlet is connected to the water inlet of the first water pump model 2; the water outlet of the second water pump model 6 is connected to the water inlet of the heat exchanger model 5; the water inlet of the hot water storage tank model 7 is connected to the water outlet of the heat exchanger model 5, and the water outlet is connected to the water inlet of the second water pump model 6. The battery stack model 1, the first water pump model 2, the radiator model 3 and the three-way valve 4 constitute a heat dissipation loop; the battery stack model 1, the first water pump model 2, the heat exchanger model 5 and the three-way valve 4 constitute a heat exchange loop; the heat exchanger model 5, the second water pump model 6 and the heat exchanger model 5 constitute a heat storage loop. The hot water storage tank model 7 is connected to a load; the first water pump model 2, the second water pump model 6 and the radiator model 3 are all electrically connected to the controller.

[0063] In this embodiment, a temperature sensor and a liquid level sensor are provided inside the hot water storage tank model 7, and both the temperature sensor and the liquid level sensor are electrically connected to the controller.

[0064] The specific mathematical models of the battery stack model 1, the cooling water tank model 8, the first water pump model 2, the radiator model 3, the heat exchanger model 5, the second water pump model 6 and the hot water storage tank model 7 are as follows:

[0065] In order to express the open circuit voltage of the proton exchange membrane fuel cell PEMFC, the Nernst voltage is calculated according to the Nernst formula:

[0066]

[0067] In the formula, E nerst represents the Nernst voltage, T fc Represents the battery temperature, and Representing the partial pressures of hydrogen and oxygen, we calculate:

[0068]

[0069] Where P a and P c represents the anode and cathode inlet pressure, RH a and RH c represents the relative humidity of the anode and cathode, i represents the working current, A represents the area of ​​the proton exchange membrane, represents the steam saturation pressure, which is a function of the operating temperature as follows:

[0070]

[0071] Taking into account the activation polarization, ohmic polarization and concentration polarization in the actual operation of PEMFC, the single cell stack model 1 is:

[0072] V cell =E nerst -V act -V ohmic -V conc

[0073] Where V act represents activation loss, V ohmic represents ohmic loss, V conc Represents concentration loss.

[0074] The activation loss can be calculated by the following formula:

[0075]

[0076] Where ξ1, ξ2, ξ3 and ξ4 are semi-empirical coefficients. is the oxygen concentration at the cathode catalytic interface, and the calculation formula is:

[0077]

[0078] In the formula, represents the oxygen partial pressure, T fc Represents the battery temperature;

[0079] The ohmic loss can be calculated by the following formula:

[0080] V ohmic =I(R m +R c )

[0081] In the formula, R c is the electron transfer equivalent resistance, R m is the equivalent impedance of the proton exchange membrane, which can be calculated by the following formula:

[0082]

[0083] Where l is the thickness of the film, ρ m is the resistivity of the membrane and λ is the water content of the membrane.

[0084] The concentration loss is determined by the following formula:

[0085]

[0086] Where b is the semi-empirical coefficient, J max is the limiting current density.

[0087] In actual engineering, the PEMFC stack consists of multiple fuel cells to meet application requirements:

[0088] V stack =n·V ceu

[0089] Where V stack is the stack voltage and n is the number of fuel cells in the stack.

[0090] The mathematical model of the cooling water tank model 8 of the proton exchange membrane fuel cell is described as follows:

[0091]

[0092] In the formula, m tank1 is the mass of water in the tank, C cl is the specific heat capacity of water, T tank1 is the temperature of the water in the tank, T st is the cooling water temperature, W cl is the mass flow rate of cooling water.

[0093] The function of the water tank is to provide cooling water and act as a heat capacity to reduce the temperature fluctuation of the outlet cooling water;

[0094] In this embodiment, the first water pump model 2 and the second water pump model 6 are described as follows:

[0095]

[0096] In the formula, q cl Indicates the coolant mass flow rate (kg / s), C cl Indicates the specific heat capacity of the coolant (KJ / kg·K), Indicates the outlet coolant temperature (K), Indicates the inlet coolant temperature (K);

[0097] For the same water pump, its speed and flow rate form a linear function relationship:

[0098]

[0099] In the formula, L0 and n0 are the maximum flow rate of the water pump and its corresponding speed respectively;

[0100] Then the flow rate under any working condition can be calculated as:

[0101]

[0102] Wherein: L1 represents the flow rate of the coolant (L / min), n1 represents the speed of the water pump (rpm); in this embodiment, the description of the radiator model 3 is as follows:

[0103] Q cl =A1K(T rad -T atm )

[0104] In the formula, A1 represents the heat exchange area of ​​the radiator m 2 ; K represents the heat transfer coefficient of the radiator W / m 2 ·K;T rad Indicates the heat exchange temperature of the radiator ℃; T atm Indicates ambient temperature ℃;

[0105] For the radiator, the heat transfer temperature of the radiator is taken as the arithmetic mean of the radiator outlet and inlet:

[0106]

[0107] In the formula, and are the radiator outlet and inlet temperatures respectively.

[0108] For the same fan, the ratio of air volume and speed has the following proportional relationship:

[0109]

[0110] Where L0 and n0 are the maximum air volume and corresponding speed of the fan respectively.

[0111] Then the air volume under any working condition can be calculated as:

[0112]

[0113] The relationship between the fan air flow rate and the heat transfer coefficient is shown in Table 1:

[0114] Table 1 Relationship between fan air flow rate and heat transfer coefficient

[0115] Air flow kg / s <![CDATA[Heat transfer coefficient W / m 2 ·K]]> 0.503 36.48 0.647 43.95 0.791 54.31 0.935 60.32 1.079 67.2 1.223 68.29 1.367 74.31 1.511 80.16 1.654 85.95

[0116] According to the data in the table above, the relationship between the heat transfer coefficient and the air flow rate is obtained:

[0117]

[0118] In the formula, qai r is the output flow rate of the cooling fan kg / s;

[0119] The above fitting evaluation parameter R 2 It is 0.9893, which reflects the accuracy of the fitting function;

[0120] In this embodiment, the mathematical description of the heat exchanger model 5 of the proton exchange membrane fuel cell is as follows:

[0121]

[0122] Q=kA2(T rad -T atm )

[0123] In the formula, C w，h is the specific heat capacity of hot water kJ / kg·K, T h K is the hot water temperature in the heat exchanger, q h is the hot water flow rate kg / s, T h，in K is the inlet hot water temperature of the heat exchanger, Q is the heat exchange capacity kW, C w，c is the specific heat capacity of cold water kJ / kg·K, q c is the cold water flow rate kg / s, T c is the cold water temperature in the heat exchanger K, T c，in K is the inlet cold water temperature of the heat exchanger, k is the heat transfer coefficient kW / m 2 K, A2 is the effective heat exchange area of ​​the heat exchanger m 2 The present invention connects multiple zero-dimensional heat exchangers in sequence in the cooling water flow direction to form a one-dimensional plate heat exchanger model 5.

[0124] In this embodiment, the mathematical description of the hot water storage tank model 7 is as follows:

[0125]

[0126] In the formula, ρ tank is the density of water in the hot water tank kg / m 3 , V tank is the volume of the hot water tank m 3 , C wis the specific heat capacity of water in the hot water tank kJ / (kg·K, T tank is the temperature of the water in the hot water tank K, Q is the heat exchange capacity of the heat exchanger kW, W use The mass flow rate of water used by the user is kg / s, T amb is the ambient temperature K.

[0127] In this embodiment, the established cogeneration system model adopts the control strategy of "following the electric load" (Following the Electric Load, FEL) means that the system's fuel cell output power gives priority to meeting the user's electricity demand, and the heat generated by the cogeneration system supplies the user's heat demand. Under this operation strategy, when the system heat supply is greater than the user's demand, the excess heat can be stored in the heat storage device in the form of hot water. When the system heat supply is insufficient, the stored heat can be used first. If the heat supply is still less than the user's heat load, other heating equipment is required to meet the user's heat demand, that is, the system output power is adjusted according to the specific user's electricity demand, and the generated heat is supplied to the user to meet the user's heat demand;

[0128] In this embodiment, the working process of the established cogeneration system model is as follows:

[0129] When the fuel cell is working, the cooling water after absorbing heat inside the battery stack model 1 is first regulated by the three-way valve 4 to flow through the heat exchange circuit, and the heat carried in the cooling water is transferred to the hot water storage tank model 7 through the heat exchanger model 5. When the hot water storage tank model 7 stores heat, the water temperature inside the hot water storage tank model 7 continues to rise, and when a certain upper temperature limit is reached, the hot water storage tank model 7 stops storing heat; the three-way valve 4 regulates so that all the cooling water flows through the heat dissipation circuit, and the heat is completely dissipated to the environment through the radiator model 3 in the heat dissipation circuit. When the hot water storage tank model 7 supplies heat to the user, the internal temperature of the hot water storage tank model 7 continues to decrease, and stops supplying heat to the user when it drops to a certain lower temperature limit. At the same time, the three-way valve 4 regulates so that the cooling water flows through the heat exchange circuit, and the hot water storage tank starts to store heat.

[0130] The present invention also provides a cooling water temperature control method for a battery stack model 1 based on a combined heat and power system model, such as Figure 2 As shown, the following steps are included:

[0131] Step 1: Adjust the output power of the battery stack model 1 according to the power demand of the load.

[0132] Step 2: According to the output power, adjust the stack output current density to adjust the stack output power.

[0133] Step 3: Obtain the temperature of the hot water storage tank model 7 by measuring, and determine whether the temperature of the hot water storage tank model 7 is greater than the upper temperature threshold value. If so, execute step 4, otherwise execute step 6;

[0134] Step 4: Control the opening of the three-way valve 4 to 0%, control the operation of the first water pump model 2, and adjust the speed of the first water pump model 2 and the radiator model 3 through the RBF-PID controller to control the inlet temperature and outlet temperature difference of the stack cooling water;

[0135] Step 5: Obtain the temperature of the hot water storage tank model 7 by measuring, and determine whether the temperature of the hot water storage tank model 7 is less than the lower limit of the temperature threshold. If so, execute step 6, otherwise return to step 4;

[0136] Step 6: Control the opening of the three-way valve 4 to 100%, control the operation of the water pump 2, exchange heat between the heat exchange loop and the heat storage loop through the heat exchanger, and adjust the first water pump model 2 and the speed of the water pump 2 through the RBF-PID controller to control the inlet temperature and the inlet and outlet temperature difference of the stack cooling water;

[0137] Step 7: Obtain the temperature of the hot water storage tank model 7 by measuring, and determine whether the temperature of the hot water storage tank model 7 is greater than the upper temperature threshold value. If so, return to step 4, otherwise return to step 6;

[0138] In the embodiment of the present disclosure, the RBF-PID structure diagram is as follows: Figure 3 As shown, the steps of establishing a radial basis function neural network RBF include:

[0139] Set the number of input layer nodes of the radial basis function neural network RBF to 3, where the input data of the first input node is u(k-1), the input data of the second input node is y(k), and the input data of the third input node is y(k-1);

[0140] The number of hidden layer nodes of the radial basis function neural network RBF is set to 4, where the activation function of the hidden layer is taken as the Gaussian basis function, that is:

[0141]

[0142] In the formula, x is the input of the radial basis function neural network RBF neural network, c j is the center point of the jth basis function, and c j =[c j1 , c j2 , …, c jn ] T , b j is a freely selectable parameter that determines the width of the basis function around the center point, and m is the number of hidden layer nodes. The derivative of the activation function is:

[0143]

[0144] Set the number of nodes in the output layer of the radial basis function neural network RBF to 1, and the output data is the identified system temperature;

[0145] Set the input of the radial basis function neural network RBF to:

[0146] x=[u(k-1),y(k),y(k-1)]

[0147] In the formula, u(k-1) is the input data of the first input node of the radial basis function neural network RBF, y(k) is the input data of the second input node of the radial basis function neural network RBF, and y(k-1) is the input data of the third input node of the radial basis function neural network RBF.

[0148] The output of the radial basis function neural network RBF is:

[0149]

[0150] In the formula, w j (k-1) is the weight from the jth hidden layer neuron to the output layer neuron at time k-1;

[0151] The signal forward propagation process of the radial basis function neural network RBF includes:

[0152] The output of the neurons in the input layer of the radial basis function neural network RBF is:

[0153]

[0154] In the formula, u(k-1) is the input data of the first input node of the radial basis function neural network RBF, y(k) is the input data of the second input node of the radial basis function neural network RBF, and y(k-1) is the input data of the third input node of the radial basis function neural network RBF.

[0155] The output of the hidden layer neurons of the radial basis function neural network RBF is:

[0156]

[0157] Where b j (k-1) represents the width of the radial basis function neural network RBF around the center point at time (k-1);

[0158] The output of the neuron in the output layer of the radial basis function neural network RBF is:

[0159]

[0160] The error of the radial basis function neural network RBF is:

[0161] e(k)=y(k)-y m (k)

[0162] In the formula, y(k) is the actual output of the system, and y(k) is also the input data of the second input node of the radial basis function neural network RBF. m (k) is the model output;

[0163] The performance index function of the radial basis function neural network RBF is:

[0164]

[0165] The δ learning algorithm is used in the error back propagation process of the radial basis function neural network RBF to adjust the weights between the layers of the radial basis function neural network RBF. According to the gradient descent method, the learning algorithm of the weights is as follows:

[0166] The weight w from the hidden layer to the output layer j (k):

[0167] From e(k)=y(k)-ym(k) and have to:

[0168]

[0169] General Substituting in:

[0170]

[0171] Then the weight w from the hidden layer to the output layer is j The learning algorithm for (k) is:

[0172]

[0173] w j (k) = w j (k-1)+Δw j (k)+α(w j (k-1)-w j (k-2))

[0174] Where α is the momentum factor α∈[0,1)), ηη>0 is the learning rate, and w j (k-1) is the weight from the jth hidden layer neuron to the output layer neuron at time k-1.

[0175] Hidden layer radial basis function neural network RBF parameter b j (k) and c ji (k):

[0176] Depend on and have to:

[0177]

[0178] Substitute the derivative equation of the activation function into the equation:

[0179]

[0180]

[0181] Then b j (k) and c ji The learning algorithm for (k) is:

[0182]

[0183] c ji (k) = c ji (k-1)+Δc ji (k)+α(c ji (k-1)-c ji (k-2))

[0184] Where b j (k-1) represents the width of the radial basis function neural network RBF around the center point at time (k-1), b j (k-2) represents the width of the radial basis function neural network RBF around the center point at time (k-2), c j (k-1) represents the center point of the jth basis function at time (k-1), c ji (k-1) represents the center point of the i-th basis function of the j-th basis function at time (k-1).

[0185] When calculating PID parameters, you need to use It is called Jacobian information. However, since the model of the controlled object is unknown, Jacobian information cannot be solved directly. It can be approximated as follows:

[0186]

[0187] Since the input of the RBF network is:

[0188] x=[x1(k),x2(k),…,x n (k)] T =[y(k-1),…,y(kn y );u(k),…,u(kn u )] T

[0189] In the formula, x n (k) represents the input value of the nth node in the input layer at time k; y(kny ) represents the nth y output value; u(kn u ) means kn u Time system nth u +1 output.

[0190] Right now Finally, we can get:

[0191]

[0192] In the formula, is the input of the network at time k; u(k) represents the input of the system at time k.

[0193] Calculate the Jacobian information of the system through radial basis function neural network RBF And provide the Jacobian information of the system to the PID self-correction algorithm;

[0194] The system parameter identification process of radial basis function neural network RBF includes:

[0195] Input the system initial data and set the initial value of RBF network parameter b j (0), c ji (0), w j (0), learning rate η, momentum factors α, β, etc.;

[0196] Use the actual system output y(k) to calculate the current network output y m (k);

[0197] Calculate the network parameter increment Δw j (k), Δb j (k), Δc ji (k);

[0198] Calculate the network parameters w j (k), b j (k), c ji (k);

[0199] Calculate the Jacobian information of the system and provide the Jacobian information of the system to the PID self-tuning algorithm;

[0200] The cycle is repeated once at sampling time k, and the next cycle is entered at time k+1;

[0201] In this embodiment, the established PID controller model is:

[0202]

[0203] In the formula, K p is the proportionality coefficient, Ki is the integration coefficient, K d is the differential coefficient, the error function e(t) is the difference between the actual speed of the water pump and the cooling fan and the target speed, and the radial basis function neural network RBF is used to realize the online identification of the system, provide the Jacobian information of the system to the PID self-correction algorithm, and realize K by the gradient descent method p , K i , K d Adaptive changes to achieve better control effect.

[0204] After the above steps, a proton exchange membrane fuel cell cogeneration system and its working process control method and controller are established. In order to reflect the actual working conditions of the system, it is operated as follows Figure 4 The typical working conditions shown are observed and the temperature control effect is analyzed and the system efficiency is analyzed;

[0205] The following table shows the results of RBF-PID controller compared with traditional PID controller under the same current disturbance curve, where IAE is the absolute error integral criterion. The smaller the IAE, the better the control effect. Its definition is as follows:

[0206] IAE=∫|e(t)|dt

[0207] As can be seen from Table 2, it can be found that the control performance of the RBF-PID controller is improved to a certain extent compared with the traditional PID controller:

[0208] Table 2 Control performance of RBF-PID controller

[0209] PID RBF-PID Inlet and outlet cooling water temperature difference IAE 208.30 150.60 Inlet cooling water temperature IAE 94.06 77.32

[0210] like Figure 5 As shown in the figure, the temperature difference between the inlet and outlet cooling water is precisely controlled. The control effect of RBF-PID is better than that of traditional PID, which is reflected in smaller overshoot and shorter adjustment time. The IAE controlled by RBF-PID is reduced by 27.7% compared with that of traditional PID.

[0211] like Figure 6 As shown in the figure, the inlet cooling water temperature is precisely controlled. RBF-PID has better control effect than traditional PID, which is reflected in smaller overshoot and shorter adjustment time. The IAE controlled by RBF-PID is reduced by 17.8% compared with traditional PID.

[0212] like Figure 7 and Figure 8As shown in Figure 2, the radial basis function neural network RBF can accurately identify the temperature of the system online. The maximum relative error of the inlet and outlet cooling water temperature difference does not exceed 3%, and the maximum relative error of the inlet cooling water temperature does not exceed 7%, ensuring the Jacobian information of the system accuracy;

[0213] In this embodiment, the above control method and controller can effectively utilize the heat carried by the fuel cell cooling water and accurately and stably control the operating temperature, thereby improving the efficiency of the fuel cell, reducing the heat loss of the fuel cell and ensuring the efficient and stable operation of the system. The fuel cell efficiency is calculated as follows:

[0214]

[0215] Where N is the number of cells in the stack, I is the current, F is the Faraday constant, which is 96485.33289±0.00059 C / mol, and ΔH is the reaction enthalpy drop.

[0216] Q 热 =mC p ΔT

[0217] Where m is the mass of water in the hot water storage tank, C p is the specific heat of water, and ΔT is the temperature rise of water in the hot water tank under operation.

[0218] W ele =∫UI dt

[0219] Where U is the fuel cell voltage and I is the operating current;

[0220] The original fuel cell system efficiency is:

[0221]

[0222] Where W ele Q is the energy of the electricity generated by the fuel cell. 总 All the energy contained in the fuel used in the fuel cell;

[0223] The efficiency of the combined heat and power system is:

[0224]

[0225] Where W ele Q is the energy of the electricity generated by the fuel cell. 执 The energy of the heat provided by the fuel cell, Q 总 All the energy contained in the fuel used in the fuel cell;

[0226] In this embodiment, it is calculated that the energy utilization rate is increased from 48.61% to 97.23% by using the combined heat and power system;

[0227] In this embodiment, in order to compare the energy saving of fuel cells before and after cogeneration, a baseline energy supply system is set, which consists of a gas boiler, a household gas wall-mounted boiler and a diesel generator; the gas boiler provides heating load, the household gas wall-mounted boiler provides domestic hot water load, and the diesel generator provides electrical load. The energy input to the diesel generator is converted into natural gas with the same calorific value for comparison; considering that natural gas hydrogen production has become a mainstream hydrogen production process, it is assumed that the hydrogen source of the PEMFC cogeneration system is provided by natural gas reforming, and the gas consumption is calculated according to the following formula:

[0228]

[0229] Where V gas-cov Natural gas consumption of each equipment in the baseline system (m 3 ),∑Q load is the cumulative heat or electricity load (kWh), η equ For equipment (gas boilers, household gas wall-mounted boilers and diesel generators) efficiency, HV NG is the lower calorific value of natural gas (MJ / m 3 ).

[0230]

[0231] In the formula, is the hydrogen consumption of PEMFC combined heat and power system m 3 , N is the number of fuel cell cells, I is the operating current of the fuel cell A, is the molar mass of hydrogen (kg / mol), F is the Faraday constant (C / mol), is the density of hydrogen kg / m 3 :

[0232]

[0233] Where V gas-PEM is the natural gas consumption of PEMFC combined heat and power system m 3 , is the hydrogen consumption of PEMFC-CHP system m 3 , The yield of hydrogen from natural gas reforming is %, that is, the hydrogen of the PEMFC cogeneration system is provided by hydrogen from natural gas reforming;

[0234] Set up a traditional heating and power supply system, which is composed of a gas boiler, a household gas wall-mounted boiler and a diesel generator;

[0235] In the traditional heating and power supply system, gas boilers provide heating load, household gas wall-mounted boilers provide domestic hot water load, and diesel generators provide electricity load;

[0236] run Figure 4 The typical operating conditions shown in the figure show that, when the same electric and thermal load is generated as the PEMFC cogeneration system, the reformed natural gas consumption is 0.17m 3 , the gas consumption of the traditional heating and power supply system is 0.40m 3 , gas consumption is saved by 57%;

[0237] The present invention also provides a computer device, including a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the cooling water temperature control method of the battery stack model 1.

[0238] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program is suitable for loading by a processor to execute the cooling water temperature control method of the battery stack model 1.

[0239] The above embodiments are only preferred specific implementation methods of the present invention, and the protection scope of the present invention is not limited thereto. Any simple changes or equivalent replacements of the technical solutions that can be obviously obtained by any technician familiar with the field within the technical scope disclosed in the present invention belong to the protection scope of the present invention.

Claims

1. A combined heat and power system model, characterized in that: include: A battery stack model (1), wherein the battery stack model (1) is connected to a load; a first water pump model (2), a water outlet of which is connected to a cooling water inlet of the battery stack model (1); The radiator model (3) has a water inlet connected to the water outlet of the battery stack model (1) through a three-way valve (4), and a water outlet connected to the water inlet of the first water pump model (2); the battery stack model (1), the first water pump model (2), the radiator model (3) and the three-way valve (4) constitute a heat dissipation circuit; The heat exchanger model (5) has a water inlet connected to the water outlet of the battery stack model (1) through a three-way valve (4), and a water outlet connected to the water inlet of the first water pump model (2); the battery stack model (1), the first water pump model (2), the heat exchanger model (5) and the three-way valve (4) constitute a heat exchange loop; A second water pump model (6), the water outlet of which is connected to the water inlet of the heat exchanger model (5); A water storage tank model (7), wherein the water inlet is connected to the water outlet of the heat exchanger model (5), and the water outlet is connected to the water inlet of the second water pump model (6); the heat exchanger model (5), the second water pump model (6) and the heat exchanger model (5) constitute a heat storage circuit, and the water storage tank model (7) is connected to a load; The controller, the first water pump model (2), the second water pump model (6) and the radiator model (3) are all electrically connected to the controller.

2. The combined heat and power system model according to claim 1, characterized in that: The hot water storage tank model (7) is internally provided with a temperature sensor and a liquid level sensor, and both the temperature sensor and the liquid level sensor are electrically connected to the controller.

3. The combined heat and power system model according to claim 1, characterized in that: The battery stack model (1) is: V cell =E nerst -V act -V ohmic -V conc in, I=i+i n V ohmic =AND int =I(R m +R c ) Where V cell Represents the output voltage, E nerst represents the Nernst voltage, V act represents activation loss, V ohmic represents ohmic loss, V conc represents the concentration loss, i n , b, J max are all coefficients, which can be calibrated by experimental data. ξ1, ξ2, ξ3 and ξ4 are all semi-empirical coefficients. n is the no-load current, R c is the electron transfer equivalent resistance, R m is the equivalent impedance of the proton exchange membrane, l is the thickness of the membrane, ρ m is the resistivity of the membrane, λ is the water content of the membrane, J max is the limiting current density, represents the partial pressure of hydrogen, is the oxygen concentration at the cathode catalytic interface, I is the total current, J is the current density, A is the membrane area, R int is the total resistance, ΔS is the reaction entropy, n is the number of electron transfers in the electrode reaction, F is the charge of a single electron, T is the actual temperature, and T ref is the reference temperature, P i is the partial pressure of component i, R is the ideal gas constant, and i represents the working current; The first water pump model (2) and the second water pump model (6) are both: In the formula, q cl Indicates the coolant mass flow rate (kg / s); C cl Indicates the specific heat capacity of the coolant (KJ / kg·K); Indicates the outlet coolant temperature (K); Indicates the inlet coolant temperature (K); The radiator model (3) is: Q cl JA1K(T rad -T atm ) In the formula, A1 represents the heat exchange area of ​​the radiator (m 2 ); K represents the heat transfer coefficient of the radiator (W / m 2 ·K); T rad Indicates the heat exchange temperature of the radiator (℃); T atm Indicates ambient temperature (℃); The heat exchanger model (5) is: Q=kA2(T h -T c ) In the formula, C w,h is the specific heat capacity of hot water (kJ / (kg·K)), T h is the hot water temperature in the heat exchanger (K), q h is the hot water flow rate (kg / s), T h,in is the inlet hot water temperature of the heat exchanger (K), Q is the heat transfer capacity (kW), C w,c is the specific heat capacity of cold water (kJ / (kg·K)), q c is the cold water flow rate (kg / s), T c is the cold water temperature in the heat exchanger (K), T c,in is the inlet cold water temperature of the heat exchanger (K), k is the heat transfer coefficient (kW / (m 2 ·K)), A2 is the effective heat exchange area of ​​the heat exchanger (m 2 ); The hot water storage tank model (7) is: Where: tank is the density of water in the hot water tank (kg / m 3 ), V tank is the volume of the hot water tank (m 3 ), C w is the specific heat capacity of water in the hot water storage tank (kJ / (kg·K)), T tank is the temperature of the water in the hot water tank (K), Q is the heat exchange capacity of the heat exchanger (kW), W use The mass flow rate of water used by the user (kg / s), T amb is the ambient temperature (K).

4. A cooling water temperature control method based on the combined heat and power system model of claim 1, characterized in that: The steps include: adjusting the output power of the battery stack model (1) according to the power demand of the load; According to the output power, the parameters of the controller are optimized using a radial basis function neural network RBF to obtain an RBF-PID controller; The temperature of the hot water storage tank model (7) is collected; it is determined whether the temperature of the hot water storage tank model (7) is greater than the upper limit of the temperature threshold; if so, the three-way valve (4) is closed, and the RBF-PID controller is used to adjust the rotation speeds of the radiator model (3) and the first water pump model (2) to control the inlet temperature and the inlet and outlet temperature difference of the cooling water of the battery stack model (1); if not, it is determined whether the temperature of the hot water storage tank model (7) is less than the lower limit of the temperature threshold; if so, the three-way valve (4) is opened, and the RBF-PID controller is used to adjust the rotation speeds of the first water pump model (2) and the second water pump model (6) to control the inlet temperature and the inlet and outlet temperature difference of the cooling water of the battery stack model (1).

5. The cooling water temperature control method according to claim 4, characterized in that: The radial basis function neural network RBF comprises an input layer, a hidden layer and an output layer. The number of nodes in the input layer is 3, the number of nodes in the hidden layer is 4, and the number of nodes in the output layer is 1.

6. The cooling water temperature control method according to claim 4, characterized in that: The RBF-PID controller is: In the formula, K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient, and e(t) is the error function, that is, the difference between the actual speed and the target speed of the water pump and cooling fan.

7. A computer device, characterized in that: It comprises a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the cooling water temperature control method according to any one of claims 4 to 6.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor to execute the cooling water temperature control method described in any one of items 4-6.

Images