Hydrogen fuel cell control method based on scene tree random model predictive control

Abstract

The invention relates to the technical field of hydrogen fuel cell control, and particularly discloses a hydrogen fuel cell control method based on scene tree random model predictive control. Comprising the following steps: S1, establishing a hydrogen fuel cell random system model; s2, collecting historical operation data of the hydrogen fuel cell, and associating the operation power of the hydrogen fuel cell with the corresponding temperature, humidity, air pressure and oxygen concentration according to a random system model; s3, performing uncertainty modeling through a Markov prediction method to obtain a Markov chain; s4, generating a scene tree according to the Markov chain; and S5, determining the optimal operation power of the hydrogen fuel cell according to the random system model of the hydrogen fuel cell established in the step S1 and the scene tree established in the step S4. According to the hydrogen fuel cell control method based on scene tree random model predictive control, the problem that the voltage of a hydrogen fuel cell obviously fluctuates is solved, and the operation stability of the hydrogen fuel cell is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrogen fuel cell control, and in particular to a hydrogen fuel cell control method based on scenario tree random model predictive control. Background Art

[0002] A hydrogen fuel cell is a device that converts hydrogen and oxygen into electricity, water and heat through an electrochemical reaction. Since it only produces water during its operation and does not produce harmful substances such as carbon monoxide, hydrocarbons and nitrogen oxides like gasoline and diesel, hydrogen fuel cells are an efficient and clean energy source. Using them in automobiles is also conducive to achieving carbon neutrality. Therefore, hydrogen fuel cells have attracted the attention and research of many scholars.

[0003] Hydrogen fuel cells often encounter various complex environments during actual use, and the operation of hydrogen fuel cells is affected by factors such as ambient temperature, humidity, and air pressure. Although there is a humidity dynamic control method and a hydrogen fuel cell for a hydrogen fuel cell disclosed in Application No. CN202310464629.0, based on a pre-constructed gas supply model and the gas flow required by the gas flow module, the gas flow of the reaction gas supplied to the gas flow module by the gas supply module is adjusted by dynamically controlling the solenoid valve; the gas flow module controls the electrochemical reaction of the reaction gas through the anode flow channel, the cathode flow channel and the pre-constructed gas flow model to determine the operating state of the hydrogen fuel cell; according to the operating state of the hydrogen fuel cell, the humidifier is controlled by a preset fuzzy control algorithm to achieve humidity control of the gas.

[0004] At the same time, temperature will also affect the operation of hydrogen fuel cells, so the existing technology also has patents for controlling the temperature of hydrogen fuel cells. In general, the interference to the operation of hydrogen fuel cells is often reduced by changing the temperature and humidity of the hydrogen fuel cells, but the hydrogen fuel cells after the reduction still have voltage fluctuations and obvious instability. Summary of the invention

[0005] The present invention aims to provide a hydrogen fuel cell control method based on scenario tree random model predictive control to solve the problem of obvious voltage fluctuations in the hydrogen fuel cell and improve the stability of the operation of the hydrogen fuel cell.

[0006] To achieve the above object, the present invention adopts the following technical solution: a hydrogen fuel cell control method based on scene tree random model predictive control, comprising the following steps:

[0007] S1, establish a hydrogen fuel cell stochastic system model, and the system state space model is established as:

[0008]

[0009] Among them, x is the state variable, which includes system load current, anode gas pressure, and cathode gas pressure; u is the control variable system anode and cathode input gas pressure; w is the random variable, which includes ambient temperature, oxygen concentration, air pressure, and humidity; y is the output variable stack output voltage.

[0010] S2, collecting historical operating data of the hydrogen fuel cell, and correlating the operating power of the hydrogen fuel cell with the corresponding temperature, humidity, air pressure and oxygen concentration according to the stochastic system model;

[0011] S3, uncertainty modeling is performed through the Markov prediction method to obtain a Markov chain; where the Markov state transfer matrix T∈R s×s for:

[0012]

[0013] Then calculate it using the maximum likelihood estimation method:

[0014]

[0015] Among them, m ij From the state To status The number of transformations, m i From historical data arrive The sum of the number of transitions;

[0016] S4, generates a scene tree based on the Markov chain;

[0017] S5, determining the optimal operating power of the hydrogen fuel cell according to the hydrogen fuel cell random system model established in step S1 and the scenario tree established in step S4, and adjusting the operation of the hydrogen fuel cell according to the optimal operating power of the hydrogen fuel cell.

[0018] The beneficial effects of this program are:

[0019] 1. Hydrogen fuel cells are randomly affected by temperature, humidity, air pressure and oxygen concentration, and are difficult to predict during control. The inventor found that the random characteristics of uncertainty modeling are consistent with the random interference changes of the hydrogen fuel cell system. Therefore, the hydrogen fuel cell random system model established in this scheme transforms the uncertainty of this random interference into mathematical classification, and finally transforms the random interference into a deterministic control problem through the component scene tree. Therefore, when it is necessary to control the hydrogen fuel cell, the most likely probability is determined by calculating Formula 3, so that the closest optimal operating power of the hydrogen fuel cell can be determined in the historical data, and then the power of the hydrogen fuel cell can be adjusted according to the optimal operating power to keep the optimal output all the time without being affected by temperature, humidity, air pressure and oxygen concentration, which leads to power reduction. That is, after adopting this scheme, the output voltage of the hydrogen fuel cell is kept near the expected value, which effectively reduces the voltage fluctuation and improves the stability of the hydrogen fuel cell.

[0020] 2. The current control idea of ​​hydrogen fuel cells is to adjust the temperature, humidity and other factors of hydrogen fuel cells to reduce the interference of these factors on hydrogen fuel cells. However, in actual implementation, adjusting the temperature, humidity, air pressure and oxygen content at the same time will undoubtedly make the entire control process too complicated, and adding several adjustment structures will also make the structure of hydrogen fuel cells more complicated. This solution does not adjust the interference factors such as temperature, humidity, air pressure and oxygen content themselves, but determines the optimal operating power of the hydrogen fuel cell under the conditions of temperature, humidity, air pressure and oxygen content by establishing a random system model and a scenario tree, so that the hydrogen fuel cell always maintains the optimal output. Therefore, the control of this solution is simpler and the structure of the hydrogen fuel cell is simpler.

[0021] 3. The construction of the scene tree also ensures that each time the optimal operating power is determined, there will be no excessive amount of calculation resulting in reduced control performance due to the need to combine interference factors such as temperature, humidity, air pressure and oxygen content at the same time. Therefore, the control method of this scheme also has the advantages of less dependence on control equipment and easier implementation.

[0022] Furthermore, in step S5, a prediction cycle is first determined, and the calculation of formula (3) is repeated in each prediction cycle to determine the optimal operating power of the hydrogen fuel cell in this prediction cycle.

[0023] The beneficial effects of this scheme are: the optimal power of the hydrogen fuel cell is re-determined in each prediction cycle, and the power of the hydrogen fuel cell is adjusted in time when interference factors such as temperature, humidity, air pressure and oxygen content change, so that the hydrogen fuel cell always maintains optimal power operation.

[0024] Furthermore, in each prediction period, an advance prediction is made on the operating power of the hydrogen fuel cell for at least three prediction periods after the prediction period.

[0025] The beneficial effect of this solution is that advance prediction can improve the accuracy of each control while reducing the amount of calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following is further described in detail through specific implementation methods:

[0028] Example

[0029] A hydrogen fuel cell control method based on scenario tree stochastic model predictive control, such as Figure 1 As shown, the following steps are included:

[0030] S1, establish a hydrogen fuel cell stochastic system model, and the system state space model is established as:

[0031]

[0032] S2, collecting historical operating data of the hydrogen fuel cell, and correlating the operating power of the hydrogen fuel cell with the corresponding temperature, humidity, air pressure and oxygen concentration according to the stochastic system model;

[0033] S3, uncertainty modeling is performed through the Markov prediction method to obtain a Markov chain; where the Markov state transfer matrix T∈R s×s for:

[0034]

[0035] Then calculate it by the maximum likelihood estimation method:

[0036]

[0037] S4, generates a scene tree based on the Markov chain;

[0038] S5, first determine a prediction cycle, in each prediction cycle, determine the optimal operating power of the hydrogen fuel cell in this prediction cycle according to the hydrogen fuel cell random system model established in the above step S1 and the scenario tree established in step S4, and adjust the operation of the hydrogen fuel cell according to the optimal operating power of the hydrogen fuel cell; at the same time, in each prediction cycle, the operating power of the hydrogen fuel cell for at least three prediction cycles after the prediction cycle is predicted in advance.

[0039] The above is only an embodiment of the present invention, and the common knowledge such as the known specific technical solutions and / or characteristics in the solution is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical solution of the present invention, several modifications and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A hydrogen fuel cell control method based on scenario tree stochastic model predictive control, characterized in that: The following steps are involved: S1, establish a hydrogen fuel cell stochastic system model, and the system state space model is established as: S2, collecting historical operating data of the hydrogen fuel cell, and correlating the operating power of the hydrogen fuel cell with the corresponding temperature, humidity, air pressure and oxygen concentration according to the stochastic system model; S3, uncertainty modeling is performed through the Markov prediction method to obtain a Markov chain; where the Markov state transfer matrix T∈R s×s for: Then calculate it using the maximum likelihood estimation method: S4, generates a scene tree based on the Markov chain; S5, determining the optimal operating power of the hydrogen fuel cell according to the hydrogen fuel cell random system model established in step S1 and the scenario tree established in step S4, and adjusting the operation of the hydrogen fuel cell according to the optimal operating power of the hydrogen fuel cell.

2. The hydrogen fuel cell control method based on scenario tree stochastic model predictive control according to claim 1 is characterized in that: In step S5, a prediction cycle is first determined, and the calculation of step formula 3 is repeated in each prediction cycle to determine the optimal operating power of the hydrogen fuel cell in this prediction cycle.

3. The hydrogen fuel cell control method based on scenario tree random model predictive control according to claim 2 is characterized in that: In each prediction period, an operating power of the hydrogen fuel cell in at least three prediction periods after the prediction period is predicted in advance.

Images