Distributed dynamic power distribution method for fuel cell ship based on autonomous frequency division

Abstract

The invention discloses a fuel cell ship distributed dynamic power distribution method based on autonomous frequency division. The method comprises the following steps: establishing a fuel cell ship multi-electric hybrid power system model taking a fuel cell, a lithium battery, a super capacitor, a DC / DC converter and a ship as loads; based on a fuel cell ship multi-electric hybrid power system model, distribution of dynamic ship load power in different power supply units of a fuel cell, a lithium battery and a super capacitor is realized by adopting a hybrid droop control method of virtual impedance, and charge state equalization of a parallel lithium battery pack is realized based on a charge state equalization method of adaptive droop control; according to the invention, ship dynamic load power can be autonomously distributed according to power units with different characteristics, balance of the charge state of the parallel lithium battery pack and stable control of the direct current bus voltage are realized, and meanwhile, distributed control is adopted, so that the communication pressure among the units is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of fuel cell ships, and relates to a decentralized dynamic power distribution method for fuel cell ships based on autonomous frequency division, and in particular to a dynamic power distribution method for a multi-electric hybrid power system of a fuel cell ship. Background Art

[0002] With the strict environmental protection regulations and the sustainable development needs of the marine transportation industry, the ship power system based on fossil fuels is gradually transforming to a more environmentally friendly solution. Fuel cells, as an efficient and low-pollution energy conversion device, are increasingly attracting attention in the application of ship power systems. In order to adapt to the complex and changeable marine environment and operation requirements, fuel cells are usually combined with lithium batteries and supercapacitors to form a multi-electric hybrid system. In this system, how to efficiently manage and distribute various energy sources under dynamically changing load conditions during navigation is the focus of technical research. The system needs to achieve optimal power distribution of fuel cells, lithium batteries and supercapacitors to meet the power requirements of different navigation stages and ensure the stability and reliability of the system. This paper will adopt a decentralized control method, the advantage of which is that each power supply component can be independently controlled by a local controller without a central controller. This autonomous performance not only reduces the pressure of communication, but also improves the reliability of the system. When a power source fails, it will not affect the operation of the entire system. In addition, during the navigation of the ship, the load demand may change frequently due to changes in speed, changes in sea conditions or switching of operating modes. Decentralized control can quickly adjust the power output locally, which not only reduces the response time, but also optimizes energy use and reduces energy consumption. Summary of the invention

[0003] In order to solve the problems of unreasonable dynamic load power distribution, unbalanced charge state of parallel lithium battery packs and communication delay in the multi-electric hybrid power system of fuel cell ships, the technical solution adopted by the present invention is: a decentralized dynamic power distribution method for fuel cell ships based on autonomous frequency division, comprising the following steps:

[0004] Establish a fuel cell ship multi-electric hybrid power system model with fuel cells, lithium batteries, supercapacitors, DC / DC converters and ships as loads;

[0005] Based on the fuel cell ship multi-electric hybrid power system model, a hybrid droop control method with virtual impedance is adopted to realize the distribution of dynamic ship load power among different power supply units such as fuel cell, lithium battery and supercapacitor.

[0006] Further: the hybrid droop control method using virtual impedance realizes the distribution of dynamic ship load power among different power supply units such as fuel cells, lithium batteries and supercapacitors as follows:

[0007] Based on the droop control of virtual inductance, the fuel cell is controlled by current loop control;

[0008] Based on the droop control of virtual resistance, the voltage and current double closed-loop PI control is used to control the lithium battery;

[0009] Based on the droop control of virtual capacitor, voltage and current double closed-loop PI control is adopted to realize the control of supercapacitor.

[0010] Furthermore: it also includes a charge state balancing method based on adaptive droop control, which realizes the charge state balancing of parallel lithium battery packs.

[0011] Furthermore, the process of implementing the control of the fuel cell by using the current loop control based on the virtual inductor droop control is as follows:

[0012] The reference output current value of the fuel cell is obtained by multiplying the difference between the nominal voltage of the fuel cell and the actual output voltage of the fuel cell by the virtual impedance of the fuel cell.

[0013] The output current value based on the fuel cell reference is multiplied by the output voltage to obtain the output power reference value, and then divided by the fuel cell input voltage to obtain the fuel cell input inductor current reference value.

[0014] The reference value of the fuel cell input inductor current is subtracted from the actual input inductor current value and then adjusted by the current loop PI regulator to obtain a pulse width modulation wave to control the converter MOSFET transistor to achieve the steady-state and low-frequency power of the fuel cell in response to the load, thereby achieving the purpose of controlling the power output of the fuel cell.

[0015] Furthermore: the droop control based on virtual resistance adopts voltage and current double closed-loop PI control to realize the process of controlling the lithium battery as follows:

[0016] Based on the nominal voltage of the lithium battery minus the product of the lithium battery output current and the lithium battery virtual impedance, the lithium battery output voltage reference value is obtained.

[0017] Based on the difference between the lithium battery output voltage reference value and the actual lithium battery voltage output value, the lithium battery inductor current reference value is obtained through voltage outer loop regulation.

[0018] Based on the difference between the lithium battery inductor current reference value and the actual lithium battery inductor current value, a pulse width modulation wave is obtained through current inner loop regulation to control the converter MOSFET transistor, so that the lithium battery responds to the intermediate frequency power of the load, thereby achieving the purpose of controlling the lithium battery power output.

[0019] Further: the droop control based on virtual capacitor adopts voltage and current double closed-loop PI control to realize the process of controlling the super capacitor as follows:

[0020] The supercapacitor output voltage reference value is obtained based on the supercapacitor nominal voltage minus the product of the supercapacitor output current and the supercapacitor virtual impedance;

[0021] Based on the difference between the supercapacitor output voltage reference value and the supercapacitor actual voltage output value, the supercapacitor inductor current reference value is obtained through voltage outer loop regulation.

[0022] Based on the difference between the supercapacitor inductor current reference value and the actual supercapacitor inductor current value, a pulse width modulation wave is obtained through current inner loop regulation to control the converter MOSFET transistor, so that the supercapacitor responds to the pulsation and high-frequency power of the load, thereby achieving the purpose of controlling the supercapacitor power output.

[0023] Furthermore: the state of charge equalization method based on adaptive droop control realizes the equalization process of the state of charge of the parallel lithium battery pack as follows:

[0024] The average state of charge, average output voltage and average virtual voltage drop of the parallel lithium battery pack are obtained through the dynamic consistency algorithm;

[0025] According to the droop coefficient formula of lithium battery, the droop coefficient is obtained;

[0026] The voltage compensation is obtained by using a PI regulator based on the difference between the average output voltage of the parallel lithium battery pack and the actual bus voltage.

[0027] The virtual voltage drop compensation is obtained by using a PI regulator based on the difference between the average virtual voltage drop and the actual virtual voltage drop of the parallel lithium battery pack.

[0028] The droop coefficient, voltage compensation amount and virtual voltage drop compensation amount are all superimposed on the voltage reference value to achieve a balanced charge state of the parallel lithium battery pack.

[0029] Further: The droop coefficient formula of the lithium battery is expressed as follows:

[0030]

[0031] Among them, SoC bati , R 0i They are the state of charge of the lithium battery and the initial value of the droop coefficient; SoC avg is the average state of charge of the lithium battery; i o,bati >0 means the lithium battery is discharged, i o,bati <0 indicates lithium battery charging; n is the speed adjustment coefficient.

[0032] The present invention provides a fuel cell ship distributed dynamic power allocation method based on autonomous frequency division. The method is an extended droop control using only a local controller, that is, a hybrid droop control strategy based on virtual impedance is adopted to realize the autonomous frequency division response of load power between fuel cells, lithium batteries and supercapacitors and bus voltage stability control;

[0033] The hybrid droop control strategy based on virtual impedance is to apply virtual inductance, virtual resistance and virtual capacitance droop control to fuel cells, lithium batteries and supercapacitors respectively to ensure the frequency division response of load power among the three power sources, thereby realizing autonomous power distribution and bus voltage stability control among power sources with different characteristics;

[0034] The state of charge balancing strategy based on adaptive droop control implements hierarchical control of the system, including communication layer, droop control layer and direct control layer. This strategy uses a dynamic consistency algorithm to calculate the average state of charge, average output voltage and average virtual voltage drop of the parallel lithium battery pack, and compensates for the voltage and virtual voltage drop through the PI regulator according to the deviation between the actual measured value and the average value, thereby ensuring the balance of the state of charge.

[0035] The proposed state-of-charge equalization strategy based on adaptive droop control is used to achieve the state-of-charge equalization of parallel lithium battery packs. At the same time, distributed control is used to reduce the problem of communication delay. Finally, a simulation model is built on the Matlab / Simulink platform to verify the effectiveness and superiority of the proposed method.

[0036] The present invention can autonomously distribute the dynamic load power of the ship, balance the charge state of the parallel lithium battery pack and stabilize the DC bus voltage according to the power units with different characteristics, and adopts decentralized control to reduce the communication pressure between the units. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 The research object of this invention is the fuel cell ship multi-electric hybrid power system

[0039] Figure 2 The distributed control architecture proposed by the present invention

[0040] Figure 3 This is a control block diagram of the hybrid droop control strategy based on virtual impedance proposed by the present invention;

[0041] Figure 4 This is a control block diagram of the state of charge balancing strategy based on adaptive droop control proposed by the present invention;

[0042] Figure 5 (a) is a load power distribution change curve of the present invention under the case 1, (b) is a charge state change curve of two groups of lithium batteries under the case 1, (c) is a bus voltage change curve of the present invention under the case 1;

[0043] Figure 6 (a) is the load power distribution change curve of the present invention in case 2, (b) is the charge state change curve of two groups of lithium batteries in case 1, and (c) is the bus voltage change curve of the present invention in case 2. DETAILED DESCRIPTION

[0044] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0045] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. 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.

[0046] A decentralized dynamic power allocation method for a fuel cell ship based on autonomous frequency division comprises the following steps:

[0047] S1: Use Matlab / Simulink to establish a fuel cell ship multi-electric hybrid power system model with fuel cells, lithium batteries, supercapacitors, DC / DC converters and ships as loads; accurately simulate the system and configure appropriate dynamic power allocation methods to achieve rational use and management of energy;

[0048] S2: Based on the fuel cell ship multi-electric hybrid power system model, a hybrid droop control method with virtual impedance is adopted to realize the distribution of dynamic ship load power among different power supply units such as fuel cells, lithium batteries and supercapacitors.

[0049] like Figure 1The figure shows the composition diagram of the multi-electric hybrid power system of a fuel cell ship, which is the object of the dynamic power allocation method implemented by the patent, and is composed of fuel cells, lithium batteries, super capacitors, DC / DC converters, DC / AC converters, bus switches, propulsion loads, auxiliary loads and fuses. During normal navigation of the ship, the bus switch is closed, and the fuel cell serves as the main power source to provide dynamic low-frequency power and most of the steady-state power for the propulsion and auxiliary loads. Lithium batteries and super capacitors serve as auxiliary power sources, responding to medium-frequency and high-frequency load power during the dynamic load power change of the ship, and their load power tends to zero in steady state.

[0050] The method is applied to a fuel cell ship multi-electric hybrid power system, which includes a group of parallel fuel cells, a group of parallel lithium batteries and a group of parallel super capacitors. The fuel cell is connected to the DC bus through a unidirectional DC / DC converter, and the lithium battery and super capacitor are connected to the DC bus through a bidirectional DC / DC converter. The fuel cell bears the low-frequency power, and the lithium battery and super capacitor bear the medium and high-frequency power respectively, thereby reducing the fluctuation of the fuel cell output power and extending its service life.

[0051] like Figure 2 The figure shows a distributed control architecture. Distributed control is a control system architecture that distributes control tasks to multiple independent local controllers, each of which is responsible for monitoring and controlling its own system. This architecture allows the system to maintain some functions in the event of a local failure, improving the reliability and fault tolerance of the system. At the same time, distributed control can improve the response speed and processing power of the system by reducing the burden on central processing. Due to its decentralized nature, distributed control significantly improves the reliability and flexibility of the fuel cell ship multi-electric hybrid power system, reduces delays caused by communication problems, and effectively avoids the paralysis of the entire system due to failure of the central controller. It achieves rapid response and efficient distribution through local control, and is particularly suitable for complex and dynamically changing load conditions.

[0052] A fuel cell ship distributed dynamic power distribution control system based on autonomous frequency division includes:

[0053] A fuel cell, a first DC / DC converter connected to the fuel cell, the first DC / DC converter connected to the fuel cell control unit and the DC bus;

[0054] The fuel cell control unit adopts droop control based on virtual inductance, designs a droop coefficient, and uses current loop control to achieve control;

[0055] A lithium battery, a second DC / DC converter connected to the lithium battery, the second DC / DC converter connected to the lithium battery control unit and the DC bus;

[0056] The lithium battery control unit is based on the droop control of the virtual resistor and uses voltage and current double closed-loop PI control to achieve control;

[0057] A supercapacitor, a third DC / DC converter connected to the supercapacitor, and the third DC / DC converter is connected to the supercapacitor control unit and the DC bus;

[0058] The supercapacitor control unit is based on the droop control of the virtual capacitor and uses voltage and current double closed-loop control to achieve control;

[0059] A ship load, a DC / AC converter connected to the ship load, and a DC / AC converter and a DC bus are connected.

[0060] like Figure 3 Shown is the control block diagram of the hybrid droop control strategy based on virtual impedance.

[0061] The hybrid droop control method based on virtual impedance realizes the distribution process of dynamic ship load power among different power supply units such as fuel cells, lithium batteries and supercapacitors as follows:

[0062] Based on the droop control of virtual inductance and configuring the appropriate droop coefficient, the current loop control is used to realize the control of the fuel cell;

[0063] Based on the droop control of virtual resistance and configuring the appropriate droop coefficient, the voltage and current double closed-loop PI control is used to control the lithium battery;

[0064] Based on the droop control of virtual capacitors and configuring appropriate droop coefficients, voltage and current dual closed-loop PI control is used to control the supercapacitor.

[0065] Fuel cells use a virtual inductor droop control method, lithium batteries use a virtual resistor droop control method, and supercapacitors use a virtual capacitor droop control method. Take one group as an example, specifically:

[0066]

[0067]

[0068]

[0069] Where V nom,fc1 (s), V ref,bat1 (s), V ref,sc1 (s) and i o,fc1 (s), i o,bat1 (s), i o,sc1(s) are the frequency domain functions corresponding to the reference voltage and output current of fuel cell power supply unit No. 1, lithium battery power supply unit No. 1 and supercapacitor power supply unit No. 1 respectively; Z * fc (s), Z * bat (s), Z * sc (s) are the expected output impedances of the fuel cell power supply unit, the lithium battery power supply unit and the supercapacitor power supply unit respectively; L vfc1 is the virtual inductor of fuel cell power supply unit No. 1, R vb1 is the virtual resistance of No. 1 lithium battery power supply unit, R vsc1 and C vsc1 They are the virtual resistance and virtual capacitance of supercapacitor power supply unit No. 1 respectively.

[0070] When the fuel cell ship multi-electric hybrid power system is in normal state, that is, when all power supply units are working normally, the load current i load The distribution relationship between the fuel cell, No. 1 lithium battery and No. 1 supercapacitor power supply unit can be derived as follows:

[0071]

[0072] Where ΔV FB (s) = V nom,fc1 (s)-V ref,bat1 (s); ΔV BS (s) = V ref,bat1 (s)-V ref,sc1 (s),S FC (s), S FB (s), S FS (s), S BAT (s), S BS (s) and S SC (s) are:

[0073]

[0074] From formula (5), we can see that by properly configuring the low-pass filter S FC (s), bandpass filter S BAT (s) and high-pass filter S SC The cut-off frequency of (s) can realize the autonomous optimization distribution of load current among different power supply units. In order to ensure that the lithium battery is in normal mode for a long time and prolong its service life, the nominal voltage of the lithium battery is set to:

[0075]

[0076] For the decentralized dynamic power allocation method for fuel cell ships based on autonomous frequency division, the selected parameters of each supply unit and droop control parameters are shown in Table 1.

[0077] Table 1 Control parameters of the decentralized dynamic power allocation method for fuel cell ships based on autonomous frequency division

[0078]

[0079]

[0080] The process of implementing the droop control based on virtual inductance and configuring a suitable droop coefficient and using current loop control to control the fuel cell is as follows:

[0081] The reference output current value of the fuel cell is obtained by multiplying the difference between the nominal voltage of the fuel cell and the actual output voltage of the fuel cell by the virtual impedance of the fuel cell.

[0082] The output current value based on the fuel cell reference is multiplied by the output voltage to obtain the output power reference value, and then divided by the fuel cell input voltage to obtain the fuel cell input inductor current reference value.

[0083] The reference value of the fuel cell input inductor current is subtracted from the actual input inductor current value and then adjusted by the current loop PI regulator to obtain a pulse width modulation wave to control the converter MOSFET transistor to achieve the steady-state and low-frequency power of the fuel cell in response to the load, thereby achieving the purpose of controlling the power output of the fuel cell.

[0084] The process of implementing the droop control based on virtual resistance and configuring a suitable droop coefficient and using voltage and current double closed-loop PI control to control the lithium battery is as follows:

[0085] Based on the nominal voltage of the lithium battery minus the product of the lithium battery output current and the lithium battery virtual impedance, the lithium battery output voltage reference value is obtained.

[0086] Based on the difference between the lithium battery output voltage reference value and the actual lithium battery voltage output value, the lithium battery inductor current reference value is obtained through voltage outer loop regulation.

[0087] Based on the difference between the lithium battery inductor current reference value and the actual lithium battery inductor current value, a pulse width modulation wave is obtained through current inner loop regulation to control the converter MOSFET transistor, so that the lithium battery responds to the intermediate frequency power of the load, thereby achieving the purpose of controlling the lithium battery power output.

[0088] The process of implementing the droop control based on virtual capacitor and configuring a suitable droop coefficient and using voltage and current double closed-loop PI control to control the super capacitor is as follows:

[0089] The supercapacitor output voltage reference value is obtained based on the supercapacitor nominal voltage minus the product of the supercapacitor output current and the supercapacitor virtual impedance;

[0090] Based on the difference between the supercapacitor output voltage reference value and the supercapacitor actual voltage output value, the supercapacitor inductor current reference value is obtained through voltage outer loop regulation.

[0091] Based on the difference between the supercapacitor inductor current reference value and the actual supercapacitor inductor current value, a pulse width modulation wave is obtained through current inner loop regulation to control the converter MOSFET transistor, so that the supercapacitor responds to the pulsation and high-frequency power of the load, thereby achieving the purpose of controlling the supercapacitor power output.

[0092] Different droop control methods are designed for units with different characteristics, that is, the fuel cell adopts virtual inductance droop control to achieve response to long-term stable load, which is suitable for handling low-frequency power fluctuations; the lithium battery adopts virtual resistance droop control to quickly adjust the medium-frequency load changes, provide necessary dynamic power supplement and storage of excess energy; the supercapacitor adopts virtual capacitor droop control to quickly respond to high-frequency instantaneous power demand, greatly improving the power management capability and operational flexibility of the ship in complex sea conditions. This strategy can adjust the response speed and stability of each energy source to load changes, thereby ensuring the overall efficiency and reliability of the fuel cell ship multi-electric hybrid power system.

[0093] like Figure 4 This is a control block diagram of the state of charge balancing strategy based on adaptive droop control, which consists of a communication layer, a droop control layer and a direct control layer.

[0094] Furthermore, the method also includes a state of charge balancing method based on adaptive droop control, which achieves state of charge balancing of parallel lithium battery packs.

[0095] The controller of lithium battery consists of communication layer, droop control layer and direct control layer.

[0096] The state of charge equalization method based on adaptive droop control realizes the process of equalizing the state of charge of parallel lithium battery packs as follows:

[0097] At the general layer, the average state of charge, average output voltage and average virtual voltage drop of the parallel lithium battery pack are obtained through the dynamic consistency algorithm;

[0098] In the droop control layer: according to the droop coefficient formula of lithium battery, the droop coefficient is obtained;

[0099] Also includes:

[0100] A voltage compensator is used to obtain a voltage compensation amount based on the difference between the average output voltage of the parallel lithium battery group and the actual bus voltage through a PI regulator;

[0101] Virtual voltage drop compensator: used to obtain virtual voltage drop compensation based on the difference between the average virtual voltage drop of the parallel lithium battery pack and the actual virtual voltage drop through the PI regulator;

[0102] Soc equalizer: used to superimpose the droop coefficient, voltage compensation and virtual voltage drop compensation on the voltage reference value, so as to achieve the balance of the charge state of the parallel lithium battery pack.

[0103] Direct control layer: used to make adaptive adjustments based on the deviation between the actual and average values, and compensate for the voltage and virtual voltage drop through the PI regulator to adjust the charge and discharge status of each battery to ensure a balanced charge state of the parallel lithium battery pack.

[0104] The present invention adopts the ampere-hour integration method to calculate the state of charge of the lithium battery, which is specifically expressed as follows:

[0105]

[0106] Where SoC0 is the initial state of charge, i o,bati is the output current of the lithium battery.

[0107] On the basis of the traditional droop control method, a state of charge equalizer and a virtual voltage drop equalizer are added. The state of charge equalizer is used to adjust the droop coefficient in real time, and the virtual voltage drop equalizer is used to dynamically eliminate the influence of line impedance on the current distribution accuracy, so that the state of charge is balanced while the current is accurately distributed in proportion to the capacity of the lithium battery. Among them, the lithium batteries communicate with each other using a dynamic consistency algorithm. Each lithium battery only needs to communicate with the lithium batteries of the adjacent nodes to obtain global information, avoiding dependence on the central controller, and enhancing the plug-and-play, dynamics and stability of the system. The virtual voltage drop equalizer is responsible for collecting the system average virtual voltage drop V vi,avg And the lithium battery output current i o,bati , droop coefficient R i , and calculate the local virtual voltage drop i o,bati R i . V vi,avg with i o,bati R i After the difference is adjusted by the PI controller, the virtual pressure drop compensation Δu I_i Superimposed on the droop control output voltage reference value V ref_i It plays a role in regulating the output current. In addition, a voltage compensator is added. Among them, the output voltage U out,bati The average output voltage V avg Used to measure the voltage level of the DC bus, V avg with U out,bati The difference between the two is passed through the PI regulator to generate the voltage compensation value ΔuV_i , to compensate for the bus voltage drop.

[0108] When a lithium battery is discharged, if the state of charge is large, it should output a larger current, that is, a smaller droop coefficient is used; if the state of charge is small, it should output a smaller current, that is, a larger droop coefficient is used.

[0109] When the lithium battery is charged, it is the opposite to when it is discharged. The expression of the droop coefficient formula of the lithium battery is shown in formula (7):

[0110]

[0111] Among them, SoCbat,i,R 0i are the state of charge of the lithium battery and the initial value of the droop coefficient; SoCavg is the average value of the state of charge of the lithium battery; i o ,bati>0 means the lithium battery is discharged, i o ,bati<0 means lithium battery charging; n is the speed adjustment coefficient.

[0112] Example 1: Assuming that the initial charge states of lithium battery No. 1 and lithium battery No. 2 are 40% and 35% respectively, and the fuel cell and supercapacitor are in normal operation. The focus is on verifying the effectiveness of the proposed decentralized dynamic power allocation method for fuel cell ships based on autonomous frequency division during the normal operation of the fuel cell ship multi-electric hybrid power system.

[0113] from Figure 5 As can be seen in (a), the fuel cell responds slowly to the dynamic load power fluctuation of the ship, and only provides low-frequency load power when the ship load power changes. Both lithium batteries and supercapacitors respond quickly to load power changes and provide medium-frequency power and high-frequency power, respectively.

[0114] from Figure 5 As can be seen in (b), the initial state of charge of the parallel lithium battery group is lower than the lower limit of the state of charge of 45% in normal operation, so both groups of lithium batteries will be charged, and the state of charge will be balanced and the charging power will be shared in about 40s. When the state of charge of the two lithium batteries is balanced, the steady-state charge and discharge power of the supercapacitor and lithium battery tends to zero, and the steady-state load power is provided by the two fuel cells.

[0115] from Figure 5 As can be seen in (c), the DC bus voltage is always within the allowable fluctuation range (±10% Vbus), and its steady-state value can be maintained at the rated value of 700V, ensuring the stable operation of the system.

[0116] In summary, the simulation results of Case 1 show that the proposed decentralized dynamic power allocation method for fuel cell ships based on autonomous frequency division can achieve dynamic power frequency division response of ship load, rapid balancing of the state of charge of parallel lithium battery packs and stable recovery of DC bus voltage, thereby ensuring safe and reliable navigation of fuel cell ships in complex marine environments.

[0117] Example 2: During the actual navigation of a ship, the fuel cell is a component with a high failure rate. Therefore, this case focuses on verifying the effectiveness of the proposed decentralized dynamic power allocation method for fuel cell ships based on autonomous frequency division after two groups of fuel cells fail in sequence. Assuming that the initial state of charge of the parallel lithium battery groups is 65%, fuel cell No. 1 and fuel cell No. 2 stop operating due to failure at t = 15s and t = 30s respectively.

[0118] from Figure 6 As can be seen in (a), the fuel cell, lithium battery and supercapacitor respond to the low-frequency, medium-frequency and high-frequency ship dynamic load power respectively. When the two groups of fuel cells stop running in turn due to a fault, the lithium battery and supercapacitor can still reasonably distribute the load power according to their respective droop coefficients.

[0119] from Figure 6 As can be seen from (b), the initial charge states of the parallel lithium battery packs are all within the normal operating range and vary in the same manner, so the steady-state charge and discharge power of the parallel lithium battery packs tends to zero.

[0120] from Figure 6 As can be seen in (c), the withdrawal of fuel cell No. 1 has almost no effect on the DC bus voltage, and its steady-state value can still be stabilized at 700 V. After the withdrawal of fuel cell No. 2, the steady-state DC bus voltage dropped by about 20 V, which is due to the steady-state voltage drop caused by the droop coefficient, but the proposed method can still maintain the DC bus voltage within the allowable range.

[0121] In summary, the simulation results of Example 2 show that the proposed decentralized dynamic power distribution method for fuel cell ships based on autonomous frequency division can achieve dynamic load power distribution and DC bus voltage stability control. However, it should be noted that the capacity of lithium batteries and supercapacitors is low and can only maintain power supply to ship loads for a short time, so they cannot be maintained for a long time under the condition of fuel cell failure. In the actual application of ship navigation, the faulty fuel cell should be replaced or repaired as soon as possible to ensure that the ship continues to sail reliably.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A decentralized dynamic power allocation method for fuel cell ships based on autonomous frequency division, characterized in that: The following steps are involved: Establish a fuel cell ship multi-electric hybrid power system model with fuel cells, lithium batteries, supercapacitors, DC / DC converters and ships as loads; Based on the fuel cell ship multi-electric hybrid power system model, a hybrid droop control method with virtual impedance is adopted to realize the distribution of dynamic ship load power among different power supply units such as fuel cell, lithium battery and supercapacitor.

2. According to claim 1, a method for distributed dynamic power allocation of fuel cell ships based on autonomous frequency division is characterized in that: The hybrid droop control method using virtual impedance is used to achieve the distribution of dynamic ship load power among different power supply units such as fuel cells, lithium batteries and supercapacitors as follows: Based on the droop control of virtual inductance, the fuel cell is controlled by current loop control; Based on the droop control of virtual resistance, the voltage and current double closed-loop PI control is used to control the lithium battery; Based on the droop control of virtual capacitor, voltage and current double closed-loop PI control is adopted to realize the control of supercapacitor.

3. According to claim 1, a decentralized dynamic power allocation method for fuel cell ships based on autonomous frequency division is characterized in that: It also includes a state of charge balancing method based on adaptive droop control, which realizes the state of charge balancing of parallel lithium battery packs.

4. According to claim 1, a method for distributed dynamic power allocation of fuel cell ships based on autonomous frequency division is characterized in that: The process of implementing the control of the fuel cell by using the current loop control based on the virtual inductor droop control is as follows: The reference output current value of the fuel cell is obtained by multiplying the difference between the nominal voltage of the fuel cell and the actual output voltage of the fuel cell by the virtual impedance of the fuel cell. The output current value based on the fuel cell reference is multiplied by the output voltage to obtain the output power reference value, and then divided by the fuel cell input voltage to obtain the fuel cell input inductor current reference value. The reference value of the fuel cell input inductor current is subtracted from the actual input inductor current value and then adjusted by the current loop PI regulator to obtain a pulse width modulation wave to control the converter MOSFET transistor to achieve the steady-state and low-frequency power of the fuel cell in response to the load, thereby achieving the purpose of controlling the power output of the fuel cell.

5. According to claim 1, a fuel cell ship decentralized dynamic power allocation method based on autonomous frequency division is characterized in that: The process of controlling the lithium battery by using the voltage and current double closed-loop PI control based on the virtual resistance droop control is as follows: Based on the nominal voltage of the lithium battery minus the product of the lithium battery output current and the lithium battery virtual impedance, the lithium battery output voltage reference value is obtained. Based on the difference between the lithium battery output voltage reference value and the actual lithium battery voltage output value, the lithium battery inductor current reference value is obtained through voltage outer loop regulation. Based on the difference between the lithium battery inductor current reference value and the actual lithium battery inductor current value, a pulse width modulation wave is obtained through current inner loop regulation to control the converter MOSFET transistor, so that the lithium battery responds to the intermediate frequency power of the load, thereby achieving the purpose of controlling the lithium battery power output.

6. According to claim 1, a method for distributed dynamic power allocation of fuel cell ships based on autonomous frequency division is characterized in that: The process of controlling the supercapacitor by using voltage and current double closed-loop PI control based on virtual capacitor droop control is as follows: The supercapacitor output voltage reference value is obtained based on the supercapacitor nominal voltage minus the product of the supercapacitor output current and the supercapacitor virtual impedance; Based on the difference between the supercapacitor output voltage reference value and the supercapacitor actual voltage output value, the supercapacitor inductor current reference value is obtained through voltage outer loop regulation. Based on the difference between the supercapacitor inductor current reference value and the actual supercapacitor inductor current value, a pulse width modulation wave is obtained through current inner loop regulation to control the converter MOSFET transistor, so that the supercapacitor responds to the pulsation and high-frequency power of the load, thereby achieving the purpose of controlling the supercapacitor power output.

7. According to claim 1, a method for distributed dynamic power allocation of fuel cell ships based on autonomous frequency division is characterized in that: The state of charge equalization method based on adaptive droop control realizes the process of equalizing the state of charge of parallel lithium battery packs as follows: The average state of charge, average output voltage and average virtual voltage drop of the parallel lithium battery pack are obtained through the dynamic consistency algorithm; According to the droop coefficient formula of lithium battery, the droop coefficient is obtained; The voltage compensation is obtained by the PI regulator based on the difference between the average output voltage of the parallel lithium battery group and the actual bus voltage; The virtual voltage drop compensation is obtained by using a PI regulator based on the difference between the average virtual voltage drop and the actual virtual voltage drop of the parallel lithium battery pack. The droop coefficient, voltage compensation amount and virtual voltage drop compensation amount are all superimposed on the voltage reference value to achieve a balanced charge state of the parallel lithium battery pack.

8. According to claim 1, a method for distributed dynamic power allocation of fuel cell ships based on autonomous frequency division is characterized in that: The droop coefficient formula of the lithium battery is expressed as follows: Among them, SoC bati , R 0i They are the state of charge of the lithium battery and the initial value of the droop coefficient; SoC avg is the average state of charge of the lithium battery; i o,bati >0 means the lithium battery is discharged, i o,bati <0 indicates lithium battery charging; n is the speed adjustment coefficient.

9. A fuel cell ship distributed dynamic power distribution control system based on autonomous frequency division is characterized by: include: A fuel cell, a first DC / DC converter connected to the fuel cell, the first DC / DC converter connected to the fuel cell control unit and the DC bus; The fuel cell control unit adopts droop control based on virtual inductance, designs a droop coefficient, and uses current loop control to achieve control; A lithium battery, a second DC / DC converter connected to the lithium battery, the second DC / DC converter connected to the lithium battery control unit and the DC bus; The lithium battery control unit is based on the droop control of the virtual resistor and uses voltage and current double closed-loop PI control to achieve control; A supercapacitor, a third DC / DC converter connected to the supercapacitor, and the third DC / DC converter is connected to the supercapacitor control unit and the DC bus; The supercapacitor control unit is based on the droop control of the virtual capacitor and uses voltage and current double closed-loop control to achieve control; A ship load, a DC / AC converter connected to the ship load, and a DC / AC converter and a DC bus are connected.

Images