A nonlinear droop control method for direct current power distribution network

Abstract

This invention relates to the field of DC distribution networks, specifically to a nonlinear i-type DC distribution network. d -v dc 2 Droop control method. Compared with the traditional dual-loop droop control method, this method improves dynamic performance to a certain extent; compared with i d -v dc This method significantly optimizes the steady-state performance of the system. The control system of this method includes: i d -v dc 2 The system comprises a droop control module, a phase-locked loop (PLL), an inductor filter, a current inner-loop PI controller, a decoupling module, a Park converter module, a PWM modulation circuit, and a DC voltage regulator capacitor. This method controls the active power of the system through the AC current in the internal loop. Based on this, it uses the squared voltage difference as the droop input and directly outputs an AC current reference value from the droop characteristic, eliminating external DC voltage or current loops. This makes the AC current directly related to the load distribution performance of the DC distribution network. The adoption of this method will effectively improve the system's voltage management performance and current / power sharing performance, while ensuring good stability and dynamic performance; therefore, its application prospects are promising.

Description

Technical Field

[0001] This invention relates to the field of DC distribution network operation control, and more specifically, to a nonlinear i-type DC distribution network. d -v dc 2 Methods for controlling drooping. Background Technology

[0002] With the increasing prevalence of renewable energy in power systems, the application of microgrids in DC distributed power sources such as photovoltaics, wind turbines, fuel cells, and energy storage systems, as well as DC loads such as electric vehicles, has received more attention and is of great significance for future smart distribution systems. Compared with AC microgrids, DC distribution networks have the ability to efficiently integrate renewable energy and storage devices, reduce power conversion losses, and eliminate the need for reactive power management and frequency synchronization. Therefore, in recent years, DC distribution networks have been widely used in the power systems of ships and multi-electric aircraft.

[0003] In multi-source DC distribution networks, parallel power sources can reduce the weight of the main generator and the redundancy of the power system. To ensure proper power distribution among parallel sources, two main strategies exist: active power sharing and passive power sharing. Active power sharing includes master-slave control, centralized control, and average current control, but these strategies rely on communication infrastructure between parallel systems and have certain limitations. Passive load sharing, such as decentralized droop control (i.e., introducing virtual resistance at the converter's output port), avoids the aforementioned communication requirements, greatly improving system reliability, and is therefore widely used in DC power systems.

[0004] The control principle diagram of the traditional current-mode linear droop strategy is as follows: Figure 3 As shown. The actual sampled DC voltage value serves as the input to the IV droop control module, and the output DC current reference value i... dc * The difference between the reference value and the actual sampled current value is adjusted by the PI control in the outer current loop to output the d-axis AC current reference value i. d * The relevant droop control equation is shown below:

[0005]

[0006] If we consider directly connecting the AC current control to the power distribution performance of the DC distribution network, we can eliminate the cascaded current loop, meaning the droop control directly outputs the d-axis current reference value i. d * To achieve a faster dynamic process, this method is called AC-DC coupled droop control strategy. Its control principle diagram is shown below. Figure 4 As shown. The actual sampled DC voltage value is used as i.d -v dc The input and output i of the droop control module d * The relevant droop control equation is shown below:

[0007]

[0008] Traditional droop control strategies strike a trade-off between voltage regulation and load sharing accuracy. DC distribution networks with a large droop factor exhibit good power sharing performance, but suffer from large bus voltage deviations. On the other hand, reducing the droop factor improves voltage regulation, but results in poor load sharing performance. Even when AC-DC coupled droop strategies effectively improve load sharing accuracy, the problem of large voltage deviations persists with a large droop factor. Summary of the Invention

[0009] The purpose of this invention is to overcome the shortcomings of the existing technologies and solutions mentioned above, and to propose a nonlinear i d -v dc 2 Droop control method. This method, based on the active power of the AC current control system within the internal loop, uses the voltage square difference as the droop input and directly outputs the AC current reference value from the droop characteristic. Compared to the traditional IV droop characteristic, it eliminates the external DC current loop, making the AC current directly related to the load distribution performance of the DC distribution network, shortening the circuit response time, and improving the system's dynamic performance and stability; while also improving the AC / DC coupling... d -v dc Compared to the droop characteristic, this method effectively enhances the voltage management performance of the system. That is, under the same droop coefficient, it greatly reduces the voltage deviation from the rated voltage, and at the same time improves the load distribution accuracy of multi-source and multi-load systems.

[0010] The principle of this invention is as follows:

[0011] A nonlinear i-type DC distribution network d -v dc 2 The method for controlling drooping includes the following steps:

[0012] S1. The active component reference value of the AC current is directly generated by the droop control, and the square difference between the sampled voltage and the rated voltage is used as the input of the droop control.

[0013] i d -v dc 2 Control principle diagram as follows Figure 5 As shown. The actual sampled DC voltage value serves as the input to the IV droop control module, and the output i... d *The relevant droop control equation is shown below:

[0014]

[0015] In the formula, V0 is the rated voltage, v dc To sample DC voltage, i d * represents the generated d-axis current reference, and k is the droop factor. (and i) d -v dc Comparing the droop control equation of the method, it can be seen that this method is equivalent to multiplying the original voltage deviation value by (V0 + v). dc This will necessarily result in i being in the same working environment and with the same k value. d -v dc 2 This method will greatly reduce the voltage drop.

[0016] S2. Based on step S1, the droop characteristics of DC voltage and DC current under steady state are derived, which can achieve superior voltage management (current and power distribution performance will be described in the specific implementation).

[0017] Depend on Figure 2 From the classic vector control diagram of the dq axis, we can see that the voltages of the d-axis and q-axis can be expressed as follows:

[0018]

[0019] In the formula, e d e q It is the bus voltage at the common coupling point; I d V d These are the d-axis current and voltage values ​​in steady state; R s L s These are the AC side resistance and inductance; ω e It is the frequency of the AC side source; V dc I dc These are the DC voltage and current values ​​under steady-state conditions, respectively.

[0020] Based on d-axis current control of active power, it is assumed that reactive power is equal to zero, i.e., e q and i q i is 0 in steady state d -v dc 2 The relevant characteristics can be expressed by the following formula:

[0021]

[0022] Combining the above three equations, we can obtain the external V-I droop characteristic of this method, as shown in the following formula:

[0023]

[0024] When the system uses a constant power load and the line impedance is ignored, the DC voltage and current satisfy the following equation:

[0025]

[0026] The formula for calculating the steady-state value of DC voltage under constant power load conditions can be obtained as follows:

[0027]

[0028] AC / DC coupling under constant power load conditions d -v dc The method can also be used to derive the formula for calculating the steady-state value of its DC voltage:

[0029]

[0030] i d -v dc 2 The method is equivalent to replacing V0 with its square value and taking the root value of the DC voltage. From the above two equations, it can be seen that under completely identical operating conditions, when the system takes a larger k value, i... d -v dc 2 This method will greatly reduce the voltage drop.

[0031] Furthermore, in order to study the number of intersection points between the droop characteristics and the constant power load curve in order to design the droop coefficient, the existence of this equilibrium point V must first be satisfied. (1) conditions, namely:

[0032]

[0033] From this, we can deduce condition 1:

[0034]

[0035] When deriving the DC voltage value under a constant power load environment, it can be found that the curve has a second intersection point. However, this value is not the operating point when the system is running stably. Its expression is:

[0036]

[0037] If V (2) It also exists, meaning that as long as the expression inside the square root is greater than 0, condition 2 can be derived:

[0038]

[0039] Under the combined constraints of conditions 1 and 2, k exists within a range where a constant power load will intersect the drooping external characteristic curve at two points. At the intersection point with a larger voltage value, the system can operate stably.

[0040] S3. Based on step S1, derive the small-signal model corresponding to this method, and obtain the source-side and load-side impedance models, thereby completing the system stability analysis:

[0041] S3.1 Model the internal current loop and the small-signal voltage of the full closed loop according to the system control block diagram.

[0042] From equation (6), we can deduce i d The transfer function of the output quantity of the internal current PI loop is:

[0043]

[0044] In the formula, v out (s) is the expression for the output of the internal current PI loop in the frequency domain.

[0045] To design the internal current loop as a simple first-order system with bandwidth ω c =1 / τ, the PI parameter can be designed as:

[0046] k pi =ω c L s ,k ii =ω c R s (16)

[0047] In the formula, k pi k is a proportional parameter. ii Let τ be the integration parameter and τ be the time constant.

[0048] Therefore, the transfer function of the internal current loop can be obtained as follows:

[0049]

[0050] Depend on Figure 1 As can be seen from the architecture, the linearized relationship between the DC output current and voltage is as follows:

[0051]

[0052] Therefore, the transfer function of the output voltage and current can be derived as follows:

[0053]

[0054] The control block diagram of the system after linearization is as follows: Figure 6 As shown, the transfer function from the control output to the system output is:

[0055]

[0056] In the formula, v dc0 i is the steady-state value of the DC voltage. d0 This represents the steady-state value of the d-axis current.

[0057] Therefore, the transfer function of the fully closed-loop voltage small-signal model can be derived as follows:

[0058]

[0059] S3.2 Model the source-side and load-side impedances according to step S3.1, and study the influence of the current inner loop bandwidth and droop coefficient on system stability based on the Middlebrook criterion.

[0060] Further analysis of the small-signal modeling above reveals that the source side can be written as an equivalent voltage source in series with an impedance:

[0061]

[0062] Where, Δv eq That is, the equivalent voltage source, Z eq This is the equivalent impedance.

[0063] Therefore, if we consider the source-side system and the load system separately, the source side is replaced by an equivalent voltage source and an impedance module in series, while the load side is selected as a controllable Buck model as a constant power load, and impedance modeling is performed separately. A simplified diagram is shown below. Figure 7 As shown.

[0064] Therefore, the impedances on both sides can be obtained as follows:

[0065]

[0066] Among them, R buck C buck ,L buck The values ​​of resistance, capacitance, and inductance of the buck circuit are ω, respectively. L This refers to the control bandwidth of the buck circuit.

[0067] And source side Z eq It can be further reduced to a parallel network of three series-connected RLC networks. The equivalent diagram of the source-side impedance RLC network is shown below. Figure 8 As shown. The specific values ​​of each parameter are as follows:

[0068]

[0069] It can be seen that L1 / R1=τ, R2*C2=τ. The derivation of this model helps to analyze system stability from a circuit perspective, such as parameters like the resonant frequency and peak values.

[0070] Compared with the prior art, the beneficial effects of the present invention are: under the same droop gain, i d -v dc 2 The method exhibits smaller voltage deviation and expands the feasible range of droop gain; the results were validated through dynamic modeling. d -v dc 2 The system under the control method has the fastest response speed and the largest stability margin. Attached Figure Description

[0071] Figure 1 This is a schematic diagram of the DC power distribution network architecture of the present invention.

[0072] Figure 2 This is a schematic diagram of the vector control principle of the present invention.

[0073] Figure 3 This is a control scheme diagram for a traditional IV droop control system.

[0074] Figure 4 For the AC / DC coupling of this invention d -V dc Control scheme diagram of the droop control system.

[0075] Figure 5 For the nonlinear i of the present invention d -V dc 2 Control scheme diagram of the droop control system.

[0076] Figure 6 This is the equivalent control block diagram of the present invention after linearization.

[0077] Figure 7 This is a simplified diagram of the source-side and load-side system of the present invention.

[0078] Figure 8 This is the equivalent diagram of the source-side impedance RLC network of the present invention.

[0079] Figure 9 The traditional IV ptosis method and the i studied in this patent d -v dc 2 Time-domain simulation diagram of the drooping method.

[0080] Figure 10 The traditional IV ptosis method and the i studied in this patent d -v dc 2Comparison of voltage management methods using the drooping approach.

[0081] Figure 11 The traditional IV ptosis method and the i studied in this patent d -v dc 2 Current time-domain simulation diagram of the droop method. Detailed Implementation

[0082] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments, thereby enabling a full understanding of how the present invention uses technical means to solve technical problems and achieve technical effects, and allowing for implementation accordingly. This embodiment is based on the technical solution of the present invention and provides detailed implementation methods and specific operating procedures; however, the scope of protection of the present invention is not limited to the following embodiments.

[0083] Example:

[0084] DC distribution network architecture such as Figure 1 As shown, multiple voltage sources powered by distributed generators are connected to the same DC bus, forming a bus voltage and supplying power to the load. i For local capacitors, and C b R is the common capacitor connected in parallel on the DC bus side. i and L i This is the line impedance connecting the source-side converter and the DC bus. In the specific implementation below, i = 1 when studying the system voltage management performance, and i = 1, 2 when studying the power distribution performance.

[0085] Figure 9 The image shows the traditional IV ptosis method and the method proposed in this patent. d -v dc 2 Time-domain simulation of the droop method. The constant power load is increased to twice the initial value at 1 second and to three times the initial value at 2 seconds. As shown in the figure, compared to the traditional IV droop method (red line), i d -v dc 2 The method (blue line) effectively reduces response time and enhances the dynamic performance of the system.

[0086] Figure 10 In the middle, (1) figure shows i d -v dc The intersection points and voltage drop of the constant power load curves when the droop method is selected with droop coefficients of 1, 20, 40, and 1000 are shown in Figure (2). d -v dc 2 The corresponding situation for the drooping method. It can be seen that when k = 1000, i d -vdc The droop curve and the CPL curve no longer intersect, meaning the system has no operating point. At this droop coefficient, i... d -v dc 2 The voltage drop exhibited by the drooping method remains relatively small. Meanwhile, for the same k value, i d -v dc 2 All are better than i d -v dc The voltage drop of this method is significantly reduced, which is a major advantage of this method.

[0087] Figure 11 In the middle, (1) figure shows i d -v dc Time-domain simulation of distributed source current and power allocation under the droop method, (2) shows i d -v dc 2 The results for the droop method are shown in Table 1. The line resistances of the two parallel power supplies are set to 0.1 and 0.2 ohms respectively, and droop coefficients that result in a bus voltage of 260V are selected for both methods, while other system parameters remain the same. It can be seen that i d -v dc 2 Slightly better than i d -v dc The current and power distribution performance of the method.

[0088] As can be seen from this embodiment, the nonlinear i-type DC distribution network proposed in this invention... d -v dc 2 Compared to the traditional dual-loop droop control method, this droop control method improves dynamic performance to some extent; compared to i d -v dc The method significantly optimizes the system's voltage management performance and current and power sharing performance.

[0089] Although the present invention has been described in detail through the above embodiments, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above content. Therefore, the scope of protection of the present invention should be defined by the appended claims.

[0090] The parameters of the main components are as follows:

[0091] AC power supply: fundamental frequency: 50Hz, peak phase voltage: 100V; inductor filter L s :3mH;

[0092] Parasitic resistance R of inductor filter s 0.2Ω;

[0093] Line inductance L line : 65μH;

[0094] Line resistance R line 0.05Ω;

[0095] Local capacitor C: 1.6mF;

[0096] Bus capacitor C b 0.6mF;

[0097] Constant power load P L 1000W (single source); 2000W (dual source) DC bus voltage reference value V0: 270V

[0098]

[0099] Table 1.

Claims

1. A nonlinear droop control method for a DC distribution network, wherein the DC distribution network comprises multiple voltage sources connected to the same DC bus to form a bus voltage and supply power to the load, characterized in that, The method includes: S1. The active component reference value of the AC current generated by the droop control, and the square difference between the sampled voltage and the rated voltage is used as the input of the droop control. S2. Assuming reactive power is zero, calculate the steady-state DC voltage V. dc With steady-state DC current I dc The drooping characteristic is expressed by the following formula: In the formula, e d It is the bus voltage at the common coupling point, R s This is the AC side resistance, where V0 is the rated voltage and k is the droop factor. S3. Determine the transfer function G of the fully closed-loop voltage small-signal model based on step S1. vdc (s), the formula is as follows: In the formula, C i For local capacitors, P L For a constant power load, G VSC (s) is the transfer function from the control output to the system output. Among them, v dc0 The steady-state value of the DC voltage, v d0 i d0 R represents the steady-state values ​​of the d-axis voltage and current. s ,L s Let τ be the resistance and inductance on the AC side, and τ be the time constant of the internal current loop.

Images