Distributed droop control method and device suitable for direct current micro-grid system and capable of improving fixed time consistency

Abstract

The invention discloses an improved fixed-time-consistency distributed droop control method and device suitable for a direct-current micro-grid system, and the method comprises the steps: constructing a distributed communication and control architecture without a central node in a distributed direct-current micro-grid system; an improved fixed time consistency algorithm is constructed on the communication and control architecture; optimizing droop control, and introducing the fixed time convergence characteristic of the improved fixed time consistency algorithm into droop control of the DC micro-grid; a simulation model with multiple distributed power supplies connected in parallel is built based on Matlab / Simulink, the distributed droop control method based on the improved fixed time consistency is adopted for control, and convergence effects of different algorithms under the same and different initial conditions are compared and analyzed; the simulation model respectively simulates control effects under normal operation and load switching, and verifies that the voltage stabilization performance and the current sharing effect of the method are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of inverter control technology, and in particular to a distributed droop control method and apparatus based on improved fixed-time consistency suitable for DC microgrid systems. Background Technology

[0002] With the global energy structure transformation and the development of renewable energy technologies (such as photovoltaics and wind power), the number of DC power devices is gradually increasing, and related research on DC microgrids is attracting more and more attention from academia and industry. DC microgrids are easy to integrate renewable energy sources, and compared with AC microgrids, they do not need to consider frequency and reactive power issues. They have broad development prospects in achieving coordinated control of distributed power sources within a microgrid.

[0003] Due to the intermittency and uncertainty of renewable energy sources, system power fluctuates, affecting the stable operation of microgrids. Microgrids, through distributed control strategies, can continue to stably supply power to critical loads during local system fluctuations or faults, significantly improving system robustness. The core of stable operation of a DC microgrid is achieving the control objective of efficient coordination of distributed generators (DGs), namely, ensuring the DC bus voltage remains stable at its rated value and the precise distribution of output current (or power) among each DG. Droop control, as a classic local control strategy, is simple and reliable to implement, offering advantages such as plug-and-play functionality and no reliance on communication networks, and is widely used in DC microgrids.

[0004] However, droop control, by introducing virtual resistance to adjust the output power of each distributed power source, inevitably causes the bus voltage to drop. Furthermore, in actual microgrids, factors such as line impedance mismatch and sensor sampling errors will also affect the accuracy of current distribution. In other words, relying solely on droop control cannot simultaneously achieve the control objectives of stable bus voltage and high-precision balanced distribution of output current. Summary of the Invention

[0005] The present invention aims to overcome the above-mentioned shortcomings of the prior art and provide an improved fixed-time consistency distributed droop control method and apparatus suitable for DC microgrid systems.

[0006] This invention improves the fixed-time consensus algorithm by adjusting its gain coefficient and convergence coefficient, and combines it with optimized droop control to obtain consistent output currents for each DG. This consistency is then used to dynamically adjust the droop coefficient, effectively solving the problem of uneven current distribution caused by line impedance mismatch and simultaneously eliminating bus voltage deviation. Furthermore, this invention combines the advantages of low communication resource consumption of the consensus algorithm with the stability and reliability of the droop control method. Compared to traditional droop control methods, the proposed control method significantly improves current sharing control accuracy, convergence response speed, bus voltage recovery speed, and bus voltage steady-state deviation.

[0007] To achieve the above objectives, a first aspect of the present invention relates to an improved fixed-time consistent distributed droop control method suitable for DC microgrid systems, characterized by comprising the following steps: S1. In a distributed DC microgrid system, a distributed communication and control architecture without a central node is constructed. The communication between each DG is abstracted into the topological relationship of a connected graph. Each DG only interacts with its adjacent nodes, which improves the single-point dependence and high communication pressure defects in the traditional centralized control method. S2. Based on the communication architecture constructed in step S1, and using the consensus algorithm, an improved fixed-time consensus algorithm is constructed by designing the algorithm framework and adjusting the gain coefficient, convergence coefficient, etc., which has an upper bound on the convergence time that is independent of the initial running state. T This overcomes the problems of slow convergence and uncertain convergence time in traditional consensus algorithms; S3. Optimize the droop control method by introducing the fixed-time convergence characteristic of the improved fixed-time consensus algorithm obtained in step S2 into the DC microgrid droop control. Adjust the droop coefficient using the consistency results of the output current of each DG. K i ' and bus compensation voltage Δ u i Simultaneous real-time dynamic adjustment is achieved, so that the output current of all DGs converges to a consistent state within the fixed convergence time upper limit, and its convergence time is independent of the initial state of the system. The bus voltage is restored through the voltage control loop, which solves the problems of bus voltage deviation caused by line impedance mismatch and uneven current distribution among DGs, and breaks through the limitations of voltage regulation and uneven current distribution in traditional droop control. S4. Based on Matlab / Simulink, build a simulation model of multiple DGs in parallel in a DC microgrid, design the basic parameters of the DC microgrid with multiple DGs, and set typical operating conditions. S5. In the model established in step S4, the microgrid is controlled using the improved fixed-time consistency distributed droop control method obtained in steps S1-S3, and experimental simulation is performed; the droop coefficient correction Δ is observed. K i Observation of reference voltage compensation amount Δ u i Simulation results, and using the DC microgrid bus voltage, etc. U bus Output current of each DG I oi The proposed method demonstrates significant improvements in voltage regulation and current sharing performance through various metrics.

[0008] Preferably, step S1 specifically includes: S101. Build a distributed communication and control framework without a central node, associate the information layer with the physical device layer, and each DG only exchanges voltage / current information with the two physical closest nodes (i.e., adopts adjacent communication) to ensure full connectivity of the communication network. S102. Abstract the communication between DGs into the topological relationships of a connected graph, using an undirected weighted graph. G ( A )=( V , E , A ) indicates that among them V Represents a non-empty set of nodes. E This represents the set of edges between nodes, containing all pairs of neighboring nodes. A= [ a ij Let be the adjacency matrix of the undirected weighted graph, containing the communication weight information between each node, satisfying: (1) In the formula, ( j , i ) E Represents a node j It is a node i One of the neighboring nodes, a ij Representing neighboring nodes i and nodes j Communication weights between them; Global coordination and optimization are achieved through information interaction between adjacent distributed units. The local communication interaction between each DG adopts adjacent communication, which does not require a central controller and can achieve convergence effect with a small communication burden.

[0009] Preferably, step S2 specifically includes: S201. Based on graph theory in step S102, the formula for the consensus algorithm is as follows: (2) In the formula, x i ( t ) is a node i exist t The state at any given moment; N i For nodes i The set of adjacent nodes. As long as the network's communication topology satisfies the connectivity graph, then equation (2) holds true. t As we approach infinity, the following condition is met: (3) That is, after running for a sufficiently long time, it can ensure that the state difference between each node is 0, thus achieving the consistency effect of different nodes; S202. Combining the fixed-time consistency theory, by adjusting the gain coefficient, changing the convergence coefficient, and other parameter settings, an improved fixed-time consistency algorithm is constructed as follows: (4) In the formula, the gain coefficient k >0, convergence coefficient and satisfy >1, 0< <1, sig( z ) = | z | sign( z ), sign(·) is the sign function; The improved fixed-time consensus algorithm will automatically determine state errors. x j ( t )- x i ( t The relationship between the value of the convergence coefficient and the error threshold 1 is used to dynamically select the convergence coefficient, which greatly accelerates the convergence speed while ensuring a fixed convergence time. Overall system convergence time T satisfy: (5) The convergence time depends only on the intrinsic coefficients and is independent of the initial input conditions. This improved fixed-time consistency algorithm achieves convergence in less than [time value missing]. T .

[0010] Preferably, step S3 specifically includes: S301, forN For a parallel DG, an equivalent model is constructed and its voltage and current droop characteristic expression is obtained: (6) In the formula, K Represents virtual impedance. r i and i oi These represent the line impedance and output current of each branch, respectively. u ref Indicates the reference voltage. u bus Indicates the bus voltage; S302. Model the droop control method and construct the dynamic characteristic curve of "output voltage-output current"; S303, Due to line impedance mismatch (i.e.) r i ≠ r j This leads to uneven current distribution, so it's necessary to increase the droop coefficient by an appropriate amount for correction. K i To dynamically adjust the droop coefficient and achieve power distribution, different DGs at this time i They have the same total droop coefficient; The fixed-time convergence characteristic of the improved fixed-time consensus algorithm obtained in step S2 is introduced into the droop control of the DC microgrid. Based on equation (4) and combined with droop control, the consistency calculation of each DG current parameter is performed to obtain the consistency iteration result of the output current. i Xi : (7) In the formula, i Li Indicates each DG i The output current, i Xi This represents the current consistency iteration result. According to consistency theory, after... t < T Time, and the iteration results satisfy: (8) The droop coefficient correction Δ is obtained by updating it using an improved fixed-time consistency algorithm. K i Real-time adjustment of DG's droop coefficient K i ': (9) To satisfy this, for different DGi They all have the same total droop coefficient. K i That is, dynamically adjusting DG using the improved fixed-time consensus algorithm obtained through S2. i The droop coefficient achieves the current equalization effect, so that the output current of all DGs converges to a consistent state within the fixed convergence time upper bound, and its convergence time is independent of the initial state of the system. S304. Due to the bus voltage drop caused by sag and line impedance, increase the reference voltage compensation. u i The bus voltage is stabilized through a secondary voltage control circuit, and after compensation, the bus voltage reaches the reference voltage, thus achieving precise voltage regulation. Based on step S303, a reference voltage compensation amount Δ is introduced into the control system. u i ,satisfy: (10) The compensation amount Δ u i By increasing the original reference voltage, the droop coefficient can be adjusted. K i Correction amount Δ K i and line impedance r i The resulting voltage drop is accurately compensated, and the compensated control strategy can completely restore the actual bus voltage to the rated reference value: (11) That is, the improved fixed-time consistency algorithm obtained through S2 dynamically adjusts the bus reference voltage compensation amount, so that the compensated bus voltage is restored to the reference voltage value, reducing steady-state voltage deviation and improving bus voltage quality; S305. Construct the optimized dynamic characteristic curve of "output voltage-output current".

[0011] Preferably, step S4 specifically includes: S401. A simulation model of a DC microgrid system with multiple DGs in parallel was built based on Matlab / Simulink. The model includes components such as distributed equivalent power sources, line impedance, loads, distributed communication networks, and electronic power converters. An improved fixed-time consistency algorithm control module was also added. S402. Design the basic parameters of a multi-DG DC microgrid and set typical operating conditions, which are divided into: Working condition 1: Performance test of the improved fixed-time consensus algorithm, namely: simulating the convergence effect under various different initial conditions, and verifying the reliability and superiority of the improved fixed-time consensus algorithm in terms of convergence speed by comparing and analyzing it with traditional consensus, finite-time consensus and other algorithms. Operating Condition 2: A sudden change in load occurs in the DC microgrid system. The system load increases at 2 seconds and decreases at 4 seconds. That is, the load power is 16000W in 0-2 seconds, 32000W in 2-4 seconds, and returns to 16000W in 4-6 seconds.

[0012] Preferably, step S5 specifically includes: S501. Based on the DC microgrid system simulation model containing 5 DGs built in step S4, the control method described in steps S1-S3 is put into use. Under the control parameters and simulation operating conditions set in step S4, the actual control effect under different operating conditions of the DC microgrid system is observed. S502, Simulation observation of droop coefficient correction Δ K i Observation of reference voltage compensation amount Δ u i The simulation results are then compared with the theoretical analyses proposed in verification steps S1-S3 to verify their effectiveness and correctness; simultaneously, the bus voltage of the DC microgrid is observed. U bus With reference voltage U ref Deviation rate, output current of each DG I oi The convergence speed and convergence effect are used as indicators to verify the significant improvement of the proposed method in voltage stabilization performance and current equalization effect.

[0013] A second aspect of the invention relates to an improved fixed-time consistent distributed droop control device for DC microgrid systems, comprising a memory and one or more processors, the memory storing executable code, which, when executed by the one or more processors, implements the improved fixed-time consistent distributed droop control method for DC microgrid systems of the invention.

[0014] A third aspect of the invention relates to a computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the improved fixed-time consistent distributed droop control of the invention for DC microgrid systems.

[0015] The innovations of this invention are: constructing a distributed communication and control architecture without a central node, where each DG completes consistency calculations only through local information interaction with neighboring nodes, eliminating the need for global communication and thus improving the consistency convergence accuracy in weak communication scenarios; proposing a distributed droop control protocol based on fixed-time consistency theory, introducing fixed-time convergence characteristics into the droop control of DC microgrids, ensuring that voltage and current consistency errors between DGs converge within a preset time, and that the convergence time is independent of the initial state of the system; and combining an optimized distributed droop control method, dynamically adjusting the droop coefficient through an improved fixed-time consistency algorithm to achieve synchronous fixed-time convergence of voltage deviation and current distribution errors between DGs, and eliminating bus voltage deviation, thus overcoming the limitations of uneven voltage regulation and current distribution in traditional droop control.

[0016] The working principle of this invention is: to adopt a communication mechanism with local communication topology and dynamically adjustable communication weights. When constructing the communication network, the information layer is associated with the physical device layer. Each DG exchanges voltage / current information only with the two physically closest nodes (i.e., adjacent communication), ensuring full connectivity of the communication network. Graph theory principles are used to transform local information into globally consistent decisions, effectively reducing communication pressure compared to traditional centralized control methods. Based on fixed-time stability theory, a consensus protocol including gain coefficients and nonlinear convergence coefficients is designed. This algorithm acquires the current information of each DG node for regulation, ensuring that each DG completes rapid convergence control within a fixed time, and that its convergence time is independent of the system's initial state, effectively overcoming the limitations of traditional consensus algorithms or finite-time constraints. The consensus algorithm suffers from slow convergence speed or convergence time that is highly dependent on the initial state difference (i.e., the greater the initial state difference, the longer the convergence time). Traditional droop control cannot solve the problem of uneven current distribution caused by line impedance mismatch, and it also causes voltage drop leading to bus voltage deviation. This algorithm is combined with optimized droop control. The algorithm adjusts the droop coefficient of each DG current according to the consistency result. The droop coefficient of each DG and the bus compensation voltage are dynamically adjusted in real time to ensure that the current (power) of all DGs converges quickly to a consistent state. It has the technical advantages of solving uneven current distribution, effectively restoring bus voltage, and eliminating steady-state voltage deviation.

[0017] Compared with existing technologies, the advantages of this invention are as follows: By improving the design of the fixed-time consensus algorithm and adjusting its gain coefficient and convergence coefficient, while optimizing the distributed droop control method, and combining the two, an improved distributed droop control method based on the fixed-time consensus algorithm is constructed. This method obtains the consistency results of the output current of each DG and uses them to dynamically adjust the droop coefficient. While achieving rapid and stable convergence of the output current of each DG, it effectively solves the problem of uneven current distribution caused by line impedance mismatch and eliminates bus voltage deviation. In addition, this invention combines the advantages of the consensus algorithm (low communication resource consumption) and the stable and reliable droop control method. Compared with traditional droop control methods, the control method proposed in this invention has significant improvements in current sharing control accuracy, convergence response speed, bus voltage recovery speed, and bus voltage steady-state deviation. Attached Figure Description

[0018] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the structure of the multi-DG used in this invention; Figure 3 This is a block diagram of the improved droop control adopted in this invention; Figure 4 These are the output voltage-output current dynamic characteristic curves under traditional droop and optimized droop control; Figure 5 This is a comparison of the convergence performance of various traditional consensus algorithms and the improved fixed-time consensus algorithm proposed in this invention under conditions where the initial states differ significantly. Figure 6 This is a comparison of the convergence performance of various traditional consensus algorithms and the improved fixed-time consensus algorithm proposed in this invention when the initial state differences are small. Figure 7 This is a waveform diagram of the bus voltage under load switching conditions according to the present invention; Figure 8 This is a waveform diagram of the output current of each DG under the load switching condition of the present invention; Figure 9 This is a waveform diagram of the bus voltage compensation under load switching conditions according to the present invention. Figure 10 This is a waveform diagram of the droop coefficient correction under load jump conditions of the present invention. Detailed Implementation

[0019] To illustrate the technical content, structural features, objectives, and effects of the present invention in detail, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0020] Example 1

[0021] This embodiment presents an improved fixed-time consistent distributed droop control method for DC microgrid systems. The DC microgrid system structure using multiple parallel DGs is shown in the appendix to the specification. Figure 1-2 As shown, it includes the following steps: S1. In a distributed DC microgrid system, a distributed communication and control architecture without a central node is constructed. The communication between each distributed generator (DG) is abstracted into the topological relationship of a connected graph. Each DG only interacts with its adjacent nodes, which improves the single-point dependence and high communication pressure defects in the traditional centralized control method. S2. Based on the communication architecture constructed in step S1, and using the consensus algorithm, an improved fixed-time consensus algorithm is constructed by designing the algorithm framework and adjusting the gain coefficient, convergence coefficient, etc., which has an upper bound on the convergence time that is independent of the initial running state. T This overcomes the problems of slow convergence and uncertain convergence time in traditional consensus algorithms; S3. Optimize the droop control method by introducing the fixed-time convergence characteristic of the improved fixed-time consensus algorithm obtained in step S2 into the DC microgrid droop control. Adjust the droop coefficient using the consistency results of the output current of each DG. K i ' and bus compensation voltage Δ u i Simultaneous real-time dynamic adjustment is achieved, so that the output current of all DGs converges to a consistent state within the fixed convergence time upper limit, and its convergence time is independent of the initial state of the system. The bus voltage is restored through the voltage control loop, which solves the problems of bus voltage deviation caused by line impedance mismatch and uneven current distribution among DGs, and breaks through the limitations of voltage regulation and uneven current distribution in traditional droop control. S4. Based on Matlab / Simulink, build a simulation model of multiple DGs in parallel in a DC microgrid, design the basic parameters of the DC microgrid with multiple DGs, and set typical operating conditions. S5. In the model established in step S4, the microgrid is controlled using the improved fixed-time consistency distributed droop control method obtained in steps S1-S3, and experimental simulation is performed; the droop coefficient correction Δ is observed. K i Observation of reference voltage compensation amount Δ u i Simulation results, and using the DC microgrid bus voltage, etc. U bus Output current of each DG I oi The proposed method demonstrates significant improvements in voltage regulation and current sharing performance through various metrics.

[0022] Preferably, step S1 specifically includes: S101. Build a distributed communication and control framework without a central node, associate the information layer with the physical device layer, and each DG only exchanges voltage / current information with the two physical closest nodes (i.e., adopts adjacent communication) to ensure full connectivity of the communication network. S102. Abstract the communication between DGs into the topological relationships of a connected graph, using an undirected weighted graph. G ( A )=( V , E , A ) indicates that among them V Represents a non-empty set of nodes. E This represents the set of edges between nodes, containing all pairs of neighboring nodes. A= [ a ij Let be the adjacency matrix of the undirected weighted graph, containing the communication weight information between each node, satisfying: (1) In the formula, ( j , i ) E Represents a node j It is a node i One of the neighboring nodes, a ij Representing neighboring nodes i and nodes j Communication weights between them; Global coordination and optimization are achieved through information interaction between adjacent distributed units. The local communication interaction between each DG adopts adjacent communication, which does not require a central controller and can achieve convergence effect with a small communication burden.

[0023] Preferably, step S2 specifically includes: S201. Based on graph theory in step S102, the formula for the consensus algorithm is as follows: (2) In the formula, x i ( t ) is a node i exist t The state at any given moment; N i For nodes i The set of adjacent nodes. As long as the network's communication topology satisfies the connectivity graph, then equation (2) holds true. t As we approach infinity, the following condition is met: (3) That is, after running for a sufficiently long time, it can ensure that the state difference between each node is 0, thus achieving the consistency effect of different nodes; S202. Combining the fixed-time consistency theory, by adjusting the gain coefficient, changing the convergence coefficient, and other parameter settings, an improved fixed-time consistency algorithm is constructed as follows: (4) In the formula, the gain coefficient k >0, convergence coefficient and satisfy >1, 0< <1, sig( z ) = | z | sign( z ), sign(·) is the sign function; The improved fixed-time consensus algorithm will automatically determine state errors. x j ( t )- x i ( t The relationship between the value of the convergence coefficient and the error threshold 1 is used to dynamically select the convergence coefficient, which greatly accelerates the convergence speed while ensuring a fixed convergence time. Overall system convergence time T satisfy: (5) The convergence time depends only on the intrinsic coefficients and is independent of the initial input conditions. This improved fixed-time consistency algorithm achieves convergence in less than [time value missing]. T .

[0024] Preferably, step S3 specifically includes: S301, as per the instruction manual Figure 3 The improved droop control block diagram shown is for N For a parallel DG, an equivalent model is constructed and its voltage and current droop characteristic expression is obtained: (6) In the formula, K Represents virtual impedance. r i and i oi These represent the line impedance and output current of each branch, respectively. u ref Indicates the reference voltage. u busIndicates the bus voltage; S302. Model the droop control method and construct the dynamic characteristic curve of "output voltage - output current", as shown in the appendix to the manual. Figure 4 The output voltage-output current dynamic characteristic curve under droop control is shown in (a); S303, Due to line impedance mismatch (i.e.) r i ≠ r j This leads to uneven current distribution, so it's necessary to increase the droop coefficient by an appropriate amount for correction. K i To dynamically adjust the droop coefficient and achieve power distribution, different DGs at this time i They have the same total droop coefficient; The fixed-time convergence characteristic of the improved fixed-time consensus algorithm obtained in step S2 is introduced into the droop control of the DC microgrid. Based on equation (4) and combined with droop control, the consistency calculation of each DG current parameter is performed to obtain the consistency iteration result of the output current. i Xi : (7) In the formula, i Li Indicates each DG i The output current, i Xi This represents the current consistency iteration result. According to consistency theory, after... t < T Time, and the iteration results satisfy: (8) The droop coefficient correction Δ is obtained by updating it using an improved fixed-time consistency algorithm. K i Real-time adjustment of DG's droop coefficient K i ': (9) To satisfy this, for different DG i They all have the same total droop coefficient. K i That is, dynamically adjusting DG using the improved fixed-time consensus algorithm obtained through S2. i The droop coefficient achieves the current equalization effect, so that the output current of all DGs converges to a consistent state within the fixed convergence time upper bound, and its convergence time is independent of the initial state of the system. S304. Due to the bus voltage drop caused by sag and line impedance, increase the reference voltage compensation. u i The bus voltage is stabilized through a secondary voltage control circuit, and after compensation, the bus voltage reaches the reference voltage, thus achieving precise voltage regulation. Based on step S303, a reference voltage compensation amount Δ is introduced into the control system. u i ,satisfy: (10) The compensation amount Δ u i By increasing the original reference voltage, the droop coefficient can be adjusted. K i Correction amount Δ K i and line impedance r i The resulting voltage drop is accurately compensated, and the compensated control strategy can completely restore the actual bus voltage to the rated reference value: (11) That is, the improved fixed-time consistency algorithm obtained through S2 dynamically adjusts the bus reference voltage compensation amount, so that the compensated bus voltage is restored to the reference voltage value, reducing steady-state voltage deviation and improving bus voltage quality; S305. Construct the optimized dynamic characteristic curve of "output voltage-output current", as shown in the instruction manual. Figure 4 The output voltage-output current dynamic characteristic curve under droop control is shown in (b).

[0025] Preferably, step S4 specifically includes: S401. A simulation model of a DC microgrid system with multiple DGs in parallel was built based on Matlab / Simulink. The model includes components such as distributed equivalent power sources, line impedance, loads, distributed communication networks, and electronic power converters. An improved fixed-time consistency algorithm control module was also added. S402. Design the basic parameters of a multi-DG DC microgrid and set typical operating conditions, which are divided into: Working condition 1: Performance test of the improved fixed-time consensus algorithm, namely: simulating the convergence effect under various different initial conditions, and verifying the reliability and superiority of the improved fixed-time consensus algorithm in terms of convergence speed by comparing and analyzing it with traditional consensus, finite-time consensus and other algorithms. Operating Condition 2: A sudden change in load occurs in the DC microgrid system. The system load increases at 2 seconds and decreases at 4 seconds. That is, the load power is 16000W in 0-2 seconds, 32000W in 2-4 seconds, and returns to 16000W in 4-6 seconds.

[0026] Preferably, step S5 specifically includes: S501. Based on the DC microgrid system simulation model containing 5 DGs built in step S4, the control method described in steps S1-S3 is put into use. Under the control parameters and simulation operating conditions set in step S4, the actual control effect under different operating conditions of the DC microgrid system is observed. Implementation results are as follows Figures 5-6 As shown. Different initial states are set. x 1 and x 2. Run the system under algorithms such as traditional consistency and finite-time consistency, with the initial state as follows: x The results of running step 1 are shown in the attached manual. Figure 5 As shown, in the initial state x 1. When the initial differences are small, taking the distribution between (-8, 8) as an example, the convergence time of the traditional consensus algorithm is about 4.5 seconds, while the convergence time of the finite-time consensus algorithm and the improved fixed-time consensus algorithm is about 2.5 seconds, showing a significant improvement in convergence speed; initial state x The results of operation 2 are shown in the attached manual. Figure 6 As shown, when the initial state changes, in the initial state... x 2. When the initial differences are large, taking the distribution between (-30, 100) as an example, the convergence speed of the finite-time consensus algorithm and the improved fixed-time consensus algorithm is still faster than the traditional consensus algorithm. However, without changing the system parameters, it can be seen that the convergence time of the finite-time consensus algorithm becomes about 4s with large fluctuations, while the convergence time of the improved fixed-time consensus algorithm is still within 2.5s. S502, Simulation observation of droop coefficient correction Δ K i Observation of reference voltage compensation amount Δ u i The simulation results are then compared with the theoretical analyses proposed in verification steps S1-S3 to verify their effectiveness and correctness; simultaneously, the bus voltage of the DC microgrid is observed. U bus With reference voltage U ref Deviation rate, output current of each DG I oi The convergence speed and convergence effect are used as indicators to verify the significant improvement of the proposed method in voltage stabilization performance and current equalization effect; Implementation results are as follows Figure 7-10 As shown. The load power is 16000W in 0-2s, 32000W in 2-4s, and returns to 16000W in 4-6s. When the load is increased in 2s, the bus voltage drops by about 3%, and adjusts back to the reference voltage value of 400V after about 0.4s. When the load is reduced in 4s, the bus voltage rises by about 3%, and adjusts back to the reference voltage value of 400V after about 0.4s.

[0027] Combination Figures 1-10 This invention improves the fixed-time consensus algorithm by adjusting its gain and convergence coefficients, and optimizes the distributed droop control method. Combining these two approaches, an improved distributed droop control method based on the fixed-time consensus algorithm is constructed. This method obtains the consistency results of the output current of each DG and uses them to dynamically adjust the droop coefficient. While achieving rapid and stable convergence of the output current of each DG, it effectively solves the problem of uneven current distribution caused by line impedance mismatch and eliminates bus voltage deviation. Furthermore, this invention integrates the advantages of the consensus algorithm (low communication resource consumption) and the stable and reliable droop control method. Compared with traditional droop control methods, the proposed control method significantly improves current sharing control accuracy, convergence response speed, bus voltage recovery speed, and bus voltage steady-state deviation.

[0028] Example 2

[0029] This embodiment relates to an improved fixed-time consistent distributed droop control device suitable for DC microgrid systems, including a memory and one or more processors. The memory stores executable code, and when the one or more processors execute the executable code, they implement the improved fixed-time consistent distributed droop control method for DC microgrid systems of Embodiment 1.

[0030] Example 3

[0031] This embodiment relates to a computer-readable storage medium storing a program that, when executed by a processor, implements the improved fixed-time consistent distributed droop control applicable to DC microgrid systems of Embodiment 1.

[0032] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. An improved fixed-time consistent distributed droop control method suitable for DC microgrid systems, characterized in that, Includes the following steps: S1. In a distributed DC microgrid system, a distributed communication and control architecture without a central node is constructed. The communication between each distributed generator (DG) is abstracted into the topological relationship of a connected graph. Each DG only interacts with its adjacent nodes, which improves the single-point dependence and high communication pressure defects in the traditional centralized control method. S2. Based on the communication architecture constructed in step S1, and using the consensus algorithm, an improved fixed-time consensus algorithm is constructed by designing the algorithm framework and adjusting the gain coefficient, convergence coefficient, etc., which has an upper bound on the convergence time that is independent of the initial running state. T This overcomes the problems of slow convergence and uncertain convergence time in traditional consensus algorithms; S3. Optimize the droop control method by introducing the fixed-time convergence characteristic of the improved fixed-time consensus algorithm obtained in step S2 into the DC microgrid droop control. Adjust the droop coefficient using the consistency results of the output current of each DG. K i ' and bus compensation voltage Δ u i Simultaneous real-time dynamic adjustment is achieved, so that the output current of all DGs converges to a consistent state within the fixed convergence time upper limit, and its convergence time is independent of the initial state of the system. The bus voltage is restored through the voltage control loop, eliminating the bus voltage deviation and uneven current distribution among DGs caused by line impedance mismatch. S4. Based on Matlab / Simulink, build a simulation model of multiple DGs in parallel in a DC microgrid, design the basic parameters of the DC microgrid with multiple DGs, and set typical operating conditions. S5. In the model established in step S4, the microgrid is controlled using the improved fixed-time consistency distributed droop control method obtained in steps S1-S3, and experimental simulation is performed; the droop coefficient correction Δ is observed. K i Observation of reference voltage compensation amount Δ u i Simulation results, and using the DC microgrid bus voltage, etc. U bus Output current of each DG I oi The indicators verify the voltage regulation performance and current sharing effect.

2. The improved fixed-time consistent distributed droop control method for DC microgrid systems as described in claim 1, characterized in that, In step S1, in the distributed DC microgrid system, a distributed communication and control architecture without a central node is constructed. The communication between each DG is abstracted as the topological relationship of a connected graph. Each DG only interacts with its adjacent nodes, improving the single-point dependency and high communication pressure defects of the traditional centralized control method. This includes the following steps: S101. Build a distributed communication and control framework without a central node, associate the information layer with the physical device layer, and each DG only exchanges voltage / current information with the two physical closest nodes (i.e., adopts adjacent communication) to ensure full connectivity of the communication network. S102. Abstract the communication between DGs into the topological relationships of a connected graph, using an undirected weighted graph. G ( A )=( V , E , A ) indicates that among them V Represents a non-empty set of nodes. E This represents the set of edges between nodes, containing all pairs of neighboring nodes. A= [ a ij Let be the adjacency matrix of the undirected weighted graph, containing the communication weight information between each node, satisfying: (1) In the formula, ( j , i ) E Represents a node j It is a node i One of the neighboring nodes, a ij Representing neighboring nodes i and nodes j Communication weights between them; Global coordination and optimization are achieved through information interaction between adjacent distributed units. The local communication interaction between each DG adopts adjacent communication, which does not require a central controller and can achieve convergence effect with a small communication burden.

3. The improved fixed-time consistent distributed droop control method for DC microgrid systems as described in claim 1, characterized in that, In step S2, based on the communication architecture constructed in step S1, and using the consensus algorithm, an improved fixed-time consensus algorithm is constructed by designing the algorithm framework and adjusting the gain coefficient, convergence coefficient, etc., which has an upper bound on the convergence time that is independent of the initial running state. T To overcome the problems of slow convergence and uncertain convergence time in traditional consensus algorithms, the following steps are included: S201. Based on graph theory in step S102, the formula for the consensus algorithm is as follows: (2) In the formula, x i ( t ) is a node i exist t The state at any given moment; N i For nodes i The set of adjacent nodes. As long as the network's communication topology satisfies the connectivity graph, then equation (2) holds true. t As we approach infinity, the following condition is met: (3) That is, after running for a sufficiently long time, it can ensure that the state difference between each node is 0, thus achieving the consistency effect of different nodes; S202. Combining the fixed-time consistency theory, by adjusting the gain coefficient, changing the convergence coefficient, and other parameter settings, an improved fixed-time consistency algorithm is constructed as follows: (4) In the formula, the gain coefficient k >0, convergence coefficient and satisfy >1, 0< <1, sig( z ) = | z | sign( z ), sign(·) is the sign function; The improved fixed-time consensus algorithm will automatically determine state errors. x j ( t )- x i ( t The relationship between the value of the convergence coefficient and the error threshold 1 is used to dynamically select the convergence coefficient, which greatly accelerates the convergence speed while ensuring a fixed convergence time. The overall convergence time of the system T satisfy: (5) The convergence time depends only on the intrinsic coefficients and is independent of the initial input conditions. This improved fixed-time consensus algorithm achieves convergence in less than [time value missing]. T .

4. The improved fixed-time consistent distributed droop control method for DC microgrid systems as described in claim 2, characterized in that, Step S3 optimizes the droop control method by incorporating the fixed-time convergence characteristic of the improved fixed-time consensus algorithm obtained in step S2 into the droop control of the DC microgrid. The droop coefficient is adjusted using the consistency results of the output current of each DG. K i ' and bus compensation voltage Δ u i To achieve simultaneous real-time dynamic adjustment and overcome the limitations of uneven voltage regulation and current distribution in traditional droop control, the following steps are included: S301, For N For a parallel DG, an equivalent model is constructed and its voltage and current droop characteristic expression is obtained: (6) In the formula, K Represents virtual impedance. r i and i oi These represent the line impedance and output current of each branch, respectively. u ref Indicates the reference voltage. u bus Indicates the bus voltage; S302. Model the droop control method and construct the dynamic characteristic curve of "output voltage-output current"; S303, Due to line impedance mismatch (i.e.) r i ≠ r j This leads to uneven current distribution, so it's necessary to increase the droop coefficient by an appropriate amount for correction. K i To dynamically adjust the droop coefficient and achieve power distribution, different DGs at this time i They have the same total droop coefficient; The fixed-time convergence characteristic of the improved fixed-time consensus algorithm obtained in step S2 is introduced into the droop control of the DC microgrid. Based on equation (4) and combined with droop control, the consistency calculation of each DG current parameter is performed to obtain the consistency iteration result of the output current. i Xi : (7) In the formula, i Li Indicates each DG i The output current, i Xi This represents the current consistency iteration result. According to consistency theory, after... t < T Time, and the iteration results satisfy: (8) The droop coefficient correction Δ is obtained by updating it using an improved fixed-time consistency algorithm. K i Real-time adjustment of DG's droop coefficient K i ': (9) To satisfy this, for different DG i They all have the same total droop coefficient. K i That is, dynamically adjusting DG using the improved fixed-time consensus algorithm obtained through S2. i The droop coefficient achieves the current equalization effect, so that the output current of all DGs converges to a consistent state within the fixed convergence time upper bound, and its convergence time is independent of the initial state of the system. S304. Due to the bus voltage drop caused by sag and line impedance, increase the reference voltage compensation. u i The bus voltage is stabilized through a secondary voltage control circuit, and after compensation, the bus voltage reaches the reference voltage, thus achieving precise voltage regulation. Based on step S303, a reference voltage compensation amount Δ is introduced into the control system. u i ,satisfy: (10) The compensation amount Δ u i By increasing the original reference voltage, the droop coefficient can be adjusted. K i Correction amount Δ K i and line impedance r i The resulting voltage drop is accurately compensated, and the compensated control strategy can completely restore the actual bus voltage to the rated reference value: (11) That is, the improved fixed-time consistency algorithm obtained through S2 dynamically adjusts the bus reference voltage compensation amount, so that the compensated bus voltage is restored to the reference voltage value, reducing steady-state voltage deviation and improving bus voltage quality; S305. Construct the optimized dynamic characteristic curve of "output voltage-output current".

5. The improved fixed-time consistent distributed droop control method for DC microgrid systems as described in claim 4, characterized in that, Step S4, which involves building a simulation model of a DC microgrid with multiple distributed generators (DGs) in parallel using Matlab / Simulink, designing the basic parameters of the DC microgrid with multiple DGs, and setting typical operating conditions, specifically includes: S401. A simulation model of a DC microgrid system with multiple DGs in parallel was built based on Matlab / Simulink. The model includes components such as distributed equivalent power sources, line impedance, loads, distributed communication networks, and electronic power converters. An improved fixed-time consistency algorithm control module was also added. S402. Design the basic parameters of a multi-DG DC microgrid and set typical operating conditions, which are divided into: Working condition 1: Performance test of the improved fixed-time consensus algorithm, namely: simulating the convergence effect under various different initial conditions, and verifying the reliability and superiority of the improved fixed-time consensus algorithm in terms of convergence speed by comparing and analyzing it with traditional consensus, finite-time consensus and other algorithms. Operating Condition 2: A sudden change in load occurs in the DC microgrid system. The system load increases at 2 seconds and decreases at 4 seconds. That is, the load power is 16000W in 0-2 seconds, 32000W in 2-4 seconds, and returns to 16000W in 4-6 seconds.

6. The improved fixed-time consistent distributed droop control method for DC microgrid systems as described in claim 5, characterized in that, In step S5, the microgrid is controlled using the improved fixed-time consistency-based distributed droop control method obtained from steps S1-S3, based on the model established in step S4. Experimental simulations are conducted to observe the droop coefficient correction Δ. K i Observation of reference voltage compensation amount Δ u i Simulation results, and using the DC microgrid bus voltage, etc. U bus Output current of each DG I oi The proposed method demonstrates significant improvements in voltage regulation and current sharing performance through various metrics, including: S501. Based on the DC microgrid system simulation model containing 5 DGs built in step S4, the control method described in steps S1-S3 is put into use. Under the control parameters and simulation operating conditions set in step S4, the actual control effect under different operating conditions of the DC microgrid system is observed. S502, Simulation observation of droop coefficient correction Δ K i Observation of reference voltage compensation amount Δ u i The simulation results are then compared with the theoretical analyses proposed in verification steps S1-S3 to verify their effectiveness and correctness; simultaneously, the bus voltage of the DC microgrid is observed. U bus With reference voltage U ref Deviation rate, output current of each DG I oi The convergence speed and convergence effect are used as indicators to verify the significant improvement of the proposed method in voltage stabilization performance and current equalization effect.

7. An improved fixed-time consistent distributed droop control device suitable for DC microgrid systems, characterized in that, The system includes a memory and one or more processors, wherein the memory stores executable code, and the one or more processors execute the executable code to implement the improved fixed-time consistent distributed droop control method for DC microgrid systems as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, It stores a program that, when executed by a processor, implements the improved fixed-time consistent distributed droop control method for DC microgrid systems as described in any one of claims 1-6.

Images