Multi-HESS distributed event trigger control system and method established on periodic communication

Abstract

The invention belongs to the technical field of new energy consumption, and discloses a multi-HESS distributed event trigger control system and method based on periodic communication, and the method comprises the steps: carrying out the droop control of a DC micro-grid, and achieving the power distribution of different HESSs through the adjustment of a virtual resistance value; a communication network between HESSs is constructed, and distributed control is adopted to carry out secondary regulation and control on DC bus voltage and power distribution; an event triggering mechanism is introduced into distributed control, and the convergence of the proposed method is verified based on a Lyapunov stability theory. According to the invention, a bottom layer uses a low-pass filter to carry out frequency division processing on reference current, so that high-frequency response of the super capacitor and low-frequency response of the storage battery are realized.

Description

Technical Field

[0001] This invention belongs to the field of new energy consumption technology, specifically relating to a multi-HESS distributed event triggering control system and method based on periodic communication. Background Technology

[0002] With the increasing penetration rate of new energy power generation and the integration of distributed energy storage, microgrids have received widespread attention and rapid development due to their ability to operate in both islanded and grid-connected modes and to absorb new energy output. DC microgrid systems have a simple control structure and do not suffer from reactive power flow and harmonic current issues found in AC microgrids. The control objective is to stabilize line voltage and control the power distribution among different DC sources. For power supply issues in remote areas, direct transmission through the main grid is costly. Building islanded DC microgrids can not only solve the electricity needs of local residents but also improve economic efficiency. However, because they are not directly controlled by the main grid, the stability of the bus voltage in islanded DC microgrids requires coordinated regulation by various energy storage systems. The randomness, volatility, and load uncertainty of new energy output in microgrids are significant, and a single type of energy storage cannot meet the regulation requirements of the microgrid. In recent years, hybrid energy storage systems composed of batteries and supercapacitors have been widely studied due to their stronger regulation capabilities compared to single-type energy storage.

[0003] To address the power control issue within a distributed hybrid energy storage system (HESS), a common approach is to use a low-pass filter to divide the HESS output power into high and low frequencies before distributing it to the batteries and supercapacitors. In practical applications, DC microgrids typically employ multiple HESSs to smooth out power fluctuations, using different distribution strategies based on the magnitude of the power fluctuations. For the control of multiple HESSs, current approaches primarily focus on centralized, decentralized, and distributed control. Centralized control uses a central controller to collect status information from all HESSs, generate control inputs, and distribute the optimization results to the HESSes according to the control strategy. This enables coordinated and economical operation among HESSes. However, centralized control requires a bidirectional communication link between the central controller and the signal source, and is significantly affected by single-point failures, exhibiting poor robustness and scalability.

[0004] Communication between HESS systems typically employs time-triggered communication, meaning sampling and communication occur at fixed intervals. To ensure control system reliability, these intervals are usually set relatively short. However, frequent data acquisition by the control system can lead to excessive energy consumption and high communication demands. Event-triggered control, on the other hand, maintains control performance while reducing the number of controller triggers, effectively decreasing communication network congestion. The trigger condition design must ensure that the time interval between any two triggers is not zero and is greater than a positive number. However, the minimum trigger interval may be lower than the system's sampling period, and the Zeno phenomenon may not be eliminated under external interference, thus reducing the control system's reliability. To avoid the Zeno phenomenon, the controller can sample at fixed time intervals, communicating only when the trigger condition is met. None of the above methods address the sampling period; an excessively large sampling period can affect system stability, while an excessively small sampling period reduces system efficiency. Summary of the Invention

[0005] To address the issues of frequent communication and heavy computational burden in current microgrid distributed control and event-triggered control, this invention provides a multi-HESS distributed event-triggered control system and method based on periodic communication. The underlying layer utilizes a low-pass filter to perform frequency division processing on the reference current, thereby achieving high-frequency response of supercapacitors and low-frequency response of batteries.

[0006] To achieve the above objectives, the present invention provides the following solution:

[0007] A multi-HESS distributed event-triggered control system based on periodic communication, the system comprising: an islanded DC microgrid containing multiple HESSs;

[0008] The photovoltaic power generation, DC load, and HESS are connected to the microgrid via a DC / DC converter. When the DC / DC converter is in boost mode, the HESS discharges to the outside. When the DC / DC converter is in buck mode, the HESS charges. The AC load and AC microgrid are connected to the system via a DC / AC converter, taking into account the line resistance between the HESS and the DC bus.

[0009] This invention also provides a multi-HESS distributed event triggering control method based on periodic communication, the method being implemented through the aforementioned system, the method comprising:

[0010] Droop control is implemented in the DC microgrid, and power distribution among different HESSs is achieved by adjusting the value of the virtual resistance.

[0011] A communication network between HESSs is constructed, and distributed control is used to perform secondary regulation of DC bus voltage and power distribution;

[0012] An event-triggered mechanism is introduced into the distributed control, and the convergence of the proposed method is verified based on Lyapunov stability theory.

[0013] Preferred, HESS i Output power meets:

[0014] P HESS,i =n ch,i P ch,i +n dis,i P dis,i ;

[0015] In the formula P ch,i P dis,i HESS i The charging and discharging power, n ch,i n dis,i n is a 0-1 variable. ch,i n dis,i 1 represents HESS i It is in a charging state or a discharging state, and satisfies n ch,i +n dis,i ≤1.

[0016] Preferably, the power allocation between different HESS includes: power allocation between HESS and power allocation within HESS;

[0017] The control methods for power distribution among HESS include:

[0018] Use the battery's SOC value to adjust the virtual resistance R i The value, expressed as:

[0019]

[0020] In the formula, and These represent the upper and lower limits of the battery's State of Charge (SOC), respectively; operation symbols These represent rounding down and rounding up, respectively; S b,i (t) represents HESS at time t. i SOC of internal storage battery, R max I is the upper limit of the virtual resistance. i For HESS i ; output current;

[0021] When the SOC value of the battery does not meet the requirements When constraining R iThe expression for adjustment is:

[0022]

[0023] The control methods for internal power distribution in HESS include:

[0024] Battery SOC meets No adjustment is made at the beginning. When the battery's SOC exceeds the upper or lower limit, compensation is made to the supercapacitor. The expression is:

[0025]

[0026] In the formula, δI i For current compensation, c i This is the control gain coefficient for the current compensation term, and its value is positively correlated with the supercapacitor's capacitance. These represent the upper and lower limits of the SOC of the supercapacitor, respectively; S sc,i (t) represents the SOC value of the supercapacitor at time t.

[0027] Preferably, by constructing a communication network between HESS and using distributed control for secondary regulation of DC bus voltage and power distribution, the following methods are employed: adding an average voltage regulation compensation term to the droop control layer for compensation regulation and introducing proportional current regulation for compensation regulation of droop control.

[0028] The methods for adding an average voltage regulation compensation term to the droop control layer for compensation include:

[0029] The average voltage observer is represented as:

[0030]

[0031] In the formula, t represents the current time. and HESS i With HESS j Average voltage observation, N i For HESS i The set of adjacent units; V i For HESS i Actual output voltage sample value; a ij Let A be an element of the adjacency matrix A, representing HESS. i With HESS j The communication status, when HESS i With HESS j When communication exists, a ij =1, or 0 if there is no communication connection;

[0032] The difference between the rated voltage and the average voltage observer output value multiplied by the control gain k V The average voltage regulation compensation term u is obtained by adjusting the input integrator. V,i , where k V >0;

[0033] Methods for compensating for droop control by introducing proportional current regulation include:

[0034] Define the proportional current difference as:

[0035]

[0036] In the formula, I j For HESS j The output current, R j For HESS j Virtual resistance, b is the reference value for the proportional current regulation circuit. i For the reference signal reception coefficient, if HESS i Can receive The value of b is then i =1, otherwise 0;

[0037] κ I,i Multiply by control gain k I The voltage compensation term u is obtained by adjusting the input integrator. I,i .

[0038] Preferably, after introducing the event triggering mechanism, the distributed event triggering control uses the estimated state information from the previous triggering moment for decision-making, i.e., the sampled signal from the latest triggering moment; the distributed event triggering control forms for average voltage regulation and proportional current regulation are as follows:

[0039]

[0040] In the formula, δu V,i Distributed event-triggered control form of average voltage regulation, δu I,i Distributed event-triggered control method for proportional current regulation. For HESS j The average voltage sample value at the previous trigger moment, V ref x is the rated voltage of the DC bus. i The difference between the rated voltage and the average voltage at the latest trigger moment. For HESS i The average voltage sample value at the previous trigger moment, For HESS i The current sample value at the previous trigger moment, For HESS jThe current sample value at the previous trigger moment, This is the current reference value for the proportional current regulation circuit;

[0041] Define variable z V,i z I,i The value is:

[0042]

[0043] Total voltage compensation term u i It becomes:

[0044] u i =∫(δu V,i +δu I,i )dt=∫(-k V x i -k I z V,i )dt;

[0045] HESS i Average terminal voltage estimate and current estimate At trigger time t k The value is determined by the actual value. I i (t k Update the state and pass the updated state information to the adjacent HESS; between two trigger times, i.e., at t∈[t k ,t k+1 The estimated value remains unchanged; HESS does not communicate with each other, but only collects its own information; HESS i HESS, the neighbor j Information reception is passive. and The value is determined by HESS j The previous trigger time determines; z V,i z I,i x i It is segmented and continuous; HESS only performs state updates and communication propagation at the trigger time.

[0046] Preferably, the trigger function is:

[0047]

[0048] In the formula, ε I,i ε represents the measurement error of the average voltage. V,i The measurement error of the average current, f I,i (t), f V,i (t) is a dummy variable, σ I,i σ is the current scaling factor, where 0 < σ I,i<1; σ V,i Let σ be the voltage scaling factor, and 0 < σ. V,i <1;

[0049] when Triggering and communication are performed at specific times, among which, To determine the derivative of the Lyapunov function of the control system, HESS i The trigger condition is set as follows:

[0050] f I,i (t)≥0∪f V,i (t)≥0;

[0051] Define non-negative trigger thresholds e1 and e2, when f is satisfied I,i (t)≥0∪f V,i When (t)≥0, the estimation errors of voltage and current also need to be compared with e1 and e2 to determine whether the event triggering is satisfied, thus improving HESS. i The triggering condition is:

[0052]

[0053] The designed distributed event-triggered control system uses sampling, detection, and calculation at fixed time intervals to determine whether the controller should be triggered. (HESS) i The next detection time t must meet the following requirements:

[0054] t = t k +lh;

[0055] In the formula, l is any positive integer, the minimum triggering time interval is limited to h, and no Zeno phenomenon occurs.

[0056] Preferably, the upper limit value of the trigger condition detection is derived based on the designed trigger function and known conditions.

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0058] This invention utilizes a low-pass filter to perform frequency division processing on the reference current, thereby achieving high-frequency response of the supercapacitor and low-frequency response of the battery. It also proposes a distributed coordinated control method for multi-HESS (Heated Shielded Storage System) over-limit adjustment of energy storage SOC, which not only achieves the control objectives of maintaining bus voltage stability and improving HESS power allocation accuracy, but also improves the reliability and economy of microgrid operation. To reduce the system's occupancy of communication and computing resources and eliminate the Zeno phenomenon in event-triggered control, a distributed event-triggered control method is proposed, and its convergence is verified based on Lyapunov stability theory. The trigger function designed under the proposed method involves fewer parameters, requires no distributed estimation operation, which is beneficial for system acquisition and design. By setting a trigger threshold, the system's occupancy of communication resources in steady state is reduced. To further reduce communication and computational loads, an upper limit for the sampling period is derived, increasing the system's sampling period and thus reducing the consumption of computing and communication resources. Attached Figure Description

[0059] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 This is a schematic diagram of a DC microgrid containing multiple HESS according to an embodiment of the present invention;

[0061] Figure 2 This is a block diagram of the droop control of the hybrid energy storage system according to an embodiment of the present invention;

[0062] Figure 3 HESS is an embodiment of the present invention. i Distributed control diagram;

[0063] Figure 4 This is a schematic diagram of the HESS event triggering control flow according to an embodiment of the present invention;

[0064] Figure 5 This is a schematic diagram comparing different control strategies in embodiments of the present invention, wherein (a) is a schematic diagram of droop control; and (b) is a schematic diagram of distributed coordinated control.

[0065] Figure 6 This is a schematic diagram of the internal power distribution and supercapacitor SOC of HESS according to an embodiment of the present invention, wherein (a) is a schematic diagram of HESS2; and (b) is a schematic diagram of HESS3.

[0066] Figure 7 This is a schematic diagram of the output current waveform when the battery's SOC exceeds the limit, according to an embodiment of the present invention.

[0067] Figure 8 This is a schematic diagram illustrating the event triggering control effect of an embodiment of the present invention;

[0068] Figure 9 The diagrams show the HESS output current under different triggering conditions in the embodiments of the present invention, wherein (a) is a schematic diagram with e1 = 0.01 and e2 = 0.001; (b) is a schematic diagram with e1 = 0.02 and e2 = 0.003; and (c) is a schematic diagram with e1 = 0.04 and e2 = 0.005.

[0069] Figure 10 The diagram below shows the output current of HESS under the same method in the embodiment of the present invention, wherein (a) is a schematic diagram of the method of the present invention; (b) is a schematic diagram of the consensus algorithm; (c) is a schematic diagram of the event-triggered control method with continuous sampling; and (d) is a schematic diagram of the event-triggered control method with periodic sampling. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0072] Example 1

[0073] This invention provides a multi-HESS distributed event-triggered control system based on periodic communication, taking an islanded DC microgrid containing multiple HESSs as the research object. Its basic structural block diagram is as follows. Figure 1 As shown, photovoltaic power generation, DC loads, and HESS are connected to the microgrid via a DC / DC converter. When the DC / DC converter is in boost mode, the HESS discharges; when the DC / DC converter is in buck mode, the HESS charges. AC loads and the AC microgrid are connected to the system via a DC / AC converter, taking into account the line resistance between the HESS and the DC bus.

[0074] In order to maximize the output of new energy sources, Figure 1 The photovoltaic power generation operates in MPPT mode and does not directly participate in bus voltage control. During islanded operation, it is disconnected from the control of the main power grid. The difference between load power demand and renewable energy output, as well as bus voltage stability, are shared by all hybrid energy storage systems. The power generation equipment independently forms a grid to supply power to the load. Figure 1 The HESS (Heated Storage System) in this context is a modular energy storage structure where batteries and supercapacitors are integrated as a whole into a DC microgrid. Each HESS responds quickly to its output power based on virtual resistance droop control, and dynamically adjusts the virtual resistance according to the charge / discharge state of each battery and its own state of charge (SOC). Internally, a low-pass filter is used to divide the reference current. To reduce the impact of line resistance and achieve distributed coordinated control of the HESS, a corresponding communication network and control system should also be provided in the microgrid system.

[0075] HESS (Host-Environmental Components) can exchange information directly and indirectly through a communication network, and generate control variables from the collected information, thereby achieving coordinated and economical operation among HESS. Introducing graph theory into the DC microgrid, the communication network can be abstracted as an undirected graph G = (V, E), where V is the set of nodes and E is the set of edges, indicating whether direct information exchange exists between HESS. If node i can exchange information with node j, then node i and node j are neighboring nodes.

[0076] Example 2

[0077] This invention also provides a multi-HESS distributed event triggering control method based on periodic communication, the method being implemented through the aforementioned system, the method comprising:

[0078] Droop control is implemented in the DC microgrid, and power distribution among different HESSs is achieved by adjusting the value of the virtual resistance.

[0079] A communication network between HESSs is constructed, and distributed control is used to perform secondary regulation of DC bus voltage and power distribution;

[0080] An event-triggered mechanism is introduced into distributed control, and the convergence of the proposed method is verified based on Lyapunov stability theory. The specific implementation process is as follows:

[0081] To address the issues of ensuring reasonable power output and internal power distribution within a HESS, and to achieve distributed control among HESSes, a multi-HESS distributed control method based on battery SOC is proposed, considering the SOC over-limit situation of the battery or supercapacitor.

[0082] In this embodiment, droop control: The droop control of any i-th hybrid energy storage system in the DC microgrid is as follows:

[0083]

[0084] In the formula, For HESS i The reference voltage output from the droop control terminal, Vref R is the rated voltage of the DC bus. i For virtual resistance, I i For HESS i The output current.

[0085] A DC microgrid control system can achieve power distribution among different HESSs by adjusting the value of the virtual resistance. Considering the line resistance R... line The following relationship exists between the output currents of each HESS when affected.

[0086]

[0087] In the formula, I j For HESS j The output current, R j For HESS j Virtual resistance, R line,i R line,i HESS i and HESS j The line resistance.

[0088] It can be seen from equation (2) that when R i >>R line,i At that time, HESS i The output current is inversely proportional to the virtual resistance, so when performing HESS power distribution, a larger virtual resistance can be set to reduce the impact of line resistance.

[0089] HESS i Output power meets:

[0090] P HESS,i =n ch,i P ch,i +n dis,i P dis,i (3)

[0091] In the formula, P ch,i P dis,i HESS i The charging and discharging power, n ch,i n dis,i n is a 0-1 variable. ch,i n dis,i 1 represents HESS i It is in a charging state or a discharging state, and satisfies n ch,i +n dis,i ≤1.

[0092] Among them, the power distribution control strategy between HESS is as follows:

[0093] Setting the virtual resistance too large will increase the steady-state voltage deviation, and in severe cases, it may even exceed the system's maximum allowable voltage deviation. To ensure that the DC bus voltage deviation meets the microgrid's operational requirements in steady state, R... i The set value must satisfy the constraint of equation (4).

[0094]

[0095] In the formula, R max ΔV represents the upper limit of the virtual resistance. max The maximum permissible voltage deviation of the bus voltage is taken as ±5% of the rated DC bus voltage in this invention. rate,i For HESS i The rated current output by the converter.

[0096] For a DC microgrid system with multiple HESS modules, the initial states of each module may differ. Adjusting the virtual resistance using a fixed value cannot achieve the distribution effect of equation (2). Moreover, if the virtual resistance operates at a fixed ratio for a long time, it will increase the SOC difference between each HESS, causing some HESS modules to prematurely shut down due to insufficient charging and discharging power. To improve the overall regulation capability of the HESS, the virtual resistance can be adjusted according to the SOC of each HESS. Since the cost of supercapacitors is relatively high, the capacity of the battery is usually much larger than that of the supercapacitor when configuring the HESS capacity. The state difference between HESS modules is mainly reflected in the battery SOC. Therefore, the battery SOC value is selected to adjust R. i The value is shown in equation (5):

[0097]

[0098] In the formula, and These represent the upper and lower limits of the battery's State of Charge (SOC), respectively. This invention uses... Operators These represent rounding down and rounding up, respectively; S b,i (t) represents HESS at time t. i State of charge (SOC) of the internal storage battery.

[0099] Considering that the SOC of a battery may exceed the upper and lower limits during initial use and during charging and discharging, when the battery's SOC value does not meet the requirements... When constrained, the present invention uses equation (6) to constrain R. i Adjustments will be made.

[0100]

[0101] Among them, the HESS internal power distribution control strategy is as follows:

[0102] To extend battery life and fully utilize the high-frequency response characteristics of supercapacitors, this invention employs a low-pass filter. Figure 2 The total current reference command in the circuit is frequency divided. The battery current inner loop calculates the difference between the current output current and the low-frequency current reference command, and the difference is input to the PI controller. Finally, the output power of the DC / DC converter is controlled by PWM. The high-frequency current reference value is the difference between the initial current reference value and the low-frequency current value. The supercapacitor current inner loop control method is similar to that of the battery section.

[0103] Because supercapacitors have relatively small capacitance, an excessively high or low State of Charge (SOC) at initial deployment can easily lead to exceeding upper and lower limits during long-term operation. This invention designs a current compensation strategy based on the supercapacitor's SOC. To fully utilize the supercapacitor's capacitance, its SOC must meet certain conditions. No adjustment is made at the time. When its SOC exceeds the upper and lower limits, the supercapacitor is compensated according to formula (7).

[0104]

[0105] In the formula, δI i For current compensation, c i This is the control gain coefficient for the current compensation term, and its value is positively correlated with the supercapacitor capacity. These represent the upper and lower limits of the SOC of the supercapacitor, respectively; S sc,i (t) represents the SOC value of the supercapacitor at time t.

[0106] The above-mentioned basic droop control structure of HESS is as follows: Figure 2 As shown.

[0107] In this embodiment, distributed control: As shown in equation (2), the unbalanced power distribution problem can be solved by adjusting the virtual resistance value, but droop control will cause the DC bus voltage to deviate from the rated value. At the same time, due to the influence of line resistance, the power distribution between HESS does not meet the preset requirements, and may even increase the state difference between energy storage systems. A distributed control scheme can be introduced on the basis of droop control to reduce the above-mentioned effects. This invention constructs a communication network between HESS, where information exchange between HESS depends on the existence of a direct information channel, and uses distributed control to perform secondary regulation of DC bus voltage and power distribution.

[0108] Among them, average voltage regulation:

[0109] As can be seen from equation (1), the terminal voltage output by the HESS deviates from its rated value under the influence of droop control, and is also affected by the uncertainty of line impedance. To reduce this effect, an average voltage regulation compensation term is added to the droop control layer. The average voltage observer can be expressed as:

[0110]

[0111] In the formula, t represents the current time. and HESS i With HESS j Average voltage observation, N i For HESS i The set of adjacent units; V i For HESS i Actual output voltage sample value; a ij Let A be an element of the adjacency matrix A, representing HESS. i With HESS j The communication status, when HESS i With HESS j When communication exists, a ij =1, or 0 if there is no communication connection.

[0112] The difference between the rated voltage and the average voltage observer output value multiplied by the control gain k V The average voltage regulation compensation term u is obtained by adjusting the input integrator. V,i Among them, k V >0.

[0113]

[0114] Among them, proportional current regulation:

[0115] As can be seen from equation (2), in order to realize the distributed coordinated control of the energy storage system, the current is distributed to the HESS according to equation (9).

[0116]

[0117] In the formula, n is the total number of HESS.

[0118] Due to the presence and difficulty in obtaining line resistance, droop control alone cannot achieve the distribution of output current to each HESS according to a given virtual resistance. Therefore, proportional current regulation can be introduced to compensate for and adjust the droop control. The proportional current difference is defined as:

[0119]

[0120] In the formula, b is the reference value for the proportional current regulation circuit. i For the reference signal reception coefficient, if HESS i Can receive The value of b is then i =1, otherwise 0. (The last part, "κ", appears to be a typo and can be omitted.) I,iMultiply by control gain k I The voltage compensation term u can be obtained by adjusting the input integrator. I,i .

[0121]

[0122] As can be seen from equations (8) and (10), distributed control only requires the HESS to communicate with adjacent HESSs through a sparse communication network, while non-adjacent HESSs achieve indirect communication, reducing the system's demand for communication resources. When distributed control fails, the HESS can still operate normally under droop control, indicating that distributed control has a plug-and-play function. Distributed control is based on droop control and does not affect the reliability of microgrid operation. At the same time, the sparse communication network reduces the communication load and improves the system's economy. The distributed control block diagram is as follows: Figure 3 As shown.

[0123] In this embodiment, distributed event triggering control:

[0124] While distributed control based on periodic sampling can effectively reduce the impact of droop control and line resistance, it requires the HESS controller to continuously communicate and make decisions according to the system sampling frequency, and each sampling requires updating its own state. Considering the large number of HESS connections in practical engineering and the high cost of building communication networks, this invention introduces an event-triggered mechanism into the distributed control and verifies the convergence of the proposed strategy based on Lyapunov stability theory. Since the HESS only communicates with neighboring HESSes when the triggering condition is met, the control system's consumption of communication resources is reduced. Furthermore, to address the heavy computational burden caused by the small sampling interval in traditional event-triggered control based on continuous-time sampling, this invention calculates an upper limit for the sampling period, thereby increasing the sampling period and reducing the control system's demand for computational resources.

[0125] In the triggering time analysis: To facilitate subsequent derivation, the difference between the rated voltage value and the average voltage value at the latest triggering time is defined as x. i Unlike traditional distributed control, which uses its latest state information for control decisions, distributed event-triggered control, by introducing an event-triggered mechanism, uses the estimated state information from the previous trigger moment—that is, the sampled signal from the latest trigger moment—for decision-making. The distributed event-triggered control forms for average voltage regulation and proportional current regulation are as follows:

[0126]

[0127] In the formula, δu V,i Distributed event-triggered control form of average voltage regulation, δu I,iDistributed event-triggered control method for proportional current regulation. For HESS j The average voltage sample value at the previous trigger moment, V ref x is the rated voltage of the DC bus. i The difference between the rated voltage and the average voltage at the latest trigger moment. For HESS i The average voltage sample value at the previous trigger moment, For HESS i The current sample value at the previous trigger moment, For HESS j The current sample value at the previous trigger moment, This is the current reference value for the proportional current regulation circuit.

[0128] To facilitate subsequent derivation and trigger function design, variable z is defined. V,i z I,i The value is:

[0129]

[0130] At this time, the total voltage compensation term u i It becomes:

[0131] u i =∫(δu V,i +δu I,i )dt=∫(-k V x i -k I z V,i )dt (15)

[0132] HESS i Average terminal voltage estimate and current estimate At trigger time t k The value is determined by the actual value. I i (t k Update the state and pass the updated state information to the adjacent HESS; between two trigger times, i.e., at t∈[t k ,t k+1 The estimated values ​​remain unchanged, and HESS do not communicate with each other, only collecting their own information. i HESS, the neighbor j Information reception is passive. and The value is determined by HESS j The previous triggering time determines z. V,i z I,i xi It is segmented and continuous; HESS only performs state updates and communication propagation at the trigger time.

[0133] Among them, the design of trigger function and trigger condition: the trigger time is determined by the trigger condition, and a brief derivation process of the event trigger condition is given based on Lyapunov stability theory.

[0134] The measurement errors for average voltage and current are defined as follows:

[0135]

[0136] The tracking errors for average voltage and current are defined as follows:

[0137]

[0138] Design the Lyapunov function as follows:

[0139]

[0140] In the formula, E1 is the set of tracking errors for the average voltage, and E1 = [e I,1 ,e I,2 ,...,e I,n ] T E2 is the tracking error set of the proportional current, and we have E2 = [e V,1 ,e V,2 ,...,e V,n ] T H = L + B, where H is the positive definite matrix related to the control system, L is the Laplace matrix of the communication network, and B = [B1, B2, ..., B]. n ] T B is the set of receiving matrices. n Let B1 be the nth signal reception matrix, where B1 = [b1, 0, ..., 0]. T b1 is the received signal coefficient of HESS1.

[0141] Differentiating equation (15) and applying Young's inequality By scaling, we can obtain:

[0142]

[0143] In the formula: c is the derivative of the Lyapunov function of the control system; i As a dummy variable, c i =|N i |+|b i | / 2, β1 and β2 are Young's inequality coefficients; |N i |For HESS iNumber of adjacent HESS.

[0144] When ε I,i ε V,i When equation (18) is satisfied, At this point, the system satisfies Lyapunov stability.

[0145]

[0146] In the formula, f I,i (t), f V,i (t) is a dummy variable, σ I,i σ is the current scaling factor, where 0 < σ I,i <1; σ V,i Let σ be the voltage scaling factor, and 0 < σ. V,i <1.

[0147] Equation (18) shows that the trigger function designed for the distributed event triggering control loop does not require additional distributed estimation operations. The method of this invention does not require obtaining the number of HESS events running in the system, and the trigger function does not involve the control gain k. I The sampling period h and the number of HESS do not require additional adjustments to the trigger function when the system size is expanded or reduced, thus reducing the design difficulty of the controller.

[0148] when Triggering and communication will be performed in time, and HESS will be used. i The trigger condition is set as follows:

[0149] f I,i (t)≥0∪f V,i (t)≥0 (21)

[0150] As can be seen from equations (18) and (19), after the distributed control objective is achieved, the triggering function becomes sensitive, and communication between each HESS remains frequent. When the system enters steady state, the measured signal changes little, and the frequent triggering in steady state consumes a large amount of communication resources. To further reduce the number of triggers, non-negative triggering thresholds e1 and e2 are defined. When equation (19) is satisfied, the estimation errors of voltage and current need to be compared with e1 and e2 to determine whether the event triggering is satisfied. Improved HESS i The triggering condition is:

[0151]

[0152] The designed distributed event-triggered control system uses sampling, detection, and calculation at fixed time intervals to determine whether the controller should be triggered. (HESS) i The next detection time t must meet the following requirements:

[0153] t = tk +lh (23)

[0154] In the formula, l is any positive integer.

[0155] As can be seen from equation (21), the minimum trigger time interval is limited to the sampling period h, and no Zeno phenomenon occurs.

[0156] Among them, the analysis of the trigger condition check cycle is as follows: Although current event trigger control methods based on continuous time sampling can reduce the communication burden of the system and avoid Zeno's phenomenon, it is known from the event trigger control execution steps that the event trigger condition needs to be judged in each sampling cycle. Reasonably setting the system sampling cycle h is beneficial to reducing communication pressure. Therefore, this invention derives the upper limit value of the trigger condition detection based on the designed trigger function and known conditions, providing a basis for the selection of the sampling cycle, thereby saving computational resources.

[0157] when At that time, according to equation (14), we have:

[0158]

[0159] In the formula: Let be the derivative of the current measurement error at time t; Let be the derivative of the average voltage measurement error at time t. Let be the derivative of the current at time t. Let be the derivative of the average voltage at time t.

[0160] From t k Integrating equation (22) up to t, we have:

[0161]

[0162] When t = t k When, the following relation holds:

[0163]

[0164] In the formula: Z I For variable z I,i The set of variables, where X is the variable x. i A set of.

[0165] By combining equations (23) and (24), we can obtain:

[0166]

[0167] In the formula, λ m Let σ be the largest eigenvalue of matrix H. I,m σ is the maximum value of the current scaling factor. I,nLet σ be the current scaling factor for the nth term. I,m =max{σ I,1 ,σ I,2 ,...,σ I,n};σ V,m σ is the maximum value of the voltage scaling factor. V,n Let σ be the voltage scaling factor for the nth term. V,m =max{σ V,1 ,σ V,2 ,...,σ V,n}

[0168] When h satisfies equation (26),

[0169]

[0170] Both the triggering condition and the design of h satisfy Lyapunov stability, thus it can be concluded that the system is stable.

[0171] In summary, the multi-HESS distributed event triggering control flow designed in this invention is as follows: Figure 4 As shown.

[0172] In this embodiment, to verify the effectiveness of the above strategy, a system was built on the MATLAB / Simulink platform and... Figure 1 The simulation model of the islanded DC microgrid system corresponding to the structure consists of new energy power generation, DC loads, and four HESS (Hybrid Energy Storage System) units composed of batteries and supercapacitors connected via a DC bus. The main parameters of the control system are shown in Table 1.

[0173] Table 1 Parameters of DC Microgrid and its Controller

[0174]

[0175] For ease of statistical analysis, the four HESS values ​​are designated as 1-4 in subsequent simulations. λ can be obtained from matrix H. m =3, in order to facilitate obtaining σ I,m σ V,m , parameter σ I,i σ V,i All values ​​are set to 0.9. The renewable energy output on each bus remains unchanged. The communication topology between the four HESSs is a ring, and the matrix H is shown below.

[0176]

[0177] Among them, the distributed coordination control effect is as follows: the initial battery SOC of the four HESS units is set to S... b1 =S b4 =0.65,S b2 =Sb3 =0.45, the initial SOC of the supercapacitor is S sc1 =S sc4 =0.6; S sc2 =0.8962; S sc3 =0.8999; The initial load demand power is the same as the output of the distributed power source. According to the adjustment strategy of the virtual resistance in equation (5), R1 = R4 and R2 = R3 are valid at the beginning. The simulation process is as follows: at t = 0s, the microgrid is disconnected from the main grid. At t = 1s, 50% of the initial load is cut off. At t = 4s, 75% of the initial load is added. At t = 7s, 70% of the initial load is cut off. At t = 10s, it is the same as t = 4s.

[0178] Comparison of the effects of droop control and distributed coordinated control: Figure 5 The average bus voltage and output current waveforms of each HESS are given for droop control and distributed coordinated control. Figure 5 It can be seen that, according to equation (4), for R max Applying constraints ensures that the average bus voltage deviation under both control strategies will not exceed the maximum permissible deviation range. Figure 5 In (a), due to the lack of average voltage regulation in droop control, the average bus voltage deviates from the reference voltage when the system recovers to steady state. Under distributed coordinated control, the average bus voltage eventually recovers to its rated value, and the overall bus voltage control effect is better than that of droop control. This indicates that the voltage drop effect in droop control is reduced by average voltage regulation, stabilizing the average bus voltage output at 220V. Figure 5 The current output waveform diagram shows that under droop control, the HESS cannot eliminate the influence of line resistance and cannot output current according to the control strategy. After adopting distributed coordinated control, when the system enters steady state, the HESS output currents of the batteries with the same SOC tend to converge, indicating that the proposed method can eliminate the influence of line resistance on power distribution and also shorten the current distribution time.

[0179] Internal power distribution within the energy storage unit: Taking HESS2 and HESS3 as examples, the total reference current is distributed to the battery and supercapacitor according to the proposed strategy. Figure 6 The internal power distribution of HESS and the SOC variation curve of the supercapacitor are presented. From Figure 6 It can be seen that when the system output power demand changes abruptly, under the frequency division effect of the low-pass filter, the output power inside the HESS is initially supported by the supercapacitor. When the system gradually reaches a steady state, the output power is mainly borne by the battery, and the output power of the supercapacitor gradually decreases to 0 in the steady state. Figure 6(b) When the SOC of the supercapacitor in HESS3 exceeds the upper limit in regions A and B, the system performs current compensation on the supercapacitor to gradually restore it to 90%. As a result of the current compensation, the battery output power response capability is also enhanced. Figure 6 In (a), the SOC of the supercapacitor HESS2 did not exceed the upper limit, so no current compensation occurred.

[0180] Battery SOC Over-Limit Adjustment Test: To verify that the proposed algorithm can still maintain the output power of each HESS at a given ratio when the SOC of some energy storage units exceeds the upper and lower limits, the initial battery SOC of the four HESSs was set as follows: S b1 =0.95, S b2 =0.15, S b3 =0.85, S b4 =0.05. Figure 7 The process of charging / discharging various hybrid energy storage systems is demonstrated when there are significant differences in the state of charge (SOC) of the batteries and some batteries exceed the upper and lower limits.

[0181] Figure 7 During charging, the charging current is mainly borne by HESS2 and HESS4, which have lower battery SOCs, while HESS1 and HESS3, with higher battery SOCs, have lower charging currents. During discharging, HESS1 and HESS3, with higher battery SOCs, bear the main discharge current, while HESS2 and HESS4, with lower battery SOCs, have lower discharge currents. The distribution follows the principle of "the capable do more," avoiding the situation where HESS batteries exit operation due to excessively high or low battery SOCs. It is worth noting that because the battery SOCs in equations (5) and (6) are rounded up / down, and the simulation time in this test was short and the battery capacity was large, the virtual resistance change caused by exceeding the battery SOC limit was not shown.

[0182] Among them, the effects of distributed event triggering control are:

[0183] Comparison of effects under different triggering conditions: As shown in equation (26), h < 0.000083, and the simulation sampling period can be set to 8e-5s. To verify the effectiveness of the proposed distributed event triggering control strategy, simulation was performed with the load size change during operation as the operating condition. The proposed strategy was added to the distributed control to obtain the average bus voltage and the output current change curves of each HESS as shown in the figure. Figure 8 As shown.

[0184] from Figure 8It can be seen that after the event triggering mechanism is added, the information exchange between HESSs only takes place when equation (20) is satisfied. The average bus voltage and the output current of each HESS can still maintain stable system control, which verifies the feasibility of the proposed control strategy.

[0185] Figure 9 The output currents of the four HESS under different triggering conditions are given. Figure 9 The small image in the upper right corner shows the output current waveforms of each HESS during the 5-6s system steady-state condition. Figure 9 (a) and Figure 9 (b) It can be seen that when the trigger threshold is relatively small, the HESS current can still remain stable; as the trigger threshold increases, Figure 9 The HESS current in (c) fluctuated slightly. To verify that the proposed control strategy can effectively reduce the number of system communications, the number of HESS communications under different trigger thresholds was statistically analyzed, and the results are shown in Table 2.

[0186] Table 2 Comparison of Communication Times

[0187]

[0188]

[0189] Table 2 shows that the number of communication calls between HESS components in the proposed method decreases as the trigger threshold increases, but an excessively high trigger threshold can make the system unstable. Under the premise of ensuring system control accuracy, e1 = 0.01 and e2 = 0.001 were selected as trigger thresholds. Compared with the method before optimization, the number of communication calls decreased by 79.07%, and the calculation frequency decreased by 37.5%.

[0190] Comparison of the proposed method with traditional methods: In order to verify the ability of the proposed method to cope with abnormal operating conditions, the simulation steps were adjusted as follows: HESS3 was damaged and exited the system between 3s and 7s. Figure 10 A comparison chart of HESS output current under four different control methods after HESS3 exits the system.

[0191] from Figure 10 It can be seen that all three control methods can maintain the stable operation of the system during load changes. After HESS3 exits at t=3s, the proposed method and consensus algorithm...

[12] The system can still maintain stable operation, and at this time the total power of the system is still distributed by the remaining HESS according to the design ratio. Figure 10 (c) Figure 10(d) are the event-triggered control method with continuous sampling and the event-triggered control method with periodic sampling, respectively. It can be seen that the output current of the remaining HESS fluctuates when HESS3 exits, and the energy storage system needs a long time to recover stable operation. The proposed method is more conducive to the safe and stable operation of the hybrid energy storage system.

[0192] Table 3 Total number of HESS communications and sampling period under different methods

[0193]

[0194] Table 3 shows the total number of HESS communications and sampling period under different control methods when HESS3 exits. The method of this invention reduces the number of communications by 76.61% and the computation frequency by 37.5% compared to the consensus algorithm. The continuous sampling event-triggered control method reduces the number of communications by 50.73% and increases the computation frequency by 400% compared to the consensus algorithm. The periodic sampling event-triggered control method reduces the number of communications by 78.64% compared to the consensus algorithm, while maintaining the same computation frequency. Figure 10 As can be seen from Table 3, the method of the present invention can achieve stable operation of the system with less communication and computing resources, verifying the feasibility of the proposed method in reducing the communication and computing burden of the system.

[0195] In summary, to address the issues of frequent communication and heavy computational burden in current microgrid distributed control and event-triggered control, this invention proposes a multi-HESS distributed event-triggered control strategy based on periodic communication, and the following conclusions are drawn:

[0196] 1) A control strategy that considers the over-limit adjustment of energy storage SOC is adopted to realize the power allocation of HESS in different states, so as to avoid the phenomenon of overcharging and over-discharging of individual units. Distributed control reduces communication pressure by constructing a sparse communication network.

[0197] 2) To address the issue of frequent communication between multiple hybrid energy storage systems, a distributed event-triggered control strategy for multiple HESS systems is proposed. This strategy reduces communication between HESS systems and also solves problems such as excessive bus voltage deviation and unbalanced power distribution caused by droop control and line resistance.

[0198] 3) Based on Lyapunov stability theory and boundary conditions of triggering functions, the upper limit of the sampling period is derived, which improves the decision interval of the triggering conditions and can effectively reduce the computation frequency of the system. At the same time, compared with the existing event-triggered control strategies, the proposed method has a significant improvement in control effect and a significant reduction in the demand for communication and computing resources.

[0199] In recent years, the security of HESS communication has received increasing attention. How to maintain the stable operation of the system under the conditions of packet loss and network attacks is a question worthy of further research.

[0200] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A multi-HESS distributed event-triggered control system based on periodic communication, characterized in that, The system includes: an islanded DC microgrid containing multiple HESS; The photovoltaic power generation, DC load, and HESS are connected to the microgrid via a DC / DC converter. When the DC / DC converter is in boost mode, the HESS discharges to the outside. When the DC / DC converter is in buck mode, the HESS charges. The AC load and AC microgrid are connected to the system via a DC / AC converter, taking into account the line resistance between the HESS and the DC bus.

2. A multi-HESS distributed event triggering control method based on periodic communication, the method being implemented using the system described in claim 1, characterized in that... The method includes: Droop control is implemented in the DC microgrid, and power distribution among different HESSs is achieved by adjusting the value of the virtual resistance. A communication network between HESSs is constructed, and distributed control is used to perform secondary regulation of DC bus voltage and power distribution; An event-triggered mechanism is introduced into the distributed control, and the convergence of the proposed method is verified based on Lyapunov stability theory.

3. The method according to claim 2, characterized in that, HESS i Output power meets: P HESS,i =n ch,i P ch,i +n dis,i P dis,i ; In the formula P ch,i P dis,i HESS i The charging and discharging power, n ch,i n dis,i n is a 0-1 variable. ch,i n dis,i 1 represents HESS i It is in a charging state or a discharging state, and satisfies n ch,i +n dis,i ≤1.

4. The method according to claim 2, characterized in that, Power allocation among different HESSs includes: power allocation between HESSs and power allocation within HESSs; The control methods for power distribution among HESS include: Use the battery's SOC value to adjust the virtual resistance R i The value, expressed as: In the formula, and These represent the upper and lower limits of the battery's State of Charge (SOC), respectively; operation symbols These represent rounding down and rounding up, respectively; S b,i (t) represents HESS at time t. i SOC of internal storage battery, R max I is the upper limit of the virtual resistance. i For HESS i ; output current; When the SOC value of the battery does not meet the requirements When constraining R i The expression for adjustment is: The control methods for internal power distribution in HESS include: Battery SOC meets No adjustment is made at the beginning. When the battery's SOC exceeds the upper or lower limit, compensation is made to the supercapacitor. The expression is: In the formula, δI i For current compensation, c i This is the control gain coefficient for the current compensation term, and its value is positively correlated with the supercapacitor's capacitance. These represent the upper and lower limits of the SOC of the supercapacitor, respectively; S sc,i (t) represents the SOC value of the supercapacitor at time t.

5. The method according to claim 4, characterized in that, By constructing a communication network between HESS and using distributed control for secondary regulation of DC bus voltage and power distribution, including: adding an average voltage regulation compensation term to the droop control layer for compensation regulation and introducing proportional current regulation for compensation regulation of droop control; The methods for adding an average voltage regulation compensation term to the droop control layer for compensation include: The average voltage observer is represented as: In the formula, t represents the current time. and HESS i With HESS j Average voltage observation, N i For HESS i The set of adjacent units; V i For HESS i Actual output voltage sample value; a ij Let A be an element of the adjacency matrix A, representing HESS. i With HESS j The communication status, when HESS i With HESS j When communication exists, a ij =1, or 0 if there is no communication connection; The difference between the rated voltage and the average voltage observer output value multiplied by the control gain k V The average voltage regulation compensation term u is obtained by adjusting the input integrator. V,i , where k V >0; Methods for compensating for droop control by introducing proportional current regulation include: Define the proportional current difference as: In the formula, I j For HESS j The output current, R j For HESS j Virtual resistance, b is the reference value for the proportional current regulation circuit. i For the reference signal reception coefficient, if HESS i Can receive The value of b is then i =1, otherwise 0; κ I,i Multiply by control gain k I The voltage compensation term u is obtained by adjusting the input integrator. I,i .

6. The method according to claim 5, characterized in that, After introducing the event triggering mechanism, distributed event triggering control uses the estimated state information from the previous triggering moment for decision-making, i.e., the sampled signal from the latest triggering moment; the distributed event triggering control forms for average voltage regulation and proportional current regulation are as follows: In the formula, δu V,i Distributed event-triggered control form of average voltage regulation, δu I,i Distributed event-triggered control method for proportional current regulation. For HESS j The average voltage sample value at the previous trigger moment, V ref x is the rated voltage of the DC bus. i The difference between the rated voltage and the average voltage at the latest trigger moment. For HESS i The average voltage sample value at the previous trigger moment, For HESS i The current sample value at the previous trigger moment, For HESS j The current sample value at the previous trigger moment, This is the current reference value for the proportional current regulation circuit; Define variable z V,i z I,i The value is: Total voltage compensation term u i It becomes: at i =∫(δu V,i +δu I,i )dt=∫(-k V x i -k I With V,i )dt; HESS i Average terminal voltage estimate and current estimate At trigger time t k The value is determined by the actual value. I i (t k Update the state and pass the updated state information to the adjacent HESS; between two trigger times, i.e., at t∈[t k ,t k+1 The estimated value remains unchanged; HESS does not communicate with each other, but only collects its own information; HESS i HESS, the neighbor j Information reception is passive. and The value is determined by HESS j The previous trigger time determines; z V,i z I,i x i It is segmented and continuous; HESS only performs state updates and communication propagation at the trigger time.

7. The method according to claim 6, characterized in that, The trigger function is: In the formula, ε I,i ε represents the measurement error of the average voltage. V,i The measurement error of the average current, f I,i (t), f V,i (t) is a dummy variable, σ I,i σ is the current scaling factor, where 0 < σ I,i <1; σ V,i Let σ be the voltage scaling factor, and 0 < σ. V,i <1; when Triggering and communication are performed at specific times, among which, To determine the derivative of the Lyapunov function of the control system, HESS i The trigger condition is set as follows: f I,i (t)≥0∪f V,i (t)≥0; Define non-negative trigger thresholds e1 and e2, when f is satisfied I,i (t)≥0∪f V,i When (t)≥0, the estimation errors of voltage and current also need to be compared with e1 and e2 to determine whether the event triggering is satisfied, thus improving HESS. i The triggering condition is: The designed distributed event-triggered control system uses sampling, detection, and calculation at fixed time intervals to determine whether the controller should be triggered. (HESS) i The next detection time t must meet the following requirements: t=t k +lh; In the formula, l is any positive integer, the minimum trigger time interval is limited to the sampling period h, and no Zeno phenomenon occurs.

8. The method according to claim 7, characterized in that, Based on the designed trigger function and known conditions, the upper limit value of the trigger condition detection is derived.

Images