Uniform quantization, coding and decoding combined double-event trigger load frequency control method and system

Abstract

The invention relates to the field of power system load frequency control, in particular to a uniform quantization and coding and decoding combined double-event trigger load frequency control method and system, and the method comprises the steps: constructing a discrete time model, calculating a load frequency error, introducing a pseudo partial derivative, and updating an estimated value of the pseudo partial derivative; the method comprises the following steps: judging whether to replace a pseudo-partial derivative according to a first event triggering condition, encoding, quantifying and decoding the pseudo-partial derivative and a load frequency error to obtain an updated control instruction, judging whether to update the control instruction according to a second event triggering condition, and finally executing power regulation after encoding, quantifying and decoding the control instruction. And enabling the load frequency error of each region to be within a preset range and circulating. According to the method, through an event triggering mechanism and a data uniform quantification design, the communication frequency is reduced, the communication cost and the energy consumption are reduced, the load frequency control of the multi-region interconnected power system is effectively realized, and the problems of limited power network communication resources and information quantification are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system load frequency control, in particular to a double-event-triggered load frequency control method and system combining uniform quantization and coding and decoding. BACKGROUND

[0002] In the modern power system operation system, frequency stability is the core element to ensure the safe and reliable operation of the power system. Frequency instability of the power system will cause deterioration of power quality on the user side, deterioration of system equipment operation condition, and even trigger a power grid cascading failure in extreme cases, leading to a complete network collapse and large-area power outage accident, and further causing serious social and economic losses. As a key technical means of power system operation control, the core function of load frequency control (LFC) is to maintain the stability of the system rated frequency and ensure that the power exchange between regions strictly follows the economic dispatching plan. Under the architecture of multi-regional interconnected power system, the core control objective of LFC is to achieve dynamic balance of power generation and load demand in each control region, while strictly controlling the system frequency deviation within the allowable threshold range.

[0003] The existing LFC technology is difficult to effectively balance the needs of model-free adaptive control, information quantization suppression and communication resource optimization scheduling. In the new energy high-penetration power grid environment, it cannot guarantee the dynamic balance of multi-regional power and the stability of system frequency, and the control efficiency is difficult to meet the actual operation demand. The three major defects can be summarized as follows:

[0004] Firstly, due to the strong nonlinearity, multivariable coupling and time-varying characteristics of the power system, accurate modeling is extremely difficult, making it difficult for traditional control strategies based on models to adapt to complex operating scenarios;

[0005] Secondly, in a multi-regional interconnected system, the communication network bandwidth and energy resources are limited, and high-frequency data transmission increases communication costs and energy consumption, while low-frequency transmission is prone to accumulate quantization errors of key information such as ACE (Area Control Error);

[0006] Thirdly, the existing control method cannot simultaneously consider model-free adaptability, information quantization suppression and communication resource optimization scheduling, and cannot effectively guarantee the dynamic balance of power generation and load in each region and the stability of system frequency. In the new energy high-penetration power grid environment, the control efficiency is particularly insufficient.

[0007] Therefore, when designing the LFC strategy for a multi-regional interconnected power system, it is necessary to consider the information quantization effect, reduce the transmission frequency by optimizing communication scheduling, and develop a model-free adaptive control strategy to address the problems of limited communication resources and information quantization, which has become an important research direction in the field of power system automation. SUMMARY

[0008] Therefore, the technical problem to be solved by the present invention is to overcome the problem that the load frequency control (LFC) of multi-region interconnected systems is difficult to achieve dynamic power balance in the region due to the difficulty in accurate modeling of power systems, the quantification of information in communication networks and the limitation of resources.

[0009] To address the aforementioned technical problems, this invention provides a dual-event triggered load frequency control method and system combining uniform quantization and encoding / decoding. The method includes the following steps:

[0010] Step S1: Construct a discrete-time model of any region operating in a multi-region interconnected power system. Each region calculates the load frequency error based on the control command received at the current moment using the discrete-time model.

[0011] Step S2: Introduce pseudo-partial derivatives to represent the dynamic response gradient of the system output to the control input. Based on the load frequency error and the control command, update the pseudo-partial derivatives at the current moment to obtain the updated pseudo-partial derivative estimate.

[0012] Step S3: Construct the first event trigger condition. Based on the load frequency error, determine whether the power system state change satisfies the first event trigger condition.

[0013] If so, replace the original pseudo-partial derivative with the updated pseudo-partial derivative estimate;

[0014] Otherwise, keep the original pseudo-partial derivatives unchanged;

[0015] Step S4: Based on the pseudo-partial derivatives obtained in step S3 and the load frequency error, perform encoding, data uniform quantization and decoding operations in sequence to obtain the decoded data, and obtain the updated control command based on the decoded data;

[0016] Step S5: Construct the second event trigger condition and determine whether the change in the control command satisfies the second event trigger condition.

[0017] If so, update the current control command using the updated control command;

[0018] Otherwise, keep the current control commands unchanged;

[0019] Step S6: The control command obtained in step S5 is sequentially encoded, data uniformly quantized, and decoded to obtain the decoded control command. Power regulation is performed according to the decoded control command so that the load frequency error of each region in the multi-region interconnected power system is within the preset error range. Then, return to step S1.

[0020] In one embodiment of the present invention, step S1, the method for constructing a discrete-time model of the operation of any region in a multi-region interconnected power system, includes:

[0021]

[0022] Where, x j (k) represents the system state vector of the j-th region at the current discrete time k, x j (k+1) represents x j The system state vector at the next discrete time k+1 of (k); A j B j C j F j All are system matrices. F represents the perturbation vector; j (k) represents the actual regional control error of the j-th region at the current discrete time k, i.e., the load frequency error at the current discrete time k, u j (k) represents the control input of the j-th region at the current discrete time k.

[0023] In one embodiment of the present invention, step S2, the method for updating the pseudo-partial derivative at the current moment based on the load frequency error and the control command to obtain the updated pseudo-partial derivative estimate includes:

[0024]

[0025] in, It is a pseudo-partial derivative ψ j The estimated value of (k), yes The pseudo-partial derivative ψ at the previous moment j The estimated value of (k); ΔE j (k) represents the change in system output, ΔE j (k)=E j (k)-E j (k-1), E j (k) represents the system output of the j-th region at the current discrete time k, E j (k-1) represents E j (k) The system output at the previous time step; Δu j (k-1) represents the change in control input, Δu j (k-1)=u j (k-1)-u j (k-2), u j (k-1) represents the control input of the j-th region at the previous discrete time k, u j (k-2) represents u jThe control input at the previous discrete time step (k-1); γ represents the gain parameter. j >0 indicates a smoothing parameter.

[0026] In one embodiment of the present invention, step S3, the method for constructing the first event triggering condition includes:

[0027]

[0028] in, This indicates the time of the qth output trigger in the j-th region. This indicates the (q-1)th output trigger time of the j-th region; This represents the change in load frequency error of the j-th region at time k. E j (k) represents the load frequency error of the j-th region at discrete time k. This indicates the output trigger time for the j-th region. The load frequency error; and For the set trigger parameters, The dynamic event triggering function for the j-th region is designed as follows:

[0029]

[0030] Where, δ j ∈(0,1) is the set trigger parameter. To set parameters.

[0031] In one embodiment of the present invention, in step S4, based on the pseudo-partial derivative obtained in step S3 and the load frequency error, encoding, data uniform quantization, and decoding operations are performed sequentially to obtain the decoded data. The method includes:

[0032] The encoder is constructed using the following function expression:

[0033]

[0034] Where E(X(k)) represents the encoded data at the current discrete time k, X(k) represents the data before encoding at the current discrete time k, ζ(k)∈(0,1) represents the scaling sequence for adjusting the encoder output amplitude; ξ(k) represents the encoder's internal state value at the current time k, ξ(k-1) represents the encoder's internal state value at the previous time step of ξ(k), and ξ(1)=0;

[0035] The encoder is used to encode the pseudo-partial derivatives obtained in step S3 and the load frequency error to obtain encoded data. A uniform quantizer is then constructed, and the encoded data is subjected to uniform quantization using the uniform quantizer to obtain quantized data. The function expression of the uniform quantizer is as follows:

[0036]

[0037] Where Q(E(X(k))) represents the quantized data, h∈Z + Indicates the quantization level parameter;

[0038] Construct a decoder, and use the decoder to process the quantized data to obtain the decoded data X. * (k):

[0039]

[0040] Among them, X * (k-1) represents the data decoded at discrete time k-1, ω(k) represents the quantization error term, and ω(k)=Q(E(X(k)))-E(X(k))≤0.06.

[0041] In one embodiment of the present invention, the encoder's internal state value ξ(k) at the current discrete time k is updated using the following formula:

[0042] ξ(k)=ζ(k)Q(E(X(k)))+ξ(k-1).

[0043] In one embodiment of the present invention, the function expression of the decoder is:

[0044] D(X(k))=D(X(k-1))+ζ(k)Q(E(X(k))).

[0045] In one embodiment of the present invention, step S4, the method for obtaining the updated control command based on the decoded data, includes:

[0046] Based on the decoded data, the updated input control command u is calculated using the following formula. j (k):

[0047]

[0048] in, This indicates the (p-1)th input trigger time of the j-th region. Input control commands, v j ∈(0,1] represents the control gain coefficient of the j-th region, α j >0 is the regularization parameter. This is the decoded feedback term at the current discrete time k. This is a pseudo-partial derivative estimate.

[0049] In one embodiment of the present invention, step S5, the method for constructing the second event triggering condition includes:

[0050]

[0051] in, This represents the trigger time of the p-th input in the j-th region. This indicates the (p-1)th input trigger time for the j-th region; This represents the predicted change in the j-th region at time k, as predicted by the control command. This represents the relative threshold coefficient for the j-th region. This represents the absolute threshold constant for the j-th region; This indicates that the j-th region was at the last trigger time. The control command value.

[0052] Based on the same inventive concept, this invention also provides a dual-event triggered load frequency control system combining uniform quantization and encoding / decoding, comprising: a discrete-time model construction module, a dynamic response gradient estimation module, an output event triggering decision module, a data processing and control command generation module, an input event triggering decision module, and a control execution and closed-loop feedback module; wherein, the discrete-time model construction module is configured to: construct a discrete-time model of operation in any region of a multi-region interconnected power system, and each region calculates the load frequency error through the discrete-time model based on the control command received at the current moment;

[0053] The dynamic response gradient estimation module is configured to: introduce pseudo-partial derivatives to represent the dynamic response gradient of the system output to the control input, update the pseudo-partial derivatives at the current time based on the load frequency error and the control command, and obtain the updated pseudo-partial derivative estimate.

[0054] The output event triggering decision module is configured to: construct a first event triggering condition, and based on the load frequency error, determine whether the power system state change meets the first event triggering condition; if so, replace the original pseudo-partial derivative with the updated pseudo-partial derivative estimate; otherwise, keep the original pseudo-partial derivative unchanged.

[0055] The data processing and control instruction generation module is configured to: based on the pseudo-partial derivative obtained by the output event triggering decision module and the load frequency error, perform encoding, data uniform quantization and decoding operations in sequence to obtain decoded data, and obtain updated control instructions based on the decoded data;

[0056] The input event triggering decision module is configured to: construct a second event triggering condition, and determine whether the change in the control command meets the second event triggering condition; if so, update the current control command through the updated control command; otherwise, keep the current control command unchanged.

[0057] The control execution and closed-loop feedback module is configured to: sequentially encode, uniformly quantize, and decode the control commands obtained by the input event triggering decision module to obtain decoded control commands; and perform power regulation according to the decoded control commands to ensure that the load frequency error of each region in the multi-region interconnected power system is within a preset error range.

[0058] The technical solution of the present invention has the following advantages compared with the prior art:

[0059] This invention avoids the modeling difficulties caused by the strong nonlinearity and multivariable coupling of power systems by introducing pseudo-partial derivatives to replace the exact mathematical model. Based on the measurable load frequency error and control commands, the estimated value of the pseudo-partial derivatives is dynamically updated, thus achieving model-free adaptive control. Addressing the information quantization problem in communication networks, a data uniform quantization rule and encoding / decoding processing chain are designed. An error compensation mechanism suppresses quantization distortion, ensuring that the decoded data accurately reflects the system state. Facing the challenge of limited communication resources, a dual-event triggering mechanism is constructed. The output trigger determines the state update requirement based on the load frequency error change and a dynamic triggering function, while the input trigger sets a composite trigger threshold based on the control command change, transmitting data only when the state change is significant, reducing redundant communication. Attached Figure Description

[0060] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein...

[0061] Figure 1 This is a schematic flowchart of a dual-event triggered load frequency control method combining uniform quantization and encoding / decoding provided in one embodiment of the present invention.

[0062] Figure 2 This is the dynamic model structure diagram of the power system in the j-th region;

[0063] Figure 3 This is a schematic diagram illustrating the specific workflow of the load frequency control method for the power system in the j-th region.

[0064] Figure 4 This is a schematic diagram of a three-region interconnected microgrid system;

[0065] Figure 5It is the triggering interval of the input and output channels under the condition of applying a step change in load, where (a) represents the triggering interval of the input channel and (b) represents the triggering interval of the input channel.

[0066] Figure 6 It is a curve showing the system load frequency deviation under a step change load condition;

[0067] Figure 7 It is the triggering interval of the input and output channels under the condition of applying dynamically changing load, where (a) represents the triggering interval of the input channel and (b) represents the triggering interval of the input channel;

[0068] Figure 8 It is a curve showing the system load frequency deviation under dynamically changing load conditions;

[0069] Figure 9 This is a schematic diagram of a dual-event triggered load frequency control system combining uniform quantization and encoding / decoding, provided in one embodiment of the present invention.

[0070] Explanation of reference numerals in the accompanying drawings: 100, Discrete-time model construction module; 200, Dynamic response gradient estimation module; 300, Output event triggering decision module; 400, Data processing and control command generation module; 500, Input event triggering decision module; 600, Control execution and closed-loop feedback module. Detailed Implementation

[0071] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0072] Example 1:

[0073] like Figure 1 As shown, this invention provides a dual-event triggered load frequency control method and system combining uniform quantization and encoding / decoding. The method includes the following steps:

[0074] Step S1: Construct a discrete-time model of any region operating in a multi-region interconnected power system. Each region calculates the load frequency error based on the control command received at the current moment using the discrete-time model.

[0075] Step S2: Introduce pseudo-partial derivatives to represent the dynamic response gradient of the system output to the control input. Based on the load frequency error and the control command, update the pseudo-partial derivatives at the current moment to obtain the updated pseudo-partial derivative estimate.

[0076] Step S3: Construct the first event trigger condition. Based on the load frequency error, determine whether the power system state change satisfies the first event trigger condition.

[0077] If so, replace the original pseudo-partial derivative with the updated pseudo-partial derivative estimate;

[0078] Otherwise, keep the original pseudo-partial derivatives unchanged;

[0079] Step S4: Based on the pseudo-partial derivatives obtained in step S3 and the load frequency error, perform encoding, data uniform quantization and decoding operations in sequence to obtain the decoded data, and obtain the updated control command based on the decoded data;

[0080] Step S5: Construct the second event trigger condition and determine whether the change in the control command satisfies the second event trigger condition.

[0081] If so, update the current control command using the updated control command;

[0082] Otherwise, keep the current control commands unchanged;

[0083] Step S6: The control command obtained in step S5 is sequentially encoded, data uniformly quantized, and decoded to obtain the decoded control command. Power regulation is performed according to the decoded control command so that the load frequency error of each region in the multi-region interconnected power system is within the preset error range. Then, return to step S1.

[0084] Furthermore, construct power systems for N regions. The dynamic model of the power system in the j-th region is as follows: Figure 2 As shown, based on Euler's approximation, the dynamic model is transformed into a discrete-time model, as follows:

[0085]

[0086] Where, x j (k)=[Δf j (k) ΔP pj (k) ΔP tie-j (k) ΔP gj (k)] T Let Δf represent the system state vector of the j-th region at the current discrete time k. j (k) represents the load frequency deviation, ΔP pj (k) represents the generator output power deviation, ΔP tie-i (k) represents the tie-line power deviation, ΔP gj (k) represents the governor valve position deviation; x j (k+1) represents x jThe system state vector at the next discrete time k+1 of (k); Let ΔP represent the disturbance vector. dj (k) represents the load change, S jv E represents the synchronous torque coefficient between region j and region v. j (k) represents the actual regional control error of the j-th region at the current discrete time k, i.e., the load frequency error at the current discrete time k, u j (k)=ΔP cj (k) represents the control input of the j-th region at the current discrete time k;

[0087] A j B j C j F j All are system matrices. T represents the sampling period. Defined as:

[0088]

[0089] H j Indicates generator inertia, Γ j T represents the frequency deviation coefficient. gj R represents the control time constant. j Indicating drooping characteristics, T tj D represents the turbine time constant. j This represents the damping coefficient of the generator set.

[0090] like Figure 3 As shown, each region calculates the load frequency error using the discrete-time model based on the control command received at the current time. In step S2, the estimator for each region receives the discrete-time k load frequency error and control command sent by the power plant in the current region, and updates the pseudo-partial derivative at the current time using a recursive estimation algorithm to obtain the updated pseudo-partial derivative estimate. The calculation method is as follows:

[0091]

[0092] in, It is a pseudo-partial derivative ψ j The estimated value of (k), yes The pseudo-partial derivative ψ at the previous moment j The estimated value of (k); ΔE j (k) represents the change in load frequency error, ΔE j (k)=E j (k)-E j (k-1), Ej (k) represents the load frequency error of the j-th region at the current discrete time k, E j (k-1) represents E j (k) The load frequency error at the previous discrete time k-1; Δu j (k-1) represents the change in control input, Δu j (k-1)=u j (k-1)-u j (k-2), u j (k-1) represents the control input of the j-th region at the previous discrete time k, u j (k-2) represents u j The control input at the previous discrete time step (k-1); ζ j ∈(0,1] represents the gain parameter, β j >0 indicates a smoothing parameter.

[0093] The estimator will use the updated pseudopartial derivative estimate. The load frequency error at time k is transmitted to the output trigger of the current region via the power network. In step S3, the method for constructing the first event trigger condition in the output trigger includes:

[0094]

[0095] in, This indicates the time of the qth output trigger in the j-th region. This indicates the (q-1)th output trigger time of the j-th region; This represents the change in load frequency error in the j-th region at discrete time k. E j (k) represents the load frequency error of the j-th region at discrete time k. This indicates that the j-th region is at the output trigger time. The load frequency error; and For the set trigger parameters, The dynamic event triggering function for the j-th region is designed as follows:

[0096]

[0097] Where, δ j ∈(0,1) is the set trigger parameter. To set parameters.

[0098] Furthermore, the output trigger determines whether the power system state change satisfies the first event triggering condition: if so, it uses the updated pseudo-partial derivative estimate. Replace the original pseudo-partial derivative; otherwise, leave the original pseudo-partial derivative unchanged.

[0099] Further, in this embodiment, in step S4, based on the pseudo-partial derivative obtained in step S3 and the load frequency error, encoding, data uniform quantization, and decoding operations are performed sequentially to obtain the decoded data. The method includes:

[0100] Construct the output encoder, whose function expression is as follows:

[0101]

[0102] Where E(x(k)) represents the encoded data at the current discrete time k, X(k) represents the data before encoding at the current discrete time k, ζ(k)∈(0,1) represents the scaling sequence for adjusting the encoder output amplitude; ξ(k) represents the internal state value of the encoder at the current discrete time k, ξ(k-1) represents the internal state value of the encoder at the previous discrete time, and ξ(1)=0;

[0103] The pseudo-partial derivatives obtained in step S3 and the load frequency error are encoded using the output encoder to obtain encoded data. A first uniform quantizer is then constructed. During the network transmission process where the output encoder transmits the encoded data to the decoder, the encoded data is uniformly quantized using the first uniform quantizer to obtain quantized data. The function expression of the first uniform quantizer is as follows:

[0104]

[0105] Where Q(E(X(k))) represents the quantized data, h∈Z + Indicates the quantization level parameter;

[0106] A decoder is constructed, with the function expression: D(X(k))=D(X(k-1))+ζ(k)Q(E(X(k))). The decoder is then used to process the quantized data to obtain the decoded data X. * (k):

[0107]

[0108] Among them, X * (k-1) represents the data decoded at discrete time k-1, ω(k) represents the quantization error term, and ω(k) = Q(E(X(k))) - E(X(k)) ≤ 0.06. The internal state value ξ(k) of the output encoder at the current time k is updated using the following formula:

[0109] ξ(k)=ζ(k)Q(E(X(k)))+ξ(k-1).

[0110] In this embodiment, in step S4, the decoder decodes the data X. * (k) The command is sent to the controller in the corresponding region. After receiving the decoded pseudo-partial derivative and the load frequency error at time k, the controller generates an updated control command. The specific method includes:

[0111] Based on the decoded data, the updated input control command u is calculated using the following formula. j (k):

[0112]

[0113] in, This indicates the (p-1)th input trigger time of the j-th region. Input control commands, v j ∈(0,1] represents the control gain coefficient of the j-th region, α j >0 is the regularization parameter. This is the decoded feedback term at time k. This is a pseudo-partial derivative estimate.

[0114] Further, in step S5, the controller sends the updated control command to the input trigger in the current area via the power network. The input trigger performs an input event mechanism judgment procedure using a constructed second event trigger condition, the expression of which is as follows:

[0115]

[0116] in, This represents the trigger time of the p-th input in the j-th region. This indicates the (p-1)th input trigger time for the j-th region; This represents the predicted change in the j-th region at time k, as predicted by the control command. This represents the relative threshold coefficient for the j-th region. This represents the absolute threshold constant for the j-th region; This indicates that the j-th region was at the last trigger time. The control command value.

[0117] The input trigger determines the amount of change in the control command. Whether the second event triggering condition is met: If yes, the current time is used as the input trigger time, and the current control instruction is updated using the updated control instruction; otherwise, the current control instruction remains unchanged. Specifically, the rules for updating the control instruction using the second event triggering condition are as follows:

[0118]

[0119] in,

[0120] Subsequently, the control command obtained in step S5 is sent to the input encoder. The input encoder encodes the received control command to obtain the encoded control command E(X(k)). The function expression for data processing by the input encoder is as follows:

[0121]

[0122] Where ζ(k)∈(0,1) represents the scaling sequence for adjusting the encoder output amplitude; ξ(k) represents the encoder's internal state value at the current time k, ξ(k-1) represents the encoder's internal state value at the previous time step of ξ(k), and ξ(1)=0;

[0123] A second uniform quantizer is constructed. During the network transmission process where the input encoder transmits the encoded control command to the input decoder, the second uniform quantizer performs a data uniform quantization operation on the encoded control command to obtain the quantized control command. The function expression of the second uniform quantizer is as follows:

[0124]

[0125] Where Q(E(u(k))) represents the quantized control command, h∈Z + Indicates the quantization level parameter;

[0126] The function expression for the input decoder processing the data is: D(X(k)) = D(X(k-1)) + ζ(k)Q(E(X(k))). Using the input decoder on the quantized control command, the decoded control command u is obtained. * (k):

[0127]

[0128] Among them, u * (k-1) represents the control command decoded at time k-1, ω(k) represents the quantization error term, and ω(k) = Q(E(u(k))) - E(u(k)) ≤ 0.06. The internal state value ξ(k) of the input encoder at the current time k is updated using the following formula:

[0129] ξ(k)=ζ(k)Q(E(u(k)))+ξ(k-1).

[0130] The input decoder will decode the control command u *(k) The control command is sent to the corresponding power plant in the region. The power plant executes the control command and calculates the load frequency error after the command is executed. The load frequency error and the controller command are sent to the estimator. Steps S1 to S6 are executed repeatedly so that the load frequency error of each region in the multi-region interconnected power system is within the preset error range.

[0131] The effectiveness of the method described in this invention will be verified through specific experiments below. Figure 4 Taking the power system structure shown as an example, the main parameters of the model are designed as follows:

[0132] D1=0.015, D2=0.016, D3=0.015, H1=0.157, H2=0.185, H3=0.124, R1=2.38, R2=2.49, R3=2.21,

[0133] T g1 =0.05,T g2 =0.07,T g3 =0.06,T t1 =0.41,T t2 =0.42,T t3 =0.38,

[0134] Γ1=0.35, Γ2=0.38, Γ3=0.36, S 12 =S 21 =0.25,S 13 =S 31 =0.35,S 23 =S 32 =0.22.

[0135] The main parameters of the controller are designed as follows:

[0136] α1=1.5, α2=1.4, α3=1.3, ζ1=0.85, ζ2=0.88, ζ3=0.84, v1=v3=0.8, v2=0.9, R1=β2=β3=1.3, λ a1 =λ a2 =λ a3 =1.2, λ b1 =λ b2 =λ b3 =1.0, λ c1 =λ c2 =λ c3 =0.5, δ1=0.05, δ2=0.04, δ T =0.05, ε=10 -5 , ζ1=ζ2=ζ3=1 / k.

[0137] Experiment 1: Verification of Control Performance under Step Load Disturbance

[0138] The initial state of the system is configured as a zero vector, and a step-change load ΔP is applied to three regions in the multi-region interconnected power system. dj =0.03. The method proposed in this invention remains effective under dynamic load changes. Figure 5 As shown in (a) and (b), the sampling results of event triggering time and triggering interval show that: out of a total of 72,000 samples in the three regions, the number of event triggers was 11,693, corresponding to a reduction in energy consumption of 83.76%. Figure 6 The medium-load frequency deviation (Δf) curve shows that the deviation converges rapidly to zero steady state under the action of the controller.

[0139] Experiment 2: Robustness verification under dynamic sinusoidal load disturbance

[0140] The initial state of the system is set to 0, and a dynamic load ΔP, including a sinusoidal component, is applied to the three regions. dj (k) = 0.03 + 0.004sin(5kπ / n). The adaptability of the method proposed in this invention to time-varying loads has been verified. Figure 7 The trigger interval data in (a) and (b) show that 33,818 triggers were performed out of 72,000 samples across the three regions, resulting in a 53.03% reduction in energy consumption. Figure 8 The medium-load frequency deviation curve shows that Δf converges rapidly to the minimum bounded neighborhood under the action of the controller, verifying the robustness of the method.

[0141] Example 2:

[0142] Based on the same inventive concept as Embodiment 1, this invention also provides a dual-event triggered load frequency control system combining uniform quantization and encoding / decoding. For example... Figure 9 As shown, the dual-event triggered load frequency control system combining uniform quantization and encoding / decoding includes: a discrete-time model construction module 100, a dynamic response gradient estimation module 200, an output event triggering decision module 300, a data processing and control instruction generation module 400, an input event triggering decision module 500, and a control execution and closed-loop feedback module 600.

[0143] The discrete-time model construction module 100 is configured to: construct a discrete-time model of any region operating in a multi-region interconnected power system, and calculate the load frequency error of each region based on the control command received at the current time through the discrete-time model;

[0144] The dynamic response gradient estimation module 200 is configured to: introduce pseudo-partial derivatives to represent the dynamic response gradient of the load frequency error to the control input; based on the load frequency error and the control command, update the pseudo-partial derivatives at the current time to obtain the updated pseudo-partial derivative estimate.

[0145] The output event triggering decision module 300 is configured to: construct a first event triggering condition, and based on the load frequency error, determine whether the power system state change meets the first event triggering condition; if so, replace the original pseudo-partial derivative with the updated pseudo-partial derivative estimate; otherwise, keep the original pseudo-partial derivative unchanged.

[0146] The data processing and control instruction generation module 400 is configured to: based on the pseudo-partial derivative obtained by the output event triggering decision module 300 and the load frequency error, perform encoding, data uniform quantization and decoding operations in sequence to obtain decoded data, and obtain updated control instructions based on the decoded data;

[0147] The input event triggering decision module 500 is configured to: construct a second event triggering condition, determine whether the change in the control command satisfies the second event triggering condition; if so, update the current control command through the updated control command; otherwise, keep the current control command unchanged.

[0148] The control execution and closed-loop feedback module 600 is configured to: sequentially encode, uniformly quantize, and decode the control command obtained by the input event triggering decision module 500 to obtain the decoded control command, and perform power regulation according to the decoded control command so that the load frequency error of each region in the multi-region interconnected power system is within a preset error range.

[0149] This embodiment proposes a dual-event triggered load frequency control system combining uniform quantization and encoding / decoding, used to implement the aforementioned dual-event triggered load frequency control method combining uniform quantization and encoding / decoding. Therefore, the specific implementation of the dual-event triggered load frequency control system combining uniform quantization and encoding / decoding can be found in the embodiment section of the aforementioned dual-event triggered load frequency control method combining uniform quantization and encoding / decoding. For example, the discrete-time model construction module 100, the dynamic response gradient estimation module 200, the output event triggering decision module 300, the data processing and control instruction generation module 400, the input event triggering decision module 500, and the control execution and closed-loop feedback module 600 are respectively used to implement steps S1, S2, S3, S4, S5, and S6 in the dual-event triggered load frequency control method combining uniform quantization and encoding / decoding in Embodiment 1. Therefore, its specific implementation can be referred to the description of the corresponding embodiment. To avoid redundancy, it will not be repeated here.

[0150] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0151] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0152] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0153] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0154] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A dual-event triggered load frequency control method combining uniform quantization and encoding / decoding, characterized in that, Includes the following steps: Step S1: Construct a discrete-time model of any region operating in a multi-region interconnected power system. Each region calculates the load frequency error based on the control command received at the current moment using the discrete-time model. Step S2: Introduce pseudo-partial derivatives to represent the dynamic response gradient of the system output to the control input. Based on the load frequency error and the control command, update the pseudo-partial derivatives at the current moment to obtain the updated pseudo-partial derivative estimate. Step S3: Construct the first event trigger condition. Based on the load frequency error, determine whether the power system state change satisfies the first event trigger condition. If so, replace the original pseudo-partial derivative with the updated pseudo-partial derivative estimate; Otherwise, keep the original pseudo-partial derivatives unchanged; Step S4: Based on the pseudo-partial derivatives obtained in step S3 and the load frequency error, perform encoding, data uniform quantization and decoding operations in sequence to obtain the decoded data, and obtain the updated control command based on the decoded data; Step S5: Construct the second event trigger condition and determine whether the change in the control command satisfies the second event trigger condition. If so, update the current control command using the updated control command; Otherwise, keep the current control commands unchanged; Step S6: The control command obtained in step S5 is sequentially encoded, data uniformly quantized, and decoded to obtain the decoded control command. Power regulation is performed according to the decoded control command so that the load frequency error of each region in the multi-region interconnected power system is within the preset error range. Then, return to step S1.

2. The dual-event triggered load frequency control method combining uniform quantization and encoding / decoding according to claim 1, characterized in that, In step S1, the method for constructing a discrete-time model of the operation of any region in a multi-region interconnected power system includes: Where, x j (k) represents the system state vector of the j-th region at the current discrete time k, x j (k+1) represents k j The system state vector at the next discrete time k+1 of (k); A j B j C j F j All are system matrices. E represents the perturbation vector; j (k) represents the actual regional control error of the j-th region at the current discrete time k, i.e., the load frequency error at the current discrete time k, u j (k) represents the control input of the j-th region at the current discrete time k.

3. The dual-event triggered load frequency control method combining uniform quantization and encoding / decoding according to claim 1, characterized in that, In step S2, the method for updating the pseudo-partial derivative at the current moment based on the load frequency error and the control command to obtain the updated pseudo-partial derivative estimate includes: in, It is a pseudo-partial derivative ψ j The estimated value of (k), yes The pseudo-partial derivative ψ at the previous moment j The estimated value of (k); ΔE j (k) represents the change in system output, ΔE j (k)=E j (k)-E j (k-1), E j (j) represents the load frequency error of the j-th region at the current discrete time k, E j (k-1) represents E j (k) The load frequency error at the previous discrete time k-1; Δu j (k-1) represents the change in control input, Δu j (k-1)=u j (k-1)-u j (k-2), u j (k-1) represents the control input of the j-th region at the previous discrete time k-1 at the current discrete time k, u j (k-2) represents u j The control input at the previous discrete time k-2 of (k-1); β represents the gain parameter. j >0 indicates a smoothing parameter.

4. The dual-event triggered load frequency control method combining uniform quantization and encoding / decoding according to claim 1, characterized in that, In step S3, the method for constructing the first event triggering condition includes: in, This indicates the time of the qth output trigger in the j-th region. This indicates the (q-1)th output trigger time of the j-th region; This represents the change in load frequency error in the j-th region at discrete time k. E j (k) represents the load frequency error of the j-th region at discrete time k. This indicates the output trigger time for the j-th region. The load frequency error; and For the set trigger parameters, The dynamic event triggering function for the j-th region is designed as follows: Where, δ j ∈(0,1) is the set trigger parameter. To set parameters.

5. The dual-event triggered load frequency control method combining uniform quantization and encoding / decoding according to claim 1, characterized in that, In step S4, based on the pseudo-partial derivatives obtained in step S3 and the load frequency error, encoding, data uniform quantization, and decoding operations are performed sequentially to obtain the decoded data. The method includes: The encoder is constructed using the following function expression: Where E(X(k)) represents the encoded data at the current discrete time k, X(k) represents the data before encoding at the current discrete time k, ζ(k)∈(0,1) represents the scaling sequence for adjusting the encoder output amplitude; ξ(k) represents the encoder's internal state value at the current time k, ξ(k-1) represents the encoder's internal state value at the previous time step of ξ(k), and ξ(1)=0; The encoder is used to encode the pseudo-partial derivatives obtained in step S3 and the load frequency error to obtain encoded data. A uniform quantizer is then constructed, and the encoded data is subjected to uniform quantization using the uniform quantizer to obtain quantized data. The function expression of the uniform quantizer is as follows: Where Q(E(X(k))) represents the quantized data, h∈Z + Indicates the quantization level parameter; Construct a decoder, and use the decoder to process the quantized data to obtain the decoded data X. * (k): Among them, X * (k-1) represents the data decoded at discrete time k-1, ω(k) represents the quantization error term, and ω(k)=Q(E(X(k)))-E(X(k))≤0.

06.

6. The dual-event triggered load frequency control method combining uniform quantization and encoding / decoding according to claim 5, characterized in that, The encoder's internal state value ξ(k) at the current discrete time k is updated using the following formula: ξ(k)=ζ(k)Q(E(X(k)))+ξ(k-1).

7. The dual-event triggered load frequency control method combining uniform quantization and encoding / decoding according to claim 5, characterized in that, The function expression of the decoder is: D(X(k))=D(X(k-1))+ζ(k)Q(E(X(k))).

8. The dual-event triggered load frequency control method combining uniform quantization and encoding / decoding according to claim 1, characterized in that, In step S4, the method for obtaining the updated control command based on the decoded data includes: Based on the decoded data, the updated input control command u is calculated using the following formula. j (k): in, This indicates the (p-1)th input trigger time of the j-th region. Input control commands, v j ∈(0,1] represents the control gain coefficient of the j-th region, α j >0 is the regularization parameter. This is the decoded feedback term at time k. This is a pseudo-partial derivative estimate.

9. The dual-event triggered load frequency control method combining uniform quantization and encoding / decoding according to claim 1, characterized in that, In step S5, the method for constructing the second event triggering condition includes: in, This represents the time of the p-th input trigger in the j-th region. This indicates the (p-1)th input trigger time for the j-th region; This represents the predicted change in the j-th region at time k, as predicted by the control command. This represents the relative threshold coefficient for the j-th region. This represents the absolute threshold constant for the j-th region; This indicates that the j-th region was at the last trigger time. The control command value.

10. A dual-event triggered load frequency control system combining uniform quantization and encoding / decoding, characterized in that, The dual-event triggered load frequency control system combining uniform quantization and encoding / decoding includes the following modules: The discrete-time model construction module is configured to: construct a discrete-time model of any region operating in a multi-region interconnected power system, and calculate the load frequency error of each region based on the control command received at the current moment through the discrete-time model; The dynamic response gradient estimation module is configured to: introduce pseudo-partial derivatives to represent the dynamic response gradient of the system output to the control input, update the pseudo-partial derivatives at the current time based on the load frequency error and the control command, and obtain the updated pseudo-partial derivative estimate. The output event triggering decision module is configured to: construct a first event triggering condition, and based on the load frequency error, determine whether the power system state change meets the first event triggering condition; if so, replace the original pseudo-partial derivative with the updated pseudo-partial derivative estimate; otherwise, keep the original pseudo-partial derivative unchanged. The data processing and control instruction generation module is configured to: based on the pseudo-partial derivative obtained by the output event triggering decision module and the load frequency error, perform encoding, data uniform quantization and decoding operations in sequence to obtain decoded data, and obtain updated control instructions based on the decoded data; The input event triggering decision module is configured to: construct a second event triggering condition, determine whether the change in the control command meets the second event triggering condition; if so, update the current control command through the updated control command; otherwise, keep the current control command unchanged. The control execution and closed-loop feedback module is configured to: sequentially encode, uniformly quantize, and decode the control commands obtained by the input event triggering decision module to obtain decoded control commands, and perform power regulation according to the decoded control commands so that the load frequency error of each region in the multi-region interconnected power system is within a preset error range.

Images