Differential protection control method and device based on information freshness, electronic equipment and storage medium

Abstract

The invention discloses a differential protection control method and device based on information freshness, electronic equipment and a storage medium, and belongs to the technical field of differential protection. The sampling time of the system state measurement parameters, the receiving time when the system state measurement parameters are received and the communication delay of the system state measurement parameters in the transmission process are obtained; calculating information freshness corresponding to the system state measurement parameters; the method comprises the following steps: constructing a target function of system differential protection by taking a system state measurement parameter as a variable, taking information freshness as a stealth regulation factor of system long-term average cost and taking the minimum system long-term average cost as a target; and solving the objective function to obtain a control strategy of differential protection of the system when the long-term average cost of the system is minimum, and performing differential protection control on the power system, so that the problem that the freshness of information in decision making is not considered in the prior art can be solved.

Description

Technical Field

[0001] The present invention relates to the field of differential protection technology, and in particular to a differential protection control method, device, electronic equipment and storage medium based on information freshness. Background Art

[0002] Differential protection is the cornerstone of power system reliability, responsible for detecting and isolating faults in transmission lines, transformers, and other critical components. Differential protection compares electrical quantities from different terminals in the protected area and triggers protective measures when the difference exceeds a predefined threshold. Traditional differential protection schemes assume near-instantaneous and lossless communication of measurements between relays and decision-making centers. However, in the context of modern smart grids, this assumption is increasingly untenable. The emergence of smart grids has brought unprecedented challenges and opportunities to traditional differential protection schemes.

[0003] On the one hand, smart grids enable real-time data collection and processing, improving fault detection capabilities. On the other hand, despite significant advances in power system protection, the integration of communication and control remains an open challenge. Traditional protection schemes typically do not consider the freshness and timeliness of information in decision-making, treating delays as unavoidable noise or assuming a fixed delay distribution, ignoring information timeliness. Limitations such as communication delays and the inability to dynamically adapt to real-time information often result in reduced fault detection accuracy and delayed system response. This can make differential protection systems susceptible to errors caused by outdated measurements, threatening the reliability and stability of the power grid. Summary of the Invention

[0004] The present invention provides a differential protection control method, device, electronic device and storage medium based on information freshness, which can solve the problem that the existing technology does not consider the freshness and timeliness of information in decision-making, making the differential protection system susceptible to errors caused by outdated measurements, threatening the reliability and stability of the power grid.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a differential protection control method based on information freshness, including:

[0006] receiving system state measurement parameters collected by a sampler, and obtaining a sampling time of the system state measurement parameters, a reception time when the system state measurement parameters are received, and a communication delay of the system state measurement parameters during transmission;

[0007] Calculating the information freshness corresponding to the system state measurement parameter based on the sampling time, the receiving time, and the communication delay;

[0008] Taking the system state measurement parameters as variables and the information freshness as an invisible adjustment factor of the system long-term average cost, an objective function of system differential protection is constructed based on the information freshness and the system state measurement parameters with the goal of minimizing the system long-term average cost;

[0009] The objective function is solved to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized, and the control strategy is mapped to a corresponding differential protection control action, and then differential protection control of the power system is performed according to the differential protection control action.

[0010] As a preferred solution, the information freshness is calculated using the following formula:

[0011] Δ(t)=tS i ,D i ≤t≤D i+1 ;

[0012] S i+1 ≥D i ;

[0013] D i =S i +Y i ;

[0014] Among them, Δ(t) is the freshness of information at time t, S i is the sampling time of the i-th system state measurement parameter group, S i+1 is the sampling time of the i+1th system state measurement parameter group, D i is the receiving time when the i-th system status measurement parameter group is received, D i+1 Y is the receiving time when the i+1th system status measurement parameter group is received, i The communication delay of the parameter packet during transmission is measured for the i-th system state.

[0015] As a preferred solution, the objective function of system differential protection is constructed based on the information freshness and the system state measurement parameters with the goal of minimizing the long-term average cost of the system, including:

[0016] According to the information freshness and the system state measurement parameters, a first objective function of the system differential protection is constructed with the goal of minimizing the long-term average cost of the system;

[0017] Taking a tuple consisting of a system state measurement parameter, a communication delay during transmission of the system state measurement parameter, and a differential protection control action generated upon receiving a previous system state measurement parameter as a corresponding sufficient statistic, and then optimizing the first objective function based on the sufficient statistic to transform it into a corresponding second objective function;

[0018] The long-term average cost solution in the second objective function is converted into the ratio of the average cost in each period to the period length to obtain the corresponding third objective function;

[0019] A preset cost parameter is added to the third objective function, the third objective function is converted into a corresponding fourth objective function, and then the fourth objective function is used as the objective function of the system differential protection.

[0020] As a preferred solution, the objective function is solved to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized, including:

[0021] Converting the objective function into a corresponding infinite-horizon MDP; wherein the infinite-horizon MDP includes: a state space, an action space, a transition probability, and a cost function;

[0022] The objective function is solved according to the state space, action space, transition probability and cost function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized.

[0023] Based on the above embodiment, another embodiment of the present invention provides a differential protection control device based on information freshness, comprising: a system state measurement parameter acquisition module, an information freshness calculation module, an objective function construction module, and a differential protection control module;

[0024] The system state measurement parameter acquisition module is used to receive the system state measurement parameter collected by the sampler, and obtain the sampling time of the system state measurement parameter, the reception time when the system state measurement parameter is received, and the communication delay of the system state measurement parameter during the transmission process;

[0025] The information freshness calculation module is used to calculate the information freshness corresponding to the system state measurement parameter based on the sampling time, receiving time and communication delay;

[0026] The objective function construction module is configured to use the system state measurement parameter as a variable and the information freshness as an invisible adjustment factor of the system long-term average cost, and to construct an objective function of the system differential protection based on the information freshness and the system state measurement parameter with the goal of minimizing the system long-term average cost;

[0027] The differential protection control module is used to solve the objective function to obtain the control strategy of the system differential protection when the long-term average cost of the system is minimized, and map the control strategy into a corresponding differential protection control action, and then perform differential protection control on the power system according to the differential protection control action.

[0028] As a preferred solution, the information freshness is calculated using the following formula:

[0029] Δ(t)=tS i ,D i ≤t≤D i+1 ;

[0030] S i+1 ≥D i ;

[0031] D i =S i +Y i ;

[0032] Among them, Δ(t) is the freshness of information at time t, S i is the sampling time of the i-th system state measurement parameter group, S i+1 is the sampling time of the i+1th system state measurement parameter group, D i is the receiving time when the i-th system status measurement parameter group is received, D i+1 Y is the receiving time when the i+1th system status measurement parameter group is received, i The communication delay of the parameter packet during transmission is measured for the i-th system state.

[0033] As a preferred solution, the objective function of system differential protection is constructed based on the information freshness and the system state measurement parameters with the goal of minimizing the long-term average cost of the system, including:

[0034] According to the information freshness and the system state measurement parameters, a first objective function of the system differential protection is constructed with the goal of minimizing the long-term average cost of the system;

[0035] Taking a tuple consisting of a system state measurement parameter, a communication delay during transmission of the system state measurement parameter, and a differential protection control action generated upon receiving a previous system state measurement parameter as a corresponding sufficient statistic, and then optimizing the first objective function based on the sufficient statistic to transform it into a corresponding second objective function;

[0036] The long-term average cost solution in the second objective function is converted into the ratio of the average cost in each period to the period length to obtain the corresponding third objective function;

[0037] A preset cost parameter is added to the third objective function, the third objective function is converted into a corresponding fourth objective function, and then the fourth objective function is used as the objective function of the system differential protection.

[0038] As a preferred solution, the objective function is solved to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized, including:

[0039] Converting the objective function into a corresponding infinite-horizon MDP; wherein the infinite-horizon MDP includes: a state space, an action space, a transition probability, and a cost function;

[0040] The objective function is solved according to the state space, action space, transition probability and cost function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized.

[0041] Based on the above embodiments, another embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the differential protection control method based on information freshness described in the above embodiment of the invention is implemented.

[0042] Based on the above embodiments, another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the differential protection control method based on information freshness described in the above invention embodiment.

[0043] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0044] The present invention provides a differential protection control method based on information freshness, which receives system state measurement parameters collected by a sampler, and obtains the sampling time of the system state measurement parameters, the receiving time when the system state measurement parameters are received, and the communication delay of the system state parameters during the transmission process; according to the sampling time, the receiving time, and the communication delay, the information freshness corresponding to the system state measurement parameters is calculated; the system state measurement parameters are used as variables, and the information freshness is used as an invisible adjustment factor of the long-term average cost of the system. According to the information freshness and the system state measurement parameters, the objective function of the system differential protection is constructed with the goal of minimizing the long-term average cost of the system; the objective function is solved to obtain the control strategy of the system differential protection when the long-term average cost of the system is minimized, and the control strategy is mapped to the corresponding differential protection control action, and then the power system is differentially protected and controlled according to the differential protection control action. The present invention calculates the corresponding information freshness through the sampling time, reception time and communication delay of the system state measurement parameters, and fully considers the influence of information freshness, that is, information timeliness, when constructing the objective function of the system differential protection. This makes the differential protection system less susceptible to errors caused by outdated measurements, and can improve the reliability and stability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of a differential protection control method based on information freshness provided by one embodiment of the present invention;

[0046] Figure 2 It is an MDP framework diagram of the power system differential protection control method of the present invention;

[0047] Figure 3 It is the overall flow chart of the differential protection control of the present invention;

[0048] Figure 4 This is a structural diagram of a differential protection control device based on information freshness provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions in this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0051] In the description of the embodiments of this application, the technical terms "first" and "second" are used only to distinguish different objects and should not be understood to indicate or imply relative importance or implicitly specify the quantity, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, the meaning of "plurality" is more than two, unless otherwise clearly and specifically defined.

[0052] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0053] In the description of the embodiments of this application, the term "and / or" is simply a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent the following three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0054] In the description of the embodiments of the present application, the terms "multiple" and "several" refer to more than two (including two). Similarly, "multiple groups" refer to more than two groups (including two groups), and "multiple pieces" refer to more than two pieces (including two pieces).

[0055] In the description of the embodiments of the present application, unless otherwise expressly specified or limited, technical terms such as "installed," "connected," "connected," and "fixed" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integration; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; internal connections between two components or interactions between two components. Those skilled in the art can understand the specific meanings of the above terms in the embodiments of the present application based on specific circumstances.

[0056] Example 1

[0057] Please refer to Figure 1To address the problem that existing technologies fail to consider the freshness and timeliness of information in decision-making, making differential protection systems susceptible to errors caused by outdated measurements and threatening the reliability and stability of the power grid, one embodiment of the present invention provides a flow chart of a differential protection control method based on information freshness. The present invention proposes an optimized differential protection strategy based on the age of information (AoI), which is a metric reflecting information freshness. By modeling the differential protection problem as a Markov decision process (MDP) (where the next state depends only on the current state and the current control action), the AoI is used as a key factor to capture the freshness of decision-making information under random delays. Treating the AoI as auxiliary information rather than an independent optimization objective can significantly improve protection performance and achieve faster and more accurate fault detection. Specifically, in a differential protection scenario, the current state includes the current operating parameters of the power system (such as current and voltage) and communication delays. Actions include control actions (such as tripping and blocking) and sampling decisions (such as whether to generate a new data packet). State transitions indicate that the next state is determined by both the current state and the action. The present invention uses AoI and system status as state parameters of MDP, dynamically balancing information freshness and control objectives, including the following specific steps:

[0058] S1. Receive system state measurement parameters collected by a sampler, and obtain a sampling time of the system state measurement parameters, a reception time when the system state measurement parameters are received, and a communication delay of the system state measurement parameters during transmission;

[0059] For details, please refer to Figure 2 , which is the MDP framework diagram of the power system differential protection control method of the present invention, including power generation equipment, power system components, samplers, communication networks and decision terminals.

[0060] The sampler collects parameters related to differential protection judgment in power system components (i.e., system state measurement parameters); and transmits the system state measurement parameters to the decision-making end through the communication network (there is a communication delay when data is transmitted in the communication network).

[0061] like Figure 2 As shown in the figure, the proposed information freshness-sensitive MDP framework demonstrates the real-time nature of communication delay and information. The monitored power system components are represented by X. Its state evolves according to the Markov process, and its transition probability (represents the state in state X) t Next action u t Then, transfer to state X t+1 The probability) is given by Pr(X t+1 |X t ,u t ) control, where is the control action taken at time t. The system state measurement parameters are sampled by the sampler and transmitted to the remote decision maker via the communication network. The communication network introduces random delays, which means that at time t, the decision maker observes the delayed state X t-Δ(t) , where Δ(t) represents the AoI, which quantifies the freshness of the recently received information, where a larger AOI indicates that the data may be more outdated.

[0062] Wherein, the state measurement quantity is X t ,∈ includes the variables required for judging protection (such as voltage and current). For example, the principle of current differential protection is to input the current vector difference at both ends of the current transformer, and start the action element when the set action value is reached.

[0063] S2. Calculating the information freshness corresponding to the system state measurement parameter based on the sampling time, the receiving time, and the communication delay;

[0064] Preferably, the information freshness is calculated by the following formula:

[0065] Δ(t)=tS i ,D i ≤t≤D i+1 ;

[0066] S i+1 ≥D i ;

[0067] D i =S i +Y i ;

[0068] Among them, Δ(t) is the freshness of information at time t, S i is the sampling time of the i-th system state measurement parameter group, S i+l is the sampling time of the i+1th system state measurement parameter group, D i is the receiving time when the i-th system status measurement parameter group is received, D i+1 Y is the receiving time when the i+1th system status measurement parameter group is received, i The communication delay of the parameter packet during transmission is measured for the i-th system state.

[0069] Specifically, at each time step t, the sampler decides whether to generate a new measurement packet. If a new sample (and system state measurement parameter) is generated, it is encapsulated into a packet for transmission, which results in random communication delay. And at time D i =S i +Y i Receive group (i.e. system status measurement parameter group), where Si is the sampling time of the ith packet. The AoI at time t is defined as the distance between t and the most recent sampling time S i The time difference between

[0070] Δ(t)=tS i ,D i ≤t≤D i+1 (1)

[0071] To ensure that the communication channel is not overloaded, the system enforces the constraint that a new sample is generated only after the previous data packet has been successfully transmitted. The sampler must wait for the previous data packet to be transmitted before it can take the next sample, i.e., group:

[0072] S i+1 ≥D i (2)

[0073] S3. Using the system state measurement parameters as variables and the information freshness as an invisible adjustment factor of the system's long-term average cost, constructing an objective function for system differential protection based on the information freshness and the system state measurement parameters with the goal of minimizing the system's long-term average cost;

[0074] Preferably, the objective function of the system differential protection is constructed based on the information freshness and the system state measurement parameters with the goal of minimizing the long-term average cost of the system, including: constructing a first objective function of the system differential protection based on the information freshness and the system state measurement parameters with the goal of minimizing the long-term average cost of the system; taking a tuple consisting of the system state measurement parameters, the communication delay of the system state measurement parameters during transmission, and the differential protection control action generated when the previous system state measurement parameters are received as a corresponding sufficient statistic, and then optimizing the first objective function according to the sufficient statistics to convert it into a corresponding second objective function; converting the long-term average cost solution in the second objective function into a ratio solution of the average cost in each cycle and the cycle length to obtain a corresponding third objective function; adding a preset cost parameter to the third objective function, converting the third objective function into a corresponding fourth objective function, and then using the fourth objective function as the objective function of the system differential protection.

[0075] Specifically, the decision maker selects the differential protection control action based on the most recently received state and AoI. The observation value at time t is represented by a pair Composition, of which It is at time D i The received state, Δ(t) is the corresponding AoI. Decision strategy π t Map the current observation to the action space The differential protection control action u selected in t :

[0076]

[0077] Mapping form: π t As a decision strategy (using reinforcement learning to achieve mapping), input Output u t , and then gradually minimize the long-term cost through training.

[0078] The control action is updated only when a new sample arrives. Between the arrival of new samples, the control action remains constant:

[0079]

[0080] Among them U i is the control action determined when the i-th packet is received.

[0081] 1. Optimization objectives:

[0082] At each time step, the system incurs a cost c(X t ,u t ), which reflects the penalty for deviations from desired operating conditions or inefficient control actions. The system's goal is to minimize the long-term average cost. Therefore, the system state measurement parameters are used as variables, and the information freshness is used as an implicit adjustment factor for the system's long-term average cost. Based on the information freshness and the system state measurement parameters, with the goal of minimizing the system's long-term average cost, the first objective function of the system differential protection is constructed:

[0083]

[0084] The goal is therefore to find a control policy π that minimizes the expected long-term cost while taking into account AoI and communication delay.

[0085] In the formula, c(X t ,u t ) is the cost function, which is a function related to the state variables and control actions. Summing and then dividing by the number of sums is the average. inf represents the maximum lower bound of a set, and sup represents the minimum upper bound of a set. Limsupt→∞ represents the supremum of the limit as t tends to infinity. infπ represents the minimum value for the control π policy. Therefore, the overall expression represents the optimality of the π policy in infinite time and is used to calculate the minimum value in the long-term time limit.

[0086] 2. Optimal sampling and decision-making strategy for differential protection:

[0087] This paper develops an optimal strategy for joint sampling and decision making in differential protection systems, addressing the challenges posed by random communication delays and evolving AoI. The optimization problem is reformulated to simplify its solution:

[0088] 2.1. Sufficient statistics of differential protection:

[0089] To design an effective policy, one must reduce computational complexity by identifying sufficient statistics that can capture all necessary information from the observation history without losing any relevant details. This allows the system to make optimal decisions without having to process an ever-expanding observation history.

[0090] At time t, the historical information of observations and actions is expressed as These include things like sampling status Sampling time S j , delivery time D j , communication delay Y j , sampling action and the control action A for each packet received until time t j However, tracking the entire history is computationally complex as its size grows over time.

[0091] To solve this problem, we define a sufficient statistic S at time t t , it can be seen from history All relevant information is captured in the statistic without losing decision-relevant details. t To summarize the key information needed for decision making. In the case of differential protection, sufficient statistics can be expressed as a tuple:

[0092] S t =(X t-Δ(t) ,Δ(t),A t-1 ), (6)

[0093] where X t-Δ(t) is the most recent state observation, Δ(t) is AoI, A t-1 is the control action taken at the previous time step.

[0094] Since the control action is updated only when new samples arrive and remains constant between arrivals, the interval [D i ,D i+1 ) can be expressed as:

[0095]

[0096] in is the sampling time S iObserved state, Y i =D i -S i is the communication delay, A i-1 Is the control action taken when the previous packet is received.

[0097] By introducing sufficient statistics, the decision-making process is greatly simplified, allowing the system to make optimal decisions without having to process continuously growing historical observation data, ensuring the efficiency and accuracy of modern power grids.

[0098] 2.2. Reformulation of the optimization problem:

[0099] Therefore, using sufficient statistics A decision strategy can be represented as a mapping:

[0100]

[0101] The input of this step is sufficient statistics, which respectively represent the system state at the iith sampling (such as the differential current value), the communication delay at the iith sampling (such as the transmission time of 2ms), and the control action taken when the last data packet arrived (such as "lock protection"). The output is the joint probability distribution The mapping goal is to Calculate a probability distribution, indicating the choice of an action pair under the current information ( and a t The probability that the states of each can have different combinations).

[0102] This mapping represents the mapping from state to action. The sufficient statistic is the state, and the latter is the action. Because of the Markov property, the next state only depends on the current state and action, so the strategy only needs to consider the current state.

[0103] Among them, {0,1} corresponds to the sampling decision and Corresponding to control action a t The optimization problem can then be reformulated as the following second objective function to focus on these sufficient statistics:

[0104]

[0105] This reconstruction exploits sufficient statistics to reduce the original problem complexity.

[0106] P1 optimization is based on a complete observation history, requiring many historical states and increasing computational complexity over time. P2 introduces sufficient statistics to compress historical information. The reconstructed problem satisfies the Markov property (the current decision is independent of the history), making MDP theory applicable. It retains all decision-related information in the history and discards redundant details. Provides the latest status information; Y i The freshness of information is quantified, which represents the age of information (AoI). i-1 Records the impact of the last action on the current state. And in the interval [D i ,D i+1 ) The internal control action is fixed to a t =A i .

[0107] 2.3. Existence of the optimal strategy:

[0108] Reformulated question satisfies the standard conditions for MDPs with bounded costs. Therefore, by the Bellman equation, there exists an optimal stationary deterministic policy π * This strategy will take each sufficient statistic Mapped to a specific action pair No randomization or time dependence is required:

[0109]

[0110] 2.4、The meaning of differential protection:

[0111] sufficient statistics Significantly reduces the computational complexity of the optimization problem. The decision-making process does not depend on the entire history This enables the optimal strategy to be implemented in real time in differential protection systems.

[0112] By integrating AoI into the decision-making process, the framework ensures timely and accurate fault detection even under random delays.

[0113] 2.5. Reformulation of the protection issue:

[0114] The objective of differential protection is to minimize the long-term average cost, which represents the combined cost of delayed fault detection and incorrect protection action. Exploiting the regenerative nature of the sampling process, the problem is reformulated as the following third objective function:

[0115]

[0116] Among them Ci represents the cumulative cost incurred in the i-th period, represents the expected cost of a single cycle, Z i +Y i Indicates the total interval time (cycle length) between two consecutive decisions, Indicates the expected duration of a single cycle.

[0117] The regenerative property of the sampling process is used to transform the long-term average problem into a problem of the ratio of the average cost of each cycle to the cycle length. In the regeneration theory, the system will "regenerate" at certain moments, that is, the state returns to a certain reference point, making each cycle independent of each other. In this way, the long-term average can be decomposed into the average of each independent cycle. P3 is to minimize the average cost per unit time. Each regeneration cycle contains the cumulative cost and the cycle length. P2 is to minimize the time average cost. From P2 to P3, the time average cost of the infinite time domain is converted into the statistical average of the finite cycle, and the optimization variable is converted from the continuous time strategy π to the periodic strategy sequence φ0:∞.

[0118] Why can P2 be reformulated as P3: 1) Reproducibility ensures that each period can be considered independent and identically distributed, 2) Due to the law of large numbers, if the period tends to infinity, the fractional form of P3 converges to the same limit as P2.

[0119] For the differential protection problem, the expected cost C i It can be broken down into:

[0120]

[0121] Among them A i is the control action determined when the i-th sample is received.

[0122] At the same time, using Dinkelbach's method, a parameter λ is introduced to It becomes another auxiliary optimization problem with λ, which can be reformulated as the following fourth objective function:

[0123]

[0124] S4. Solve the objective function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized, map the control strategy into a corresponding differential protection control action, and then perform differential protection control on the power system according to the differential protection control action.

[0125] Preferably, solving the objective function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized includes: converting the objective function into a corresponding infinite time domain MDP; wherein the infinite time domain MDP includes: state space, action space, transition probability and cost function; solving the objective function according to the state space, action space, transition probability and cost function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized.

[0126] Specifically, when solving the objective function, the problem is further It is transformed into a standard infinite-horizon MDP, suitable for efficient computation.

[0127] Auxiliary Problem (Equivalent MDP for Differential Protection) can be represented as a standard MDP with the following content:

[0128] State Space:

[0129] Action space:

[0130] Transition probability: defined as Capture X t Dynamic and random delay Y i+1 ;

[0131] Cost function:

[0132]

[0133] in reflects the expected cost within a sampling period, q(.) reflects the control performance such as misoperation cost and delay penalty, Strive to reduce waiting time and communication delay to balance cost and real-time performance.

[0134] The core idea behind converting to an MDP is to treat each regeneration cycle i as a decision-making stage in an MDP. By defining states, actions, transition probabilities, and a cost function, P4 can be mapped to a standard MDP. The four elements of an MDP are all reflected in P4, so P4 can be rewritten as an MDP, meaning that P4 is mapped to a standard MDP. Their goal is to find the optimal strategy to minimize long-term cumulative cost.

[0135] Differential protection optimization, solving the optimal strategy:

[0136] For equivalent MDPs, if the state space and action space are finite and satisfy the single-chain property, then there exists an optimal stationary deterministic policy that minimizes the long-term average cost.

[0137] An algorithm for solving infinite-time domain MDPs in the context of differential protection, aiming to determine the optimal cost parameter h * , so that U(h * )=0, thus ensuring cost-effective and timely decision-making in modern smart grids.

[0138] The optimal cost parameter is actually represents the optimal average cost per unit time.

[0139] Since the fractional optimization problem P3 is difficult to solve, the parameter λ is introduced to transform P3 into P4. Finding the optimal solution of P3 (the minimum value of the ratio) is equivalent to finding the zero point of the function P4.

[0140] The present invention proposes a grid-aware fixed-point iteration (GFP-Iteration) algorithm that transforms the problem into a set of fixed-point equations and solves them directly.

[0141] (Fixed point representation): Using the Bayer equation, the root problem U(λ) = 0 is equivalent to solving the following nonlinear fixed point equation:

[0142]

[0143] in is a state-action tuple, γ ref is an arbitrarily chosen reference state-action tuple.

[0144] GFP iterative algorithm iteratively updates W * (γ) and h * Until convergence, as shown in Algorithm 1.

[0145] The GFP iterative algorithm uses fixed-point iteration to directly approximate the solution of the equations, alternately updating the strategy and value function until convergence, and finally obtains the optimal strategy π * , avoiding the overhead of nested iteration. MDP (h * ) represents the parameterized MDP model, including the optimal strategy π * Related structures (such as state transitions, reward functions, etc.)

[0146] Algorithm 1: Grid-aware fixed-point (GFP) iteration:

[0147] (1) Input: threshold ∈>0, MDPP MDP (h * );

[0148] (2) Initial value: W0=0, h0=0;

[0149] (3)while||W k -Wk-1 ||≥∈;

[0150] (4)h k =H(W k-1 );

[0151] (5)W k =T(W k-1 ,h k );

[0152] (6)End;

[0153] (7) Output: h * =h k .

[0154] The optimal strategy finally obtained has the following parameters:

[0155] 1. State parameters (input) The function is to serve as decision input, representing the current system state and historical actions;

[0156] 2. Action parameters (output), for each state Generate an action pair, including Zi waiting time decision and Ai control action. Decision output controls sampling frequency and protection action;

[0157] 3. Optimal cost parameter H*;

[0158] 4. Value function parameter W * (γ).

[0159] Among them, the optimal strategy is π * , is the mapping rule from state to action, which can directly generate control and sampling actions to achieve rapid response to faults. The steps are: data acquisition (obtaining the latest system state and calculating AoI) → state generation gi (compressing historical information and retaining key decision features) → generating action pairs, including Zi waiting time decision and Ai control action → executing action.

[0160] Please refer to Figure 3, is the overall flow chart of the differential protection control of the present invention. It can be seen that the present invention provides a differential protection control method based on information freshness. The present invention calculates the corresponding information freshness through the sampling time, reception time and communication delay of the system state measurement parameters, and fully considers the influence of information freshness, that is, information timeliness, when constructing the objective function of the system differential protection, so that the differential protection system is not easily affected by errors caused by outdated measurements, which can improve the reliability and stability of the power grid. Unlike the standard MDP formula that assumes instantaneous state observation, the present invention defines AoI as a dynamically evolving random process and defines it as a function of AoI. This framework not only captures the impact of communication delay, but also can make adaptive decisions based on the real-time state of the communication network; a new protection mechanism is proposed, which proves that even under poor network conditions, the real-time and accuracy of fault detection can be simultaneously improved by optimizing information timeliness.

[0161] Example 2

[0162] Please refer to Figure 4 , is a schematic structural diagram of a differential protection control device based on information freshness provided by an embodiment of the present invention, the device comprising: a system state measurement parameter acquisition module, an information freshness calculation module, an objective function construction module, and a differential protection control module;

[0163] The system state measurement parameter acquisition module is used to receive the system state measurement parameter collected by the sampler, and obtain the sampling time of the system state measurement parameter, the reception time when the system state measurement parameter is received, and the communication delay of the system state measurement parameter during the transmission process;

[0164] The information freshness calculation module is used to calculate the information freshness corresponding to the system state measurement parameter based on the sampling time, receiving time and communication delay;

[0165] The objective function construction module is configured to use the system state measurement parameter as a variable and the information freshness as an invisible adjustment factor of the system long-term average cost, and to construct an objective function of the system differential protection based on the information freshness and the system state measurement parameter with the goal of minimizing the system long-term average cost;

[0166] The differential protection control module is used to solve the objective function to obtain the control strategy of the system differential protection when the long-term average cost of the system is minimized, and map the control strategy into a corresponding differential protection control action, and then perform differential protection control on the power system according to the differential protection control action.

[0167] Preferably, the information freshness is calculated by the following formula:

[0168] Δ(t)=tS i ,D i ≤t≤D i+1 ;

[0169] S i+1 ≥D i ;

[0170] D i =S i +Y i ;

[0171] Among them, Δ(t) is the freshness of information at time t, S i is the sampling time of the i-th system state measurement parameter group, S i+1 is the sampling time of the i+1th system state measurement parameter group, D i is the receiving time when the i-th system status measurement parameter group is received, D i+1 Y is the receiving time when the i+1th system status measurement parameter group is received, i The communication delay of the parameter packet during transmission is measured for the i-th system state.

[0172] Preferably, constructing the objective function of system differential protection based on the information freshness and the system state measurement parameters with the goal of minimizing the long-term average cost of the system includes:

[0173] According to the information freshness and the system state measurement parameters, a first objective function of the system differential protection is constructed with the goal of minimizing the long-term average cost of the system;

[0174] Taking a tuple consisting of a system state measurement parameter, a communication delay during transmission of the system state measurement parameter, and a differential protection control action generated upon receiving a previous system state measurement parameter as a corresponding sufficient statistic, and then optimizing the first objective function based on the sufficient statistic to transform it into a corresponding second objective function;

[0175] The long-term average cost solution in the second objective function is converted into the ratio of the average cost in each period to the period length to obtain the corresponding third objective function;

[0176] A preset cost parameter is added to the third objective function, the third objective function is converted into a corresponding fourth objective function, and then the fourth objective function is used as the objective function of the system differential protection.

[0177] Preferably, solving the objective function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized includes:

[0178] Converting the objective function into a corresponding infinite-horizon MDP; wherein the infinite-horizon MDP includes: a state space, an action space, a transition probability, and a cost function;

[0179] The objective function is solved according to the state space, action space, transition probability and cost function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized.

[0180] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive effort.

[0181] Those skilled in the art will clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0182] Example 3

[0183] Accordingly, an embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the differential protection control method based on information freshness described in the above-mentioned embodiment of the invention.

[0184] The electronic device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The device may include, but is not limited to, a processor and a memory.

[0185] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the device and connects various parts of the entire device using various interfaces and lines.

[0186] Example 4

[0187] Accordingly, an embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the differential protection control method based on information freshness described in the above-mentioned embodiment of the invention.

[0188] The memory can be used to store the computer program, and the processor realizes various functions of the device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0189] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned various method embodiments. The computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0190] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A differential protection control method based on information freshness, characterized in that: include: receiving the system state measurement parameters collected by the sampler, and obtaining the sampling time of the system state measurement parameters, the reception time when the system state measurement parameters are received, and the communication delay of the system state measurement parameters during the transmission process; Calculating the information freshness corresponding to the system state measurement parameter based on the sampling time, the receiving time, and the communication delay; Taking the system state measurement parameters as variables and the information freshness as an invisible adjustment factor of the system long-term average cost, an objective function of system differential protection is constructed based on the information freshness and the system state measurement parameters with the goal of minimizing the system long-term average cost; The objective function is solved to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized, and the control strategy is mapped to a corresponding differential protection control action, and then differential protection control of the power system is performed according to the differential protection control action.

2. The differential protection control method based on information freshness according to claim 1, characterized in that: The freshness of the information is calculated using the following formula: Δ(t)=t-S i ,D i ≤t≤D i+1 ; S i+1 ≥D i ; D i =S i +Y i ; Among them, Δ(t) is the freshness of information at time t, S i is the sampling time of the i-th system state measurement parameter group, S i+1 is the sampling time of the i+1th system state measurement parameter group, D i is the receiving time when the i-th system status measurement parameter group is received, D i+1 Y is the receiving time when the i+1th system status measurement parameter group is received, i The communication delay of the parameter packet during transmission is measured for the i-th system state.

3. The differential protection control method based on information freshness according to claim 1, characterized in that: The objective function of the system differential protection is constructed based on the information freshness and the system state measurement parameters with the goal of minimizing the long-term average cost of the system, including: According to the information freshness and the system state measurement parameters, a first objective function of the system differential protection is constructed with the goal of minimizing the long-term average cost of the system; Taking a tuple consisting of a system state measurement parameter, a communication delay during transmission of the system state measurement parameter, and a differential protection control action generated upon receiving a previous system state measurement parameter as a corresponding sufficient statistic, and then optimizing the first objective function based on the sufficient statistic to transform it into a corresponding second objective function; The long-term average cost solution in the second objective function is converted into the ratio of the average cost in each period to the period length to obtain the corresponding third objective function; A preset cost parameter is added to the third objective function, the third objective function is converted into a corresponding fourth objective function, and then the fourth objective function is used as the objective function of the system differential protection.

4. The differential protection control method based on information freshness according to claim 1, characterized in that: By solving the objective function, the control strategy of the system differential protection is obtained when the long-term average cost of the system is minimized, including: Converting the objective function into a corresponding infinite-horizon MDP; wherein the infinite-horizon MDP includes: a state space, an action space, a transition probability, and a cost function; The objective function is solved according to the state space, action space, transition probability and cost function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized.

5. A differential protection control device based on information freshness, characterized in that: include: System state measurement parameter acquisition module, information freshness calculation module, objective function construction module and differential protection control module; The system state measurement parameter acquisition module is used to receive the system state measurement parameter collected by the sampler, and obtain the sampling time of the system state measurement parameter, the reception time when the system state measurement parameter is received, and the communication delay of the system state measurement parameter during the transmission process; The information freshness calculation module is used to calculate the information freshness corresponding to the system state measurement parameter based on the sampling time, receiving time and communication delay; The objective function construction module is configured to use the system state measurement parameter as a variable and the information freshness as an invisible adjustment factor of the system long-term average cost, and to construct an objective function of the system differential protection based on the information freshness and the system state measurement parameter with the goal of minimizing the system long-term average cost; The differential protection control module is used to solve the objective function to obtain the control strategy of the system differential protection when the long-term average cost of the system is minimized, and map the control strategy into a corresponding differential protection control action, and then perform differential protection control on the power system according to the differential protection control action.

6. The differential protection control device based on information freshness according to claim 5, characterized in that: The freshness of the information is calculated using the following formula: Δ(t)=t-S i ,D i ≤t≤D i+1 ; S i+1 ≥D i ; D i =S i +Y i ; Among them, Δ(t) is the freshness of information at time t, S i is the sampling time of the i-th system state measurement parameter group, S i+1 is the sampling time of the i+1th system state measurement parameter group, D i is the receiving time when the i-th system status measurement parameter group is received, D i+1 Y is the receiving time when the i+1th system status measurement parameter group is received, i The communication delay of the parameter packet during transmission is measured for the i-th system state.

7. The differential protection control device based on information freshness according to claim 5, characterized in that: The objective function of the system differential protection is constructed based on the information freshness and the system state measurement parameters with the goal of minimizing the long-term average cost of the system, including: According to the information freshness and the system state measurement parameters, a first objective function of the system differential protection is constructed with the goal of minimizing the long-term average cost of the system; Taking a tuple consisting of a system state measurement parameter, a communication delay during transmission of the system state measurement parameter, and a differential protection control action generated upon receiving a previous system state measurement parameter as a corresponding sufficient statistic, and then optimizing the first objective function based on the sufficient statistic to transform it into a corresponding second objective function; The long-term average cost solution in the second objective function is converted into the ratio of the average cost in each period to the period length to obtain the corresponding third objective function; A preset cost parameter is added to the third objective function, the third objective function is converted into a corresponding fourth objective function, and then the fourth objective function is used as the objective function of the system differential protection.

8. The differential protection control device based on information freshness according to claim 5, characterized in that: By solving the objective function, the control strategy of the system differential protection is obtained when the long-term average cost of the system is minimized, including: Converting the objective function into a corresponding infinite-horizon MDP; wherein the infinite-horizon MDP includes: a state space, an action space, a transition probability, and a cost function; The objective function is solved according to the state space, action space, transition probability and cost function to obtain a control strategy for system differential protection when the long-term average cost of the system is minimized.

9. An electronic device, characterized in that: The invention comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the differential protection control method based on information freshness according to any one of claims 1 to 4 is implemented.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the differential protection control method based on information freshness according to any one of claims 1 to 4.

Images