Substation communication disaster recovery switching method and system based on satellite Internet

Abstract

The embodiment of the present application provides a satellite Internet-based substation communication disaster recovery switching method and system, which relates to the field of power system communication, obtains the communication demand and historical transmission data of the substation, determines the communication data priority of the substation according to the communication demand and the historical transmission data, collects the communication network status data of the substation, determines the network status of the substation according to the communication network status data, and performs necessary disaster-time switching based on weighted rank sum ratio or constant preferential switching based on DQN on the communication network of the substation according to the communication data priority and the network status, thereby improving the stability and reliability of the communication network, achieving accurate control and timely response to the network status, ensuring reasonable switching under different network conditions, enhancing the adaptability and robustness of the substation communication system, improving the communication disaster recovery capability of the substation, and ensuring the safe operation and stability of the power system.

Description

Technical Field

[0001] The present invention relates to the field of power system communication technology, and in particular to a satellite Internet-based substation communication disaster recovery switching method and system. Background Art

[0002] In recent years, with the proposal of dual carbon goals and the continuous deepening of smart substation construction, substation business has become more and more complex. When the station is affected by a disaster, the communication network is easily damaged, which will affect the overall stability of the power grid system. Therefore, there are newer and higher requirements for substation network communication. Traditional substation communication systems mainly rely on ground networks, such as optical fiber communication and microwave communication. Such communication methods are easily damaged in the face of natural disasters, emergencies and extreme weather conditions, resulting in communication interruption of the substation, which in turn affects the normal operation of the power system.

[0003] With the rapid development of satellite communication technology, satellite Internet has gradually become an important means of communication. Compared with traditional ground networks, satellite Internet communication has higher disaster resistance and coverage. At present, the satellite-ground fusion network mainly switches the power station network during the high-speed movement of mobile terminals. Once the substation is affected by a disaster, it not only requires emergency maintenance, but also affects the stability of the power grid. The disaster recovery communication of the substation requires higher and more stable communication quality requirements than mobile terminals. Therefore, there is an urgent need for a communication method that can ensure smooth communication of substations more quickly during disasters and give priority to the transmission of key data. Summary of the invention

[0004] In view of this, the purpose of the present invention is to overcome the deficiencies in the prior art and to provide a substation communication disaster recovery switching method and system based on satellite Internet.

[0005] The present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present disclosure provides a method for disaster recovery switching of substation communication based on satellite Internet, the method comprising:

[0007] Acquire the communication demand and historical transmission data of the substation, and determine the communication data priority of the substation according to the communication demand and the historical transmission data;

[0008] Collecting communication network status data of the substation, and determining the network status of the substation according to the communication network status data;

[0009] According to the communication data priority and the network status, the communication network of the substation is switched in times of disaster based on a weighted rank sum ratio or switched in a regular preferential manner based on a DQN.

[0010] According to a specific implementation method disclosed in the present application, the step of obtaining the communication demand and historical transmission data of the substation, and determining the communication data priority of the substation according to the communication demand and the historical transmission data includes:

[0011] Dividing the communication data into broadband service data, narrowband service data and short message service data according to the communication requirements of the substation;

[0012] Acquire the transmission preference of the broadband service data, narrowband service data and short message service data in the station according to the historical transmission data of the substation;

[0013] Determine the normal station priority and the disaster station priority according to the communication demand of the substation and the transmission preference;

[0014] The communication data priority of the substation is determined according to the normal station priority, the disaster station priority and the historical transmission data.

[0015] According to a specific implementation method disclosed in the present application, the step of determining the communication data priority of the substation according to the normal station priority, the disaster station priority and the historical transmission data includes:

[0016] adjusting the communication data priority of the substation according to the historical transmission data;

[0017] The adjusted communication data priority is determined according to the normal station priority and the disaster station priority.

[0018] According to a specific implementation manner disclosed in the present application, the step of collecting the communication network status data of the substation and determining the network status of the substation according to the communication network status data includes:

[0019] Preprocessing the communication network status data of the substation to obtain a waveform data set and a link data set;

[0020] Using a butterfly optimization algorithm to perform network-aware feature extraction processing on the waveform data set and the link data set to obtain a network status data feature set;

[0021] The communication network state perception model is constructed using the square sum of deviation method;

[0022] The network status data feature set is input into the communication network status perception model, and the network status of the substation is output, wherein the network status includes a stable state, a fluctuating state and a disconnected state.

[0023] According to a specific implementation method disclosed in the present application, the step of performing network-aware feature extraction processing on the waveform data set and the link data set using a butterfly optimization algorithm to obtain a network status data feature set includes:

[0024] The basic model of the butterfly optimization algorithm is:

[0025] ,in, For sensory modality, is the stimulus intensity, is a power index, satisfying ∈[0,1];

[0026] The comprehensive position iteration of the butterfly optimization algorithm is:

[0027] ,in, is the current position information of the butterfly, r1 is the global search factor, satisfying r1∈[0,1], For the The smell of a butterfly, r2 is the local search factor, satisfying r2∈[0,1], and For the Different objects are randomly moved and selected in the local search, and p is the actual switching probability randomly generated, satisfying p∈[0,1].

[0028] According to a specific implementation method disclosed in the present application, the step of constructing a communication network state perception model using the square sum of deviations method includes:

[0029] The eigenvalues ​​are standardized and the The first weight matrix is ​​obtained by weighting the features. :

[0030] ;

[0031] Select n groups of network data and construct a non-negative matrix:

[0032] ;

[0033] The second weight matrix is ​​obtained by entropy weight method, so that :

[0034] ;

[0035] The combined decision scheme of the first weight matrix and the second weight matrix is ​​formed into a matrix, where the i-th decision scheme can be expressed as , let the comprehensive weight vector matrix be:

[0036] ;

[0037] Among them, α1 and α2 are the first weight combination coefficient and the second weight combination coefficient, and satisfy , weight matrix W={W1,W2},w 3,j To evaluate z i,j The weights of the weight combination coefficient matrix A={α1,α2} T ,satisfy:

[0038] ;

[0039] The comprehensive evaluation of scheme i is performed through linear weighting:

[0040] ;

[0041] Plan S i The sum of squares of deviations from other options can be expressed as:

[0042] ;

[0043] Construct the m-order symmetric matrix G as:

[0044] ;

[0045] Then the objective function of the communication network state perception model can be expressed as:

[0046] ;

[0047] Find the largest eigenvalue of matrix G and its corresponding unit eigenvector , calculate the weight set and feature set of combined weighting;

[0048] The weight set and feature set of the combined weighting are:

[0049] ;

[0050] ;

[0051] Build a communication network status awareness model:

[0052] ,in, The optimal set of weights for the combination;

[0053] According to the collection cycle, the communication network status is sensed and the preset threshold is used. , The communication network perception status of the substation is divided, wherein: :

[0054] , where NG is 1, indicating that the communication network perception state of the substation is stable; NG is 2, indicating that the communication network perception state of the substation is fluctuating; and NG is 3, indicating that the communication network perception state of the substation is disconnected.

[0055] According to a specific implementation method disclosed in the present application, the step of performing necessary disaster switching based on a weighted rank sum ratio or regular preferential switching based on a DQN on the communication network of the substation according to the communication data priority and the network status includes:

[0056] When the network state is stable or fluctuating, the communication network of the substation is switched to the optimal communication link based on DQN, and communication data is transmitted according to the communication data priority;

[0057] When the network state is disconnected, the communication network of the substation is switched to a candidate communication link when necessary according to the weighted rank sum ratio, and communication data is transmitted according to the communication data priority.

[0058] According to a specific implementation method disclosed in the present application, the step of performing DQN-based constant preferential switching of the communication network of the substation to the optimal communication link includes:

[0059] The link characteristics obtained by the butterfly optimization algorithm include received signal strength, available bit rate, transmission delay, network cost and network switching rate, and their normalized values ​​are , , , and , set the binary function for:

[0060] ;

[0061] The utility functions of the link features can be expressed as:

[0062] ;

[0063] in, , , are the upper or lower thresholds of the received signal strength, available bit rate, and network switching rate, respectively. The total utility function can be expressed as:

[0064] ;

[0065] in, , , , , is the weight of each feature;

[0066] Set the first ground link, the second ground link, the satellite Internet link, the high-orbit satellite link, and the BeiDou link to correspond to 1, 2, 3, 4, and 5 respectively, and the state space MDP dimension is 5:

[0067] ;

[0068] There are 5 choices in the action space, expressed as: , then action a∈A;

[0069] Assuming the time slot is t, the reward function for performing action a in the kth time slot can be expressed as:

[0070] ;

[0071] in, is a binary function, the value in the brackets is 1 if it is true and 0 if it is false. is the switching factor, ;

[0072] use - The greedy strategy obtains the state and reward value of the next time slot, and stores the state and reward value in the experience pool;

[0073] Set the objective function and loss function to be:

[0074] ;

[0075] ;

[0076] in, is the discount rate, Q is the deep Q function, W is the neural network parameter, and it is updated by gradient descent with a fixed learning rate:

[0077] ;

[0078] ;

[0079] Among them, α is a fixed learning rate, which is a constant value.

[0080] According to a specific implementation method disclosed in the present application, the step of performing a weighted rank sum ratio disaster necessary switching of the communication network of the substation to a candidate communication link includes:

[0081] Obtaining the station through Shannon formula evolution Access the network The data transfer rate is:

[0082] ;

[0083] in, Indicates station Access the network The bandwidth available, Representation Network The signal-to-noise ratio of , then the matching degree of different links for various services is defined as:

[0084] ;

[0085] in, Indicates the data service type By station Access the network The data transfer rate, =1,2,3 represent broadband, narrowband, and short message services respectively. and Respectively represent data service types By station Access the network The minimum and maximum transmission rates required;

[0086] The network parameters are obtained according to the transmission preference of the station. Assume that the station has M candidate networks and N characteristic indicators, where N = 6. The parameters are encoded into an original data matrix with M rows and N columns, recorded as

[0087] The original data matrix is ​​rank processed to obtain a ranked data matrix:

[0088] ;

[0089] By using the entropy weight method, we can get The weight of the indicator is , so that , then the alternative network Weighted rank sum ratio It can be expressed as:

[0090] ;

[0091] Will Sort by size to get the corresponding rank value , and calculate the cumulative frequency :

[0092] ;

[0093] The probability unit It can be expressed as the standard normal deviation function:

[0094] ,in, Expressed as a standard normal distribution function;

[0095] Weighted Rank Sum Ratio About Probability Units The linear fitting of , the first-order regression function is: ,in, are the linear function constant values ​​obtained by fitting.

[0096] In a second aspect, the present disclosure provides a satellite Internet-based substation communication disaster recovery switching system, the system comprising:

[0097] A communication priority determination module, used to obtain the communication demand and historical transmission data of the substation, and determine the communication data priority of the substation according to the communication demand and the historical transmission data;

[0098] A network status determination module, used to collect communication network status data of the substation, and determine the network status of the substation according to the communication network status data;

[0099] The communication network switching module is used to perform necessary disaster switching based on weighted rank sum ratio and normal preferential switching based on DQN on the communication network of the substation according to the communication data priority and the network status.

[0100] The above-mentioned satellite Internet-based substation communication disaster recovery switching method provided in the present application obtains the communication demand and historical transmission data of the substation, determines the communication data priority of the substation according to the communication demand and the historical transmission data, collects the communication network status data of the substation, determines the network status of the substation according to the communication network status data, and performs necessary disaster-time switching based on the weighted rank sum ratio or constant preferential switching based on DQN on the communication network of the substation according to the communication data priority and the network status, thereby improving the stability and reliability of the communication network, achieving precise control and timely response to the network status, ensuring reasonable switching under different network conditions, enhancing the adaptability and robustness of the substation communication system, improving the communication disaster recovery capability of the substation, and ensuring the safe operation and stability of the power system.

[0101] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and understandable, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work. In each of the drawings, similar components are numbered similarly.

[0103] Figure 1 A flowchart of a satellite Internet-based substation communication disaster recovery switching method provided in an embodiment of the present application is shown;

[0104] Figure 2 The embodiment of the present application shows the adjustment of data priority according to the frequency of abnormal data occurrence, wherein (a) is to adjust the data priority according to the average frequency of abnormal data occurrence, and (b) is to adjust the data priority according to the change of the average frequency of abnormal data occurrence;

[0105] Figure 3 The satellite-ground fusion communication network architecture of the substation provided by the embodiment of the present application is shown;

[0106] Figure 4 The communication network switching result provided by the embodiment of the present application is shown, wherein (a) is the network classification obtained through the network status, (b) is the network classification obtained through network perception, and (c) is the specific communication link path finally selected by each data;

[0107] Figure 5 A schematic structural diagram of a satellite Internet-based substation communication disaster recovery switching system provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0108] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0109] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. In contrast, when an element is referred to as being "directly on" another element, there is no intermediate element. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0110] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0111] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0112] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of the template are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0113] Example 1

[0114] like Figure 1 As shown, it is a flow chart of a method for disaster recovery switching of substation communication based on satellite Internet provided in an embodiment of the present application. The method for disaster recovery switching of substation communication based on satellite Internet provided in an embodiment of the present application comprises the following steps:

[0115] Step S101, obtaining the communication demand and historical transmission data of the substation, and determining the communication data priority of the substation according to the communication demand and the historical transmission data.

[0116] Specifically, the communication demand refers to the communication requirements required for the substation to communicate normally during a disaster or at ordinary times, and the historical transmission data refers to the data generated by the substation during the historical transmission process. By determining the priority of the communication data, the communication data with high priority represents more critical information in a disaster, and will be transmitted first, while the communication data with low priority will be transmitted with a delay, so as to ensure that the communication link is damaged when the substation is affected by the environment or the link is blocked due to increased communication needs of personnel, and it can still operate stably, ensuring the efficiency of the communication sequence.

[0117] The step of obtaining the communication demand and historical transmission data of the substation, and determining the communication data priority of the substation according to the communication demand and the historical transmission data comprises:

[0118] Dividing the communication data into broadband service data, narrowband service data and short message service data according to the communication requirements of the substation;

[0119] Acquire the transmission preference of the broadband service data, narrowband service data and short message service data in the station according to the historical transmission data of the substation;

[0120] Determine the normal station priority and the disaster station priority according to the communication demand of the substation and the transmission preference;

[0121] The communication data priority of the substation is determined according to the normal station priority, the disaster station priority and the historical transmission data.

[0122] Specifically, communication data includes broadband service data, narrowband service data, and short message service data. Broadband service data requires low network latency and smooth transmission, narrowband service data has the characteristics of low cost and low power consumption, and short message service data can stably transmit emergency information in times of disaster. Normal communication refers to communication under normal circumstances, which needs to ensure high reliability, real-time, security, and efficiency. The primary guarantee for communication in times of disaster is the smooth transmission of key information.

[0123] Specifically, among all the affected substations, some are in key locations for disaster relief and rescue, which can facilitate disaster relief or reduce the impact of disasters. Therefore, the data collected by them has a higher priority than other stations. The flow, duration, number of times and success rate of various business data transmitted by each station in a cycle are obtained, and the data are normalized to establish the network congestion perception of each station for different services. The three types of services are weighted to obtain the number of station periodic transmissions, so as to reflect the transmission preference of the station for different business data. According to the transmission preference of the station, the threshold of the station in normal and disaster times is divided into key stations, important stations and general stations.

[0124] The step of determining the communication data priority of the substation according to the normal station priority, the disaster station priority and the historical transmission data comprises:

[0125] adjusting the communication data priority of the substation according to the historical transmission data;

[0126] The adjusted communication data priority is determined according to the normal station priority and the disaster station priority.

[0127] For specific implementation, see Figure 2 , Figure 2 The embodiment of the present application provides for adjusting the data priority according to the frequency of abnormal data occurrence. The historical communication data transmitted by the substation contains abnormal data due to the actual environment or network environment. Such abnormal data can indicate that the substation is facing a disaster or emergency. The deviation standardization is obtained by statistically analyzing the flow, duration, number of times, and success rate of each substation transmission. , , , , establish the perception of network congestion caused by the substation to the three data types within a cycle , then the congestion perception of broadband, narrowband and short message types can be obtained respectively , , , weighted to represent the transmission volume of the substation in a period of t ,in , , Respectively represent the weights of broadband, narrowband, and short message data types, which are determined by the network status. The larger the value, the greater the transmission volume of the station in the unit cycle, that is, the richer the content, the more important the station is. Set the threshold value , , prioritize stations:

[0128] ;

[0129] On the basis of the station priority, the data priority is adjusted according to the average frequency of abnormal data and the average frequency change rate. If the frequency of abnormal data exceeds the normal range or is within the normal range but the increase rate of abnormal data is too high and exceeds the normal range, the data priority of the substation is increased; conversely, if the frequency of abnormal data is lower than the normal range or is within the normal range but the decrease rate is lower than the normal range, the data priority of the substation is lowered.

[0130] ;

[0131] Data priority is expressed as: .

[0132] Step S102: collecting communication network status data of the substation, and determining the network status of the substation according to the communication network status data.

[0133] Specifically, the network status of the substation is an important basis for determining whether the communication network needs to be switched, and the communication network status of the substation is determined based on the communication network status data.

[0134] The step of collecting the communication network status data of the substation and determining the network status of the substation according to the communication network status data comprises:

[0135] Preprocessing the communication network status data of the substation to obtain a waveform data set and a link data set;

[0136] Using a butterfly optimization algorithm to perform network-aware feature extraction processing on the waveform data set and the link data set to obtain a network status data feature set;

[0137] The communication network state perception model is constructed using the square sum of deviation method;

[0138] The network status data feature set is input into the communication network status perception model, and the network status of the substation is output, wherein the network status includes a stable state, a fluctuating state and a disconnected state.

[0139] It is understandable that the communication network status is greatly affected by the collected communication network data, so the communication network data is cleaned, including abnormal perception processing, abnormal correction processing, missing supplement processing and data normalization processing, and the communication network data after data cleaning is classified into waveform data set and link data set. The waveform data set is the valid data extracted from the communication network waveform, and the link data set is the communication link data obtained by the sensor. By performing data cleaning on the communication network data, the data accuracy and the accuracy of identifying the communication network status of the substation are improved.

[0140] The step of performing network-aware feature extraction processing on the waveform data set and the link data set using the butterfly optimization algorithm to obtain a network status data feature set includes:

[0141] The basic model of the butterfly optimization algorithm is:

[0142] ,in, For sensory modality, is the stimulus intensity, is a power index, satisfying ∈[0,1];

[0143] The comprehensive position iteration of the butterfly optimization algorithm is:

[0144] ,in, is the current position information of the butterfly, r1 is the global search factor, satisfying r1∈[0,1], For the The smell of a butterfly, r2 is the local search factor, satisfying r2∈[0,1]; and For the Different objects are randomly moved and selected in the local search, and p is the actual switching probability randomly generated, satisfying p∈[0,1].

[0145] Understandably, since the data that affects the state of the substation communication network is large and complex, the butterfly optimization algorithm is used to perform feature screening to improve the efficiency of feature determination and feature extraction. When the iteration limit is reached, the screened individuals can be obtained.

[0146] The step of constructing a communication network state perception model using the square sum of deviations method includes:

[0147] The eigenvalues ​​are standardized and the The first weight matrix is ​​obtained by weighting the features. :

[0148] ;

[0149] Select n groups of network data and construct a non-negative matrix:

[0150] ;

[0151] The second weight matrix is ​​obtained by entropy weight method, so that :

[0152] ;

[0153] The combined decision scheme of the first weight matrix and the second weight matrix is ​​formed into a matrix, where the i-th decision scheme can be expressed as , let the comprehensive weight vector matrix be:

[0154] ;

[0155] Among them, α1 and α2 are the first weight combination coefficient and the second weight combination coefficient, and satisfy , weight matrix W={W1,W2},w 3,j To evaluate z i,j The weights of the weight combination coefficient matrix A={α1,α2} T ,satisfy:

[0156] ;

[0157] The comprehensive evaluation of scheme i is performed through linear weighting:

[0158] ;

[0159] Plan S iThe sum of squares of deviations from other options can be expressed as:

[0160] ;

[0161] Construct the m-order symmetric matrix G as:

[0162] ;

[0163] Then the objective function of the communication network state perception model can be expressed as:

[0164] ;

[0165] Find the largest eigenvalue of matrix G and its corresponding unit eigenvector , calculate the weight set and feature set of combined weighting;

[0166] The weight set and feature set of the combined weighting are:

[0167] ;

[0168] ;

[0169] Build a communication network status awareness model:

[0170] ,in, The optimal set of weights for the combination;

[0171] According to the collection cycle, the communication network status is sensed and the preset threshold is used. , The communication network perception status of the substation is divided into: :

[0172] , where NG is 1, indicating that the communication network perception state of the substation is stable; NG is 2, indicating that the communication network perception state of the substation is fluctuating; and NG is 3, indicating that the communication network perception state of the substation is disconnected.

[0173] Specifically, the larger the sum of squares of the comprehensive evaluation values ​​of all schemes, the more dispersed the comprehensive evaluation of the decision schemes is, and the more accurate the comprehensive weights are. Constructing an m-order symmetric matrix Based on the extracted features, an m-order symmetric matrix is ​​constructed, where m is the number of features, including bandwidth utilization, packet loss rate, delay and jitter. The maximum eigenvalue root of the matrix and its corresponding unit eigenvector are calculated using eigenvalue decomposition. The components of the unit eigenvector are used as the weights of the features, reflecting the importance of each feature in determining the network status. According to the determined weights and eigenvalues, the comprehensive evaluation index is calculated, and the mapping relationship between the communication network status and the comprehensive evaluation index is established using the sum of squares method. Network perception is performed according to the collection cycle, and multiple groups of data are collected for analysis. The preset threshold for network status division is determined according to the analysis results, and the network status is divided into three states: stable, fluctuating and disconnected according to the preset threshold for division.

[0174] Preferably, the communication network status of the substation can be intuitively displayed through indicator lights and charts, and network status reports can be regularly generated for reference by substation staff to improve the working efficiency of the substation system.

[0175] Step S103, performing necessary disaster switching based on weighted rank sum ratio or regular preferential switching based on DQN on the communication network of the substation according to the communication data priority and the network status.

[0176] Specifically, the communication of substations involves functions such as information collection, data transmission, monitoring and dispatching. Normal communication is when the network status is stable or fluctuating, based on DQN (Deep Q Network) to select links with better transmission conditions to improve the operation efficiency and management level of the station. Disaster communication is when the network status is completely disconnected, and it is necessary to transmit information and take necessary measures in an emergency in a timely manner to minimize losses and harm.

[0177] The step of performing necessary disaster switching based on a weighted rank sum ratio or regular preferential switching based on a DQN on the communication network of the substation according to the communication data priority and the network status comprises:

[0178] When the network state is stable or fluctuating, the communication network of the substation is switched to the optimal communication link based on DQN, and communication data is transmitted according to the communication data priority;

[0179] When the network state is disconnected, the communication network of the substation is switched to a candidate communication link when necessary according to the weighted rank sum ratio, and communication data is transmitted according to the communication data priority.

[0180] For specific implementation, see Figure 3 , Figure 3 The satellite-ground fusion communication network architecture of the substation provided in the embodiment of the present application is to set up multiple substations within a range, each of which has the function of data transmission, determine the station priority according to the historical transmission data, divide the data priority in the event of a disaster, and select a more suitable network for data transmission according to the network status. The data information collected or sensed by each substation is transmitted back to the target control center respectively. The data transmission methods include ground links and satellite links. The ground links are connected through ground base stations. The satellite links select different satellite systems for transmission according to the business type of the data. The satellite systems include satellite Internet, high-orbit satellites and Beidou system.

[0181] The step of performing DQN-based constant preferential switching of the communication network of the substation to the optimal communication link comprises:

[0182] The link characteristics obtained by the butterfly optimization algorithm include received signal strength, available bit rate, transmission delay, network cost and network switching rate, and their normalized values ​​are , , , and , set the binary function for:

[0183] ;

[0184] The utility functions of the link features can be expressed as:

[0185] ;

[0186] in, , , are the upper or lower thresholds of the received signal strength, available bit rate, and network switching rate, respectively. The total utility function can be expressed as:

[0187] ;

[0188] in, , , , , is the weight of each feature;

[0189] Set the first ground link, the second ground link, the satellite Internet link, the high-orbit satellite link, and the BeiDou link to correspond to 1, 2, 3, 4, and 5 respectively, and the state space MDP dimension is 5:

[0190] ;

[0191] There are 5 choices in the action space, expressed as: , then action a∈A;

[0192] Assuming the time slot is t, the reward function for performing action a in the kth time slot can be expressed as:

[0193] ;

[0194] in, is a binary function, the value in the brackets is 1 if it is true and 0 if it is false. is the switching factor, ;

[0195] use - The greedy strategy obtains the state and reward value of the next time slot, and stores the state and reward value in the experience pool;

[0196] Set the objective function and loss function to be:

[0197] ;

[0198] ;

[0199] in, is the discount rate, Q is the deep Q function, W is the neural network parameter, and it is updated by gradient descent with a fixed learning rate:

[0200] ;

[0201] ;

[0202] Among them, α is a fixed learning rate, which is a constant value.

[0203] Specifically, the first ground link refers to a link that is closer to the base station, and the second ground link refers to a link that is farther from the base station. The features that have a key impact on the switching decision, namely the network status features, are selected from the numerous link features, and the total utility function is constructed based on the features. Define a state space, including network status and link quality, etc. The action space is the possible switching actions, namely the selection process of the transmission link. The switching factor is used to balance the benefits and costs brought by the switching to avoid frequent ping-pong switching. After multiple iterative training, DQN can learn to select the optimal communication link under different states.

[0204] The step of performing a weighted rank sum ratio on the communication network of the substation and switching it to a candidate communication link when necessary comprises:

[0205] Obtaining the station through Shannon formula evolution Access the network The data transfer rate is:

[0206] ;

[0207] in, Indicates station Access the network The bandwidth available, Representation Network The signal-to-noise ratio of , then the matching degree of different links for various services is defined as:

[0208] ;

[0209] in, Indicates the data service type By station Access the network The data transfer rate, =1,2,3 represent broadband, narrowband, and short message services respectively. and Respectively represent data service types By station Access the network The minimum and maximum transmission rates required;

[0210] The network parameters are obtained according to the transmission preference of the station. Assume that the station has M candidate networks and N characteristic indicators, where N = 6. The parameters are encoded into an original data matrix with M rows and N columns, recorded as

[0211] The original data matrix is ​​rank processed to obtain a ranked data matrix:

[0212] ;

[0213] By using the entropy weight method, we can get The weight of the indicator is , so that , then the alternative network Weighted rank sum ratio It can be expressed as:

[0214] ;

[0215] Will Sort by size to get the corresponding rank value , and calculate the cumulative frequency :

[0216] ;

[0217] The probability unit It can be expressed as the standard normal deviation function:

[0218] ,in, Expressed as a standard normal distribution function;

[0219] Weighted Rank Sum Ratio About Probability Units The linear fitting of , the first-order regression function is: ,in, are the linear function constant values ​​obtained by fitting.

[0220] Specifically, the network parameters obtained according to the transmission preference of the station include service matching degree, received signal strength, available bit rate, signal-to-noise ratio, signal transmission time and transmission cost. The network parameters are calculated according to the data type of the transmission preference of the station. The network parameters include benefit parameters and cost parameters. Substitute Get the fitted , sort by value from large to small, the largest The corresponding candidate network is the most suitable network for emergency communications during disasters.

[0221] Preferably, the communication network switching result is as follows: Figure 4 , Figure 4 The communication network switching result provided in the embodiment of the present application, (a) Ⅰ indicates that the network state is disconnected, II indicates that the network state is fluctuating, and III indicates that the network state is stable. In the vertical axis, 1 indicates broadband signal, 2 indicates narrowband signal, 3 indicates short message signal, and the horizontal axis indicates time. The data signal in the disaster period needs to be adjusted according to the data priority method. Light gray indicates that the priority increases, dark gray indicates that the priority remains unchanged, and black indicates that the priority decreases. (b) represents the network classification obtained through network perception. In the vertical axis, 1 indicates that the network is normal, 2 indicates that the network fluctuates, and 3 indicates that the network is disconnected. The horizontal axis indicates time. (c) represents the specific communication link path finally selected for each data. 1 represents the communication link where the first ground base station is farther away, 2 represents the communication link where the second ground base station is closer, 3 represents the satellite Internet link, 4 represents the high-orbit satellite link, and 5 represents the Beidou system link. In the event of a disaster, the priority of the data signal is adjusted according to the data priority to ensure the priority transmission of key data. The network classification results obtained through network perception are used to determine the network status in real time. According to the weighted rank sum ratio, it is necessary to switch to the candidate communication link in the event of a disaster, thereby realizing flexible switching of data between different links, ensuring the continuity and reliability of data transmission, improving the communication security of the power grid system, and ensuring the operation efficiency of the power system.

[0222] The above-mentioned satellite Internet-based substation communication disaster recovery switching method provided in the present application obtains the communication demand and historical transmission data of the substation, determines the communication data priority of the substation according to the communication demand and the historical transmission data, collects the communication network status data of the substation, determines the network status of the substation according to the communication network status data, and performs necessary disaster switching based on a weighted rank sum ratio or constant preferential switching based on DQN on the communication network of the substation according to the communication data priority and the network status. In normal times, the ping-pong effect is reduced through constant preferential switching of DNQ. In disasters, the most suitable emergency communication network is quickly selected through necessary disaster switching based on a weighted rank sum ratio to ensure timely transmission of key data, improve the operating efficiency and safety of the site, reduce losses and hazards during disasters, improve the communication security of the power grid system, and ensure the operating efficiency of the power system.

[0223] Example 2

[0224] like Figure 5 As shown, it is a structural diagram of a substation communication disaster recovery switching system 500 based on satellite Internet provided in an embodiment of the present application, and the system includes:

[0225] The communication priority determination module 501 is used to obtain the communication demand and historical transmission data of the substation, and determine the communication data priority of the substation according to the communication demand and the historical transmission data;

[0226] A network status determination module 502, configured to collect communication network status data of the substation, and determine the network status of the substation according to the communication network status data;

[0227] The communication network switching module 503 is used to perform necessary disaster switching based on weighted rank sum ratio and normal preferential switching based on DQN on the communication network of the substation according to the communication data priority and the network status.

[0228] The satellite Internet-based substation communication disaster recovery switching system 500 provided in this embodiment can implement the satellite Internet-based substation communication disaster recovery switching method shown in Example 1, which will not be described again here to avoid repetition.

[0229] The satellite Internet-based substation communication disaster recovery switching system provided in the embodiment of the present application obtains the communication demand and historical transmission data of the substation, determines the communication data priority of the substation according to the communication demand and the historical transmission data, collects the communication network status data of the substation, determines the network status of the substation according to the communication network status data, and performs necessary disaster-time switching based on weighted rank sum ratio or constant preferential switching based on DQN on the communication network of the substation according to the communication data priority and the network status, thereby improving the stability and reliability of the communication network, achieving precise control and timely response to the network status, ensuring reasonable switching under different network conditions, enhancing the adaptability and robustness of the substation communication system, improving the communication disaster recovery capability of the substation, and ensuring the safe operation and stability of the power system.

[0230] In several embodiments provided in the present application, it should be understood that the disclosed system and method can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and structure diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in an alternative implementation, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the structure diagram and / or the flow diagram, and the combination of boxes in the structure diagram and / or the flow diagram, can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0231] In addition, the functional modules or units in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0232] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a smart phone, a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0233] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A satellite Internet-based substation communication disaster recovery switching method, characterized in that: The method comprises: Acquire the communication demand and historical transmission data of the substation, and determine the communication data priority of the substation according to the communication demand and the historical transmission data; Collecting communication network status data of the substation, and determining the network status of the substation according to the communication network status data; According to the communication data priority and the network status, the communication network of the substation is switched in times of disaster based on a weighted rank sum ratio or switched in a regular preferential manner based on a DQN.

2. The satellite Internet-based substation communication disaster recovery switching method according to claim 1 is characterized in that: The step of obtaining the communication demand and historical transmission data of the substation, and determining the communication data priority of the substation according to the communication demand and the historical transmission data comprises: Dividing the communication data into broadband service data, narrowband service data and short message service data according to the communication requirements of the substation; Acquire the transmission preference of the broadband service data, narrowband service data and short message service data in the station according to the historical transmission data of the substation; Determine the normal station priority and the disaster station priority according to the communication demand of the substation and the transmission preference; The communication data priority of the substation is determined according to the normal station priority, the disaster station priority and the historical transmission data.

3. The satellite Internet-based substation communication disaster recovery switching method according to claim 2 is characterized in that: The step of determining the communication data priority of the substation according to the normal station priority, the disaster station priority and the historical transmission data comprises: adjusting the communication data priority of the substation according to the historical transmission data; The adjusted communication data priority is determined according to the normal station priority and the disaster station priority.

4. The satellite Internet-based substation communication disaster recovery switching method according to claim 1, characterized in that: The step of collecting the communication network status data of the substation and determining the network status of the substation according to the communication network status data comprises: Preprocessing the communication network status data of the substation to obtain a waveform data set and a link data set; Using a butterfly optimization algorithm to perform network-aware feature extraction processing on the waveform data set and the link data set to obtain a network status data feature set; The communication network state perception model is constructed using the square sum of deviation method; The network status data feature set is input into the communication network status perception model, and the network status of the substation is output, wherein the network status includes a stable state, a fluctuating state and a disconnected state.

5. The satellite Internet-based substation communication disaster recovery switching method according to claim 4 is characterized in that: The step of performing network-aware feature extraction processing on the waveform data set and the link data set using the butterfly optimization algorithm to obtain a network status data feature set includes: The basic model of the butterfly optimization algorithm is: ,in, For sensory modality, is the stimulus intensity, is a power index, satisfying ∈[0,1]; The comprehensive position iteration of the butterfly optimization algorithm is: ,in, is the current position information of the butterfly, r1 is the global search factor, satisfying r1∈[0,1], For the The smell of a butterfly, r2 is the local search factor, satisfying r2∈[0,1], and For the Different objects are randomly moved and selected in the local search, and p is the actual switching probability randomly generated, satisfying p∈[0,1].

6. The satellite Internet-based substation communication disaster recovery switching method according to claim 4 is characterized in that: The step of constructing a communication network state perception model using the square sum of deviations method includes: The eigenvalues ​​are standardized and the The first weight matrix is ​​obtained by weighting the features. : ； Select n groups of network data and construct a non-negative matrix: ； The second weight matrix is ​​obtained by entropy weight method, so that : ； The combined decision scheme of the first weight matrix and the second weight matrix is ​​formed into a matrix, where the i-th decision scheme can be expressed as , let the comprehensive weight vector matrix be: ； Among them, α1 and α2 are the first weight combination coefficient and the second weight combination coefficient, and satisfy , weight matrix W={W1,W2},w 3,j To evaluate z i,j The weights of the weight combination coefficient matrix A={α1,α2} T ,satisfy: ； The comprehensive evaluation of scheme i is performed through linear weighting: ； Plan S i The sum of squares of deviations from other options can be expressed as: ； Construct the m-order symmetric matrix G as: ； Then the objective function of the communication network state perception model can be expressed as: ； Find the largest eigenvalue of matrix G and its corresponding unit eigenvector , calculate the weight set and feature set of combined weighting; The weight set and feature set of the combined weighting are: ； ； Build a communication network status awareness model: ,in, The optimal set of weights for the combination; According to the collection cycle, the communication network status is sensed and the preset threshold is used. , The communication network perception status of the substation is divided, wherein: : , where NG is 1, indicating that the communication network perception state of the substation is stable; NG is 2, indicating that the communication network perception state of the substation is fluctuating; and NG is 3, indicating that the communication network perception state of the substation is disconnected.

7. The satellite Internet-based substation communication disaster recovery switching method according to claim 1, characterized in that: The step of performing necessary disaster switching based on a weighted rank sum ratio or regular preferential switching based on a DQN on the communication network of the substation according to the communication data priority and the network status comprises: When the network state is stable or fluctuating, the communication network of the substation is switched to the optimal communication link based on DQN, and communication data is transmitted according to the communication data priority; When the network state is disconnected, the communication network of the substation is switched to a candidate communication link when necessary according to the weighted rank sum ratio, and communication data is transmitted according to the communication data priority.

8. The satellite Internet-based substation communication disaster recovery switching method according to claim 7, characterized in that: The step of performing DQN-based constant preferential switching of the communication network of the substation to the optimal communication link comprises: The link characteristics obtained by the butterfly optimization algorithm include received signal strength, available bit rate, transmission delay, network cost and network switching rate, and their normalized values ​​are , , , and , set the binary function for: ； The utility functions of the link features can be expressed as: ； in, , , are the upper or lower thresholds of the received signal strength, available bit rate, and network switching rate, respectively. The total utility function can be expressed as: ； in, , , , , is the weight of each feature; Set the first ground link, the second ground link, the satellite Internet link, the high-orbit satellite link, and the BeiDou link to correspond to 1, 2, 3, 4, and 5 respectively, and the state space MDP dimension is 5: ； There are 5 choices in the action space, expressed as: , then action a∈A; Assuming the time slot is t, the reward function for performing action a in the kth time slot can be expressed as: ； in, is a binary function, the value in the brackets is 1 if it is true and 0 if it is false. is the switching factor, ; use - The greedy strategy obtains the state and reward value of the next time slot, and stores the state and reward value in the experience pool; Set the objective function and loss function to be: ； ； in, is the discount rate, Q is the deep Q function, W is the neural network parameter, and it is updated by gradient descent with a fixed learning rate: ； ； Among them, α is a fixed learning rate, which is a constant value.

9. The satellite Internet-based substation communication disaster recovery switching method according to claim 7, characterized in that: The step of performing a weighted rank sum ratio on the communication network of the substation and switching it to a candidate communication link when necessary comprises: Obtaining the station through Shannon formula evolution Access the network The data transfer rate is: ； Among them, B uv Indicates station Access the network The bandwidth available, Representation Network The signal-to-noise ratio of , then the matching degree of different links for various services is defined as: ； in, Indicates the data service type By station Access the network The data transfer rate, =1,2,3 represent broadband, narrowband, and short message services respectively. and Respectively represent data service types By station Access the network The minimum and maximum transmission rates required; The network parameters are obtained according to the transmission preference of the station. Assume that the station has M candidate networks and N characteristic indicators, where N = 6. The parameters are encoded into an original data matrix with M rows and N columns, recorded as ; The original data matrix is ​​rank processed to obtain a ranked data matrix: ； By using the entropy weight method, we can get The weight of the indicator is , so that , then the alternative network Weighted rank sum ratio It can be expressed as: ； Will Sort by size to get the corresponding rank value , and calculate the cumulative frequency : ； The probability unit It can be expressed as the standard normal deviation function: ,in, Expressed as a standard normal distribution function; Weighted Rank Sum Ratio About Probability Units The linear fitting of , the first-order regression function is: ,in, are the linear function constant values ​​obtained by fitting.

10. A satellite Internet-based substation communication disaster recovery switching system, characterized in that: The system comprises: A communication priority determination module, used to obtain the communication demand and historical transmission data of the substation, and determine the communication data priority of the substation according to the communication demand and the historical transmission data; A network status determination module, used to collect communication network status data of the substation, and determine the network status of the substation according to the communication network status data; The communication network switching module is used to perform necessary disaster switching based on weighted rank sum ratio and normal preferential switching based on DQN on the communication network of the substation according to the communication data priority and the network status.

Images