Overcurrent protection setting method and system for rectifier transformer of power system

Abstract

The present invention discloses an overcurrent protection setting method and system for a rectifier transformer of an electric power system, and relates to the technical field of reinforcement learning models. By collecting a system topology diagram, power data is collected in real time for any rectifier transformer, and when an overcurrent phenomenon occurs, current time data is collected in real time; based on the power data, a current level grade is obtained, based on the current time data and the system topology diagram, a duration grade is obtained, for each rectifier transformer, an overcurrent score is obtained based on the current level grade and the duration grade, a state space, an action space and a reward value function of a protection decision model are constructed, a protection decision strategy is output by the protection decision model, and the electric power system performs overcurrent protection according to the protection decision strategy; accurate and efficient overcurrent protection of the rectifier transformer is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of reinforcement learning models, and in particular to an overcurrent protection setting method and system for a rectifier transformer of an electric power system. Background Art

[0002] The rectifier transformer in the power system is one of the key devices. Its main function is to convert AC power into DC power to supply power to the DC system. However, in actual operation, the rectifier transformer may have overcurrent, that is, the secondary current exceeds its rated value. Overcurrent can cause equipment overheating and insulation aging. In severe cases, it may cause permanent damage to the equipment, thus affecting the reliable operation of the entire power system.

[0003] Therefore, it is necessary to perform overcurrent protection on the rectifier transformer. Traditional overcurrent protection schemes are usually based on fixed overcurrent setting values, and the protection action is immediately triggered when the secondary current exceeds the setting value. This method has some shortcomings:

[0004] The differences in current levels and the impact of current duration are ignored. Different degrees of overcurrent have different degrees of harm to equipment, and different protection strategies should be adopted. Short-term overcurrent spikes have limited impact on equipment, while long-term overcurrent is more dangerous.

[0005] The Chinese patent with the authorization announcement number CN112785037B discloses a method for optimizing the setting value of the inverse time overcurrent protection of the distribution network, which includes the following steps: Step 1: Establishing the inverse time overcurrent protection model of the distribution network and its constraints; Step 2: Based on the improved gray wolf algorithm, the optimal setting value of the parameter to be set in the inverse time overcurrent protection model of the distribution network is calculated under the constraints. However, this method fails to take into account the influence of the overcurrent size and the overcurrent duration;

[0006] To this end, the present invention proposes an overcurrent protection setting method and system for a rectifier transformer of an electric power system. Summary of the invention

[0007] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an overcurrent protection setting method and system for a rectifier transformer of a power system, which realizes accurate and efficient overcurrent protection for the rectifier transformer.

[0008] To achieve the above object, a method for setting overcurrent protection of a rectifier transformer of a power system is proposed, comprising the following steps:

[0009] Step 1: Collect system topology diagram;

[0010] Step 2: For any rectifier transformer, collect power data in real time, and determine whether overcurrent occurs based on the power data; if overcurrent occurs, go to step 3, otherwise, repeat step 2;

[0011] Step 3: For each rectifier transformer, collect current time data in real time; obtain the current level grade based on the power data, and obtain the duration grade based on the current time data and the system topology diagram;

[0012] Step 4: For each rectifier transformer, based on the current level grade and the duration grade, obtain an overcurrent score from a preset two-dimensional score table;

[0013] Step 5: construct the state space, action space and reward value function of the protection decision model based on the overcurrent score, power data and duration level; the protection decision model outputs a protection decision strategy, and the power system performs overcurrent protection according to the protection decision strategy; the protection decision model is trained based on the current state and the next state after the overcurrent protection;

[0014] The collection system topology diagram includes the following steps:

[0015] Step 11: Obtain a power line schematic diagram of the entire power system;

[0016] Step 12: extracting power equipment and power line connection relationships between the power equipment from the power line schematic diagram to form a topological connection relationship;

[0017] Step 13: Obtain the capacity of each rectifier transformer and the capacity of each power line;

[0018] Step 14: Take each power device as a topological node, the edge between each two topological nodes with a connection relationship as a topological edge, the capacity of each power device as the attribute of the corresponding topological node, and the capacity of each power line as the attribute of the corresponding topological edge, and construct a system topology graph;

[0019] The method of collecting power data in real time is:

[0020] Collect the primary side current, secondary side current, voltage value, voltage fluctuation, current change rate and frequency of each rectifier transformer to form power data;

[0021] The method of judging whether an overcurrent phenomenon occurs based on power data is:

[0022] Collect the rated current value of each rectifier transformer;

[0023] An excess threshold H is preset. If the result of multiplying the rated current value by the excess threshold H is less than or equal to the secondary side current in the power data, it is determined that overcurrent has occurred. If the result of multiplying the rated current value by the excess threshold H is greater than the secondary side current in the power data, it is determined that no overcurrent has occurred. The excess threshold H is a constant greater than 1.

[0024] The method of collecting current time data in real time is:

[0025] The overcurrent duration of each rectifier transformer that has an overcurrent phenomenon is collected as current-time data;

[0026] The method of obtaining the current level based on the power data is as follows:

[0027] The current level is divided into N1 power levels, and a power level threshold is set for each power level; N1 is the number of preset power levels;

[0028] Based on the product of the power level threshold and the rated current, the current range of each power level is divided;

[0029] According to the magnitude of the secondary side current in the power data, the power level of each rectifier transformer is determined, which is expressed as a current level level;

[0030] The method of obtaining the duration level based on the current time data and the system topology diagram is as follows:

[0031] The number of the rectifier transformer is marked as i, and the capacity of the i-th rectifier transformer is marked as Ci;

[0032] Calculate the sum of the capacities of the power lines connected to the i-th rectifier transformer, and mark the sum of the capacities as CTi;

[0033] The capacity ratio CBi of the i-th rectifier transformer is calculated. The capacity ratio CBi is calculated as follows:

[0034] N2 duration levels are preset, and a duration level threshold is set for each duration level according to the standard endurance time, wherein the duration level threshold is a multiple of the standard endurance time; N2 is the number of preset duration levels;

[0035] Based on the duration level threshold, the duration range of each duration level is divided;

[0036] N3 ratio levels are preset for the capacity ratio, and a ratio threshold and a ratio coefficient are preset for each ratio level; according to the ratio threshold, the ratio range of each ratio level is divided; N3 is the number of preset ratio levels;

[0037] According to the capacity ratio CBi, the ratio level of the i-th rectifier transformer is determined as the capacity ratio level, and the ratio coefficient corresponding to the capacity ratio level is marked as ki;

[0038] Then the duration range of each duration level of the i-th rectifier transformer is reset to [tmin×(1+ki), tmax×(1+ki)]; where tmin is the lower limit of the duration range, and tmax is the upper limit of the duration range;

[0039] According to the current time data of each rectifier transformer and the duration range of each corresponding duration level, the duration level of each rectifier transformer is determined and expressed as a duration level;

[0040] The method of obtaining the overcurrent score from the preset two-dimensional score table based on the current level grade and the duration grade is as follows:

[0041] A two-dimensional scoring table is preset; each row of the two-dimensional scoring table represents a time level, each column represents a power level, and the element value corresponding to each row and column represents an overcurrent score;

[0042] According to the current level level and duration level of each rectifier transformer, find the overcurrent score corresponding to the power level and duration level from the two-dimensional score table;

[0043] The method of constructing the state space, action space and reward value function of the protection decision model based on the overcurrent score, power data and duration level is as follows:

[0044] The Actor-Critic model was selected as the protection decision model;

[0045] Design a state space S, wherein the state space S reflects the real-time operating state of the entire power system, and the state space S includes power data, current-time data, and overcurrent scores of all rectifier transformers;

[0046] Designing an action space A; the action space A is a system-level protection decision control strategy output by the Actor model, and the actions in the action space A include an adjustment value of an overcurrent setting parameter of each rectifier transformer;

[0047] Design a reward value function Q; the reward value function Q = r (s, a), which measures the effect of performing action a under the current state s;

[0048] The protection decision model is trained based on the current state and the next state after overcurrent protection as follows:

[0049] The power data, current time data and overcurrent score of each rectifier transformer in the power system are taken as the current state s; the current state s is input to the Actor model of the protection decision model, and the Actor model outputs an action in the action space A so that the state-behavior value output by the Critic model is maximized, and the output action is used as the protection decision control strategy;

[0050] Execute the protection decision control strategy to obtain the next state s' and the actual reward value;

[0051] The critic model outputs the state-behavior value after executing the protection decision control strategy in the current state, calculates the TD error δ of the critic model, updates the critic model parameters according to δ, and minimizes the critic loss function; the state-behavior value output by the critic model is obtained by calculating the state-behavior value function generated by the time difference optimization method;

[0052] The Actor model takes the state-behavior value output by the Critic model as the output target and uses the policy gradient update method to update the parameters of the Actor network model.

[0053] An overcurrent protection setting system for a rectifier transformer of a power system is proposed, comprising an overcurrent judgment module, a scoring calculation module and a protection decision module; wherein each module is electrically connected;

[0054] The overcurrent judgment module collects the system topology diagram and the power data of any rectifier transformer in real time, and judges whether an overcurrent phenomenon occurs based on the power data; if an overcurrent phenomenon occurs, the system topology diagram and power data are sent to the scoring calculation module and the protection decision module;

[0055] The scoring calculation module collects the current-time data in real time for each rectifier transformer; obtains the current level grade based on the power data, obtains the duration grade based on the current-time data and the system topology diagram, and obtains the overcurrent score from a preset two-dimensional scoring table for each rectifier transformer based on the current level grade and the duration grade; and sends the overcurrent score to the protection decision module;

[0056] A protection decision module constructs a state space, action space and reward value function of a protection decision model based on overcurrent scores, power data and duration levels; the protection decision model outputs a protection decision strategy, and the power system performs overcurrent protection according to the protection decision strategy; the protection decision model is trained based on the current state and the next state after overcurrent protection.

[0057] An electronic device is proposed, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0058] The processor executes the above-mentioned overcurrent protection setting method for the rectifier transformer of the power system by calling the computer program stored in the memory.

[0059] A computer-readable storage medium is provided, on which a rewritable computer program is stored;

[0060] When the computer program is executed on a computer device, the computer device is enabled to execute the above-mentioned overcurrent protection setting method for a rectifier transformer of a power system.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention collects power data in real time for any rectifier transformer by collecting the system topology diagram, and judges whether an overcurrent phenomenon occurs based on the power data; if an overcurrent phenomenon occurs, the current time data is collected in real time for each rectifier transformer; based on the power data, the current level grade is obtained, based on the current time data and the system topology diagram, the duration grade is obtained, for each rectifier transformer, based on the current level grade and the duration grade, the overcurrent score is obtained from a preset two-dimensional score table, and the state space, action space and reward value function of the protection decision model are constructed based on the overcurrent score, power data and duration grade; the protection decision model outputs a protection decision strategy, and the power system performs overcurrent protection according to the protection decision strategy; the protection decision model is trained based on the current state and the next state after the overcurrent protection. By establishing a protection decision model based on deep reinforcement learning, the two key factors of the overcurrent size and duration are fully considered, the overcurrent risk of the rectifier transformer in the entire power system is dynamically evaluated, and the system-level adaptive protection decision control strategy is output, so as to realize accurate, efficient and forward-looking overcurrent protection for the rectifier transformer, thereby significantly improving the reliability and safe operation level of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flow chart of the overcurrent protection setting method of the rectifier transformer of the power system in Embodiment 1 of the present invention;

[0064] Figure 2 A module connection relationship diagram of an overcurrent protection setting system for a rectifier transformer of a power system in Embodiment 2 of the present invention;

[0065] Figure 3 This is a schematic diagram of the structure of an electronic device in Embodiment 3 of the present invention;

[0066] Figure 4 This is a schematic diagram of the computer-readable storage medium structure in Example 4 of the present invention. DETAILED DESCRIPTION

[0067] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] Example 1

[0069] like Figure 1 As shown, the overcurrent protection setting method of the rectifier transformer of the power system includes the following steps:

[0070] Step 1: Collect system topology diagram;

[0071] Step 2: For any rectifier transformer, collect power data in real time, and determine whether overcurrent occurs based on the power data; if overcurrent occurs, go to step 3, otherwise, repeat step 2;

[0072] Step 3: For each rectifier transformer, collect current time data in real time; obtain the current level grade based on the power data, and obtain the duration grade based on the current time data and the system topology diagram;

[0073] Step 4: For each rectifier transformer, based on the current level grade and the duration grade, obtain an overcurrent score from a preset two-dimensional score table;

[0074] Step 5: Construct the state space, action space and reward value function of the protection decision model based on the overcurrent score, power data and duration level; the protection decision model outputs a protection decision strategy, and the power system performs overcurrent protection according to the protection decision strategy; the protection decision model is trained based on the current state and the next state after the overcurrent protection.

[0075] Wherein, the collection system topology diagram includes the following steps:

[0076] Step 11: Obtain a power line schematic diagram of the entire power system; specifically, the power line schematic diagram may be exported from an energy management system (EMS) or a geographic information system (GIS);

[0077] Step 12: extracting power equipment and power line connection relationships between the power equipment from the power line schematic diagram to form a topological connection relationship; the power equipment includes but is not limited to a rectifier transformer;

[0078] Step 13: Obtain the capacity of each rectifier transformer and the capacity of each power line;

[0079] Step 14: Take each power device as a topological node, the edge between each two connected topological nodes as a topological edge, the capacity of each power device as the attribute of the corresponding topological node, and the capacity of each power line as the attribute of the corresponding topological edge, and construct a system topology graph.

[0080] Furthermore, the method of collecting power data in real time is:

[0081] Collect the primary side current, secondary side current, voltage value, voltage fluctuation, current change rate and frequency of each rectifier transformer to form power data;

[0082] It is understandable that each power parameter in the power data can be obtained in real time through the corresponding power sensor, for example, the current is collected through the current sensor, and the voltage is collected through the power sensor.

[0083] In another preferred embodiment, the power data may also include other auxiliary factors that affect the overcurrent impact range, such as ambient temperature, ambient humidity, etc.

[0084] The method of judging whether an overcurrent phenomenon occurs based on power data is as follows:

[0085] Collect the rated current value of each rectifier transformer; it is understandable that the rated current value is a fixed constant of the rectifier transformer and can be directly obtained by reading the technical parameters in the production instructions of the rectifier transformer;

[0086] Preset an excess threshold H. If the result of multiplying the rated current value by the excess threshold H is less than or equal to the secondary side current in the power data, it is determined that overcurrent has occurred. If the result of multiplying the rated current value by the excess threshold H is greater than the secondary side current in the power data, it is determined that no overcurrent has occurred.

[0087] The excess threshold H is a constant greater than 1, preferably, it can be set to 1.2;

[0088] It should be noted that the degree of harm caused by overcurrent is not only related to the current amplitude, but also closely related to the duration. Moreover, the time when overcurrent occurs in each rectifier transformer in the power system is not consistent. When making overcurrent protection decisions, it is necessary to comprehensively consider the duration of overcurrent in each rectifier transformer, so it is also necessary to monitor and analyze the duration of overcurrent.

[0089] Furthermore, the method of collecting current time data in real time is:

[0090] The overcurrent duration of each rectifier transformer that has an overcurrent phenomenon is collected as current-time data;

[0091] Furthermore, the method of obtaining the current level based on the power data is:

[0092] The current level is divided into N1 power levels, and a power level threshold is set for each power level; N1 is the number of preset power levels;

[0093] Based on the product of the power level threshold and the rated current, the current range of each power level is divided;

[0094] Preferably, N1 can be set to 3, corresponding to three power levels of slight overcurrent, moderate overcurrent, and severe overcurrent;

[0095] According to the magnitude of the secondary side current in the power data, the power level of each rectifier transformer is determined, which is expressed as a current level level;

[0096] The power level threshold can be set to 1.2, 1.5 and 1.8 respectively. The current range of mild overcurrent is between 1.2 times the rated current and 1.5 times the rated current (including 1.2 times), the current range of moderate overcurrent is between 1.5 times the rated current and 1.8 times the rated current (including 1.5 times), and the current range of severe overcurrent is 1.8 times the rated current and above.

[0097] It should be noted that when an overcurrent fault occurs, the fault current not only comes from one side of the transformer, but also from other power sources connected to the transformer through busbars, lines, etc. These additional power supplies will increase the duration of the overcurrent. Therefore, the larger the capacity of other equipment and lines connected to the transformer, the easier it is to support overcurrent for a longer time. Conversely, the smaller the capacity, the shorter the duration of the overcurrent.

[0098] Therefore, the method of obtaining the duration level based on the current time data and the system topology diagram is:

[0099] The number of the rectifier transformer is marked as i, and the capacity of the i-th rectifier transformer is marked as Ci;

[0100] Calculate the sum of the capacities of the power lines connected to the i-th rectifier transformer, and mark the sum of the capacities as CTi;

[0101] The capacity ratio CBi of the i-th rectifier transformer is calculated. The capacity ratio CBi is calculated as follows: It can be understood that the larger the CBi, the greater the capacity of the line is, which means that the capacity of the line is much greater than the capacity of the rectifier transformer, and thus it can withstand overcurrent for a longer period of time;

[0102] Preset standard bearing time t;

[0103] N2 duration levels are preset, and a duration level threshold is set for each duration level according to the standard endurance time, wherein the duration level threshold is a multiple of the standard endurance time; N2 is the number of preset duration levels;

[0104] Based on the duration level threshold, the duration range of each duration level is divided;

[0105] Preferably, the standard endurance time t can be set to 30 seconds, and N2 can be set to 3, corresponding to the three duration levels of short duration, medium duration, and long duration respectively;

[0106] The duration level threshold can be set to 0, 1, and 10 respectively; the duration range for short duration is 0 seconds to 30 seconds (including 0), the duration range for medium duration is 30 seconds to 5 minutes (including 30 seconds), and the duration range for long duration is 5 minutes and above;

[0107] N3 ratio levels are preset for the capacity ratio, and a ratio threshold and a ratio coefficient are preset for each ratio level; according to the ratio threshold, the ratio range of each ratio level is divided; N3 is the number of preset ratio levels;

[0108] Preferably, N3 can be set to 5, and the proportion thresholds corresponding to the 5 proportion levels are 0, 0.5, 1, 2, and 5 respectively; the corresponding proportion ranges are 0 to 0.5 (including 0), 0.5 to 1 (including 0.5), 1 to 2 (including 1), 2 to 5 (including 2), and 5 and above; and the proportion coefficients corresponding to each proportion range can be -0.5, -0.25, 0, 0.5, and 1;

[0109] According to the capacity ratio CBi, the ratio level of the i-th rectifier transformer is determined as the capacity ratio level, and the ratio coefficient corresponding to the capacity ratio level is marked as ki;

[0110] Then the duration range of each duration level of the i-th rectifier transformer is reset to [tmin×(1+ki), tmax×(1+ki)]; where tmin is the lower limit of the duration range, and tmax is the upper limit of the duration range;

[0111] For example, when the capacity ratio of a rectifier transformer is 0.75, the duration ranges of the corresponding duration levels are updated to [0, 30×(1-0.25)=22.5), [22.5, 225), [225, +∞);

[0112] According to the current time data of each rectifier transformer and the duration range of each corresponding duration level, the duration level of each rectifier transformer is determined and expressed as a duration level;

[0113] The method of obtaining the overcurrent score from the preset two-dimensional score table based on the current level grade and the duration grade is as follows:

[0114] Preset two-dimensional scoring table;

[0115] Preferably, each row of the two-dimensional scoring table represents a time level, each column represents a power level, and the element value corresponding to each row and column represents an overcurrent score;

[0116] The overcurrent score is used to measure the current value and duration of the current overcurrent and the danger of the overcurrent. The overcurrent score is preset based on expert experience.

[0117] Preferably, an example of the two-dimensional scoring table is shown in Table 1;

[0118] Table 1 Two-dimensional scoring table

[0119] Slight overcurrent Medium overcurrent Severe overcurrent Short duration 1 2 3 Medium sustained 2 3 4 Long-term sustainability 4 5 6

[0120] According to the current level grade and duration grade of each rectifier transformer, find the overcurrent score corresponding to the power grade and duration grade from the two-dimensional score table;

[0121] By combining the two key factors of current amplitude and duration, the degree of overcurrent danger can be quantified and comprehensively assessed;

[0122] Furthermore, the state space, action space and reward value function of the protection decision model based on the overcurrent score, power data and duration level are constructed as follows:

[0123] The Actor-Critic model is selected as the protection decision model; the Actor model is used to output recommended installation parameters, the Critic model is used to evaluate the scores of the recommended installation parameters output by the Actor model, and the Actor model updates the model parameters based on the scores;

[0124] Design a state space S, wherein the state space S reflects the real-time operating state of the entire power system, and the state space S includes power data, current-time data, and overcurrent scores of all rectifier transformers;

[0125] Designing an action space A; the action space A is a system-level protection decision control strategy output by the Actor model, and the actions in the action space A include adjustment values ​​of overcurrent setting parameters of each rectifier transformer;

[0126] The overcurrent setting parameters include but are not limited to overcurrent starting current value, overcurrent delay time, locking current value and reset time; specifically:

[0127] The overcurrent starting current value is the threshold setting for judging overcurrent, which is usually set to 1.05-1.2 times the rated current. This ratio can be adjusted according to specific circumstances;

[0128] Overcurrent delay time means that after overcurrent is detected, the protection action is delayed for a period of time before starting. This delay can be set to different values, such as 0.1s, 1s, 5s, etc. Generally, the longer the duration, the shorter the delay.

[0129] For some differential protections, a locking value needs to be set. When the difference between the internal and external currents is less than this value, the protection will not operate.

[0130] The reset time is the time it takes for the protection device to reset after the overcurrent fault is eliminated.

[0131] Design a reward value function Q; the reward value function Q = r (s, a), which measures the effect of performing action a under the current state s;

[0132] Preferably, the calculation formula of the reward value function Q can be Q=r(s,a)=w1×r1+w2×r2+w3×r3;

[0133] Wherein, w1, w2 and w3 are all preset proportional coefficients greater than 0, and w1+w2+w3=1; preferably, w1=0.5, w2=0.3, w3=0.2;

[0134] Wherein, r1 = -α×FD-β×AU-θ×EC; wherein FD represents the maximum duration of overcurrent, AU represents the number of affected users, and EC represents the operating cost; α, β, and θ are all preset proportional coefficients, and α+β+θ=1;

[0135] Where r2 = ∑ i (δi×ΔV_i+εi×ΔF_i); wherein ΔV_i represents the deviation of the voltage of the rectifier transformer from the rated value; ΔF_i represents the deviation of the frequency of the rectifier transformer from the rated value; δi and εi are both preset proportional coefficients; and δi+εi=1;

[0136] Where r3 = -∑ j λj×Pj; j is the number of each element (overcurrent score) in the two-dimensional score table, λj is the overcurrent score of the jth element, and Pj is the number of rectifier converters whose overcurrent score is the jth element (overcurrent score); it can be understood that the more rectifier converters with high overcurrent scores, the greater the value of r3, and therefore, the smaller the reward value function Q is, which tends to be processed as early as possible to reduce the overcurrent danger;

[0137] It can be understood that the setting of the above reward value function Q is only a simple example provided in this embodiment. In practical applications, the reward value function Q can be adaptively modified, deleted or added;

[0138] Furthermore, the protection decision model is trained based on the current state and the next state after the overcurrent protection as follows:

[0139] The power data, current time data and overcurrent score of each rectifier transformer in the power system are taken as the current state s; the current state s is input to the Actor model of the protection decision model, and the Actor model outputs an action in the action space A so that the state-behavior value output by the Critic model is maximized, and the output action is used as the protection decision control strategy;

[0140] Execute the protection decision control strategy to obtain the next state s' and the actual reward value;

[0141] The critic model outputs the state-behavior value after executing the protection decision control strategy in the current state, calculates the TD error δ of the critic model, updates the critic model parameters according to δ, and minimizes the critic loss function; the state-behavior value output by the critic model is obtained by calculating the state-behavior value function generated by the time difference optimization method;

[0142] Specifically, the state-behavior value function generated by the temporal difference optimization method can be:

[0143] Q(s,a)=r(s,a)+γ×V(s'), γ is the discount factor, V(s') is the estimated value in state s'; generally, the estimated value can be replaced by Q(s',a'), that is, V(s') represents the estimated value of the state-behavior value function corresponding to the next state s' and action a', so as to ensure that Q(s,a) can consider the long-term instead of only the current state;

[0144] Then the TD error δ is δ=(r(s,a)+γ×V(s')-Q(s,a)) 2 ;

[0145] The Actor model takes the state-behavior value output by the Critic model as the output target and uses the policy gradient update method to update the parameters of the Actor network model.

[0146] Example 2

[0147] like Figure 2As shown, the overcurrent protection setting system of the rectifier transformer of the power system includes an overcurrent judgment module, a scoring calculation module and a protection decision module; wherein the modules are electrically connected;

[0148] The overcurrent judgment module collects the system topology diagram and the power data of any rectifier transformer in real time, and judges whether an overcurrent phenomenon occurs based on the power data; if an overcurrent phenomenon occurs, the system topology diagram and power data are sent to the scoring calculation module and the protection decision module;

[0149] The scoring calculation module collects the current-time data in real time for each rectifier transformer; obtains the current level grade based on the power data, obtains the duration grade based on the current-time data and the system topology diagram, and obtains the overcurrent score from a preset two-dimensional scoring table for each rectifier transformer based on the current level grade and the duration grade; and sends the overcurrent score to the protection decision module;

[0150] A protection decision module constructs a state space, action space and reward value function of a protection decision model based on overcurrent scores, power data and duration levels; the protection decision model outputs a protection decision strategy, and the power system performs overcurrent protection according to the protection decision strategy; the protection decision model is trained based on the current state and the next state after overcurrent protection.

[0151] Example 3

[0152] Figure 3 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 3 As shown, according to another aspect of the present application, an electronic device 100 is also provided. The electronic device 100 may include one or more processors and one or more memories. The memories store computer readable codes, and when the computer readable codes are executed by one or more processors, the overcurrent protection setting method for the rectifier transformer of the power system as described above may be executed.

[0153] The method or device according to the embodiment of the present application can also be used by Figure 3 The electronic device architecture shown in FIG. Figure 3 As shown, the electronic device 100 may include a bus 101, one or more CPUs 102, a ROM 103, a RAM 104, a communication port 105 connected to a network, an input / output component 106, a hard disk 107, etc. A storage device in the electronic device 100, such as the ROM 103 or the hard disk 107, may store the overcurrent protection setting method for the rectifier transformer of the power system provided in the present application.

[0154] Furthermore, the electronic device 100 may further include a user interface 108. Figure 3 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 3 One or more components of an electronic device are shown.

[0155] Example 4

[0156] Figure 4 Schematic diagram of a computer-readable storage medium structure provided by an embodiment of the present application. Figure 4 As shown, a computer-readable storage medium 200 according to an embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 200. When the computer-readable instructions are executed by the processor, the overcurrent protection setting method of the rectifier transformer of the power system according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium 200 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0157] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, the non-transitory machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided by the present application, and when the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0158] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0159] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0160] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0161] The above preset parameters or preset thresholds are all set by technicians in this field according to actual conditions or obtained through large-scale data simulation.

[0162] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for setting overcurrent protection of a rectifier transformer in a power system, characterized in that: The following steps are involved: Step 1: Collect system topology diagram; Step 2: For any rectifier transformer, collect power data in real time, and determine whether overcurrent occurs based on the power data; If overcurrent occurs, go to step 3, otherwise, repeat step 2; Step 3: For each rectifier transformer, collect current-time data in real time; Based on the power data, the current level level is obtained, and based on the current time data and the system topology diagram, the duration level is obtained; Step 4: For each rectifier transformer, based on the current level grade and the duration grade, obtain an overcurrent score from a preset two-dimensional score table; Step 5: Construct the state space, action space and reward value function of the protection decision model based on the overcurrent score, power data and duration level; the protection decision model outputs a protection decision strategy, and the power system performs overcurrent protection according to the protection decision strategy; the protection decision model is trained based on the current state and the next state after the overcurrent protection.

2. The overcurrent protection setting method for the rectifier transformer of the power system according to claim 1, characterized in that: The collecting system topology diagram comprises the following steps: Step 11: Obtain a power line schematic diagram of the entire power system; Step 12: extracting power equipment and power line connection relationships between the power equipment from the power line schematic diagram to form a topological connection relationship; Step 13: Obtain the capacity of each rectifier transformer and the capacity of each power line; Step 14: Take each power device as a topological node, the edge between each two connected topological nodes as a topological edge, the capacity of each power device as the attribute of the corresponding topological node, and the capacity of each power line as the attribute of the corresponding topological edge, and construct a system topology graph.

3. The overcurrent protection setting method for the rectifier transformer of the power system according to claim 2 is characterized in that: The method of judging whether an overcurrent phenomenon occurs based on power data is: Collect the rated current value of each rectifier transformer; A preset excess threshold H is set. If the result of multiplying the rated current value by the excess threshold H is less than or equal to the secondary side current in the power data, it is judged that overcurrent has occurred. If the result of multiplying the rated current value by the excess threshold H is greater than the secondary side current in the power data, it is judged that no overcurrent has occurred.

4. The overcurrent protection setting method for the rectifier transformer of the power system according to claim 3 is characterized in that: The method of collecting current time data in real time is: The overcurrent duration of each rectifier transformer that has an overcurrent phenomenon is collected as current-time data; The method of obtaining the current level based on the power data is as follows: The current level is divided into N1 power levels, and a power level threshold is set for each power level; N1 is the number of preset power levels; Based on the product of the power level threshold and the rated current, the current range of each power level is divided; According to the magnitude of the secondary-side current in the power data, the power level of each rectifier transformer is determined and expressed as a current level.

5. The overcurrent protection setting method for the rectifier transformer of the power system according to claim 4, characterized in that: The method of obtaining the duration level based on the current time data and the system topology diagram is as follows: The number of the rectifier transformer is marked as i, and the capacity of the i-th rectifier transformer is marked as Ci; Calculate the sum of the capacities of the power lines connected to the i-th rectifier transformer, and mark the sum of the capacities as CTi; The capacity ratio CBi of the i-th rectifier transformer is calculated. The capacity ratio CBi is calculated as follows: N2 duration levels are preset, and a duration level threshold is set for each duration level according to the standard endurance time, wherein the duration level threshold is a multiple of the standard endurance time; N2 is the number of preset duration levels; Based on the duration level threshold, the duration range of each duration level is divided; N3 ratio levels are preset for the capacity ratio, and a ratio threshold and a ratio coefficient are preset for each ratio level; according to the ratio threshold, the ratio range of each ratio level is divided; N3 is the number of preset ratio levels; According to the capacity ratio CBi, the ratio level of the i-th rectifier transformer is determined as the capacity ratio level, and the ratio coefficient corresponding to the capacity ratio level is marked as ki; Then the duration range of each duration level of the i-th rectifier transformer is reset to [tmin×(1+ki), tmax×(1+ki)]; where tmin is the lower limit of the duration range, and tmax is the upper limit of the duration range; According to the current-time data of each rectifier transformer and the duration range of each corresponding duration level, the duration level of each rectifier transformer is determined and expressed as a duration level.

6. The overcurrent protection setting method for the rectifier transformer of the power system according to claim 5, characterized in that: The method of obtaining the overcurrent score from the preset two-dimensional score table based on the current level grade and the duration grade is as follows: A two-dimensional scoring table is preset; each row of the two-dimensional scoring table represents a time level, each column represents a power level, and the element value corresponding to each row and column represents an overcurrent score; According to the current level grade and duration grade of each rectifier transformer, the overcurrent score corresponding to the power grade and duration grade is found from the two-dimensional score table.

7. The overcurrent protection setting method for the rectifier transformer of the power system according to claim 6, characterized in that: The method of constructing the state space, action space and reward value function of the protection decision model based on the overcurrent score, power data and duration level is as follows: The Actor-Critic model was selected as the protection decision model; Design a state space S, wherein the state space S reflects the real-time operating state of the entire power system, and the state space S includes power data, current-time data, and overcurrent scores of all rectifier transformers; Designing an action space A; the action space A is a system-level protection decision control strategy output by the Actor model, and the actions in the action space A include an adjustment value of an overcurrent setting parameter of each rectifier transformer; Design a reward value function Q; the reward value function Q = r(s, a) measures the effect of performing action a in the current state s.

8. The overcurrent protection setting method for the rectifier transformer of the power system according to claim 7, characterized in that: The protection decision model is trained based on the current state and the next state after overcurrent protection as follows: The power data, current time data and overcurrent score of each rectifier transformer in the power system are taken as the current state s; Input the current state s to the Actor model of the protection decision model, and the Actor model outputs an action in the action space A so that the state-behavior value output by the Critic model is maximized, and the output action is used as the protection decision control strategy; Execute the protection decision control strategy to obtain the next state s' and the actual reward value; The Critic model outputs the state-behavior value after executing the protection decision control strategy in the current state, calculates the TD error δ of the Critic model, updates the Critic model parameters according to δ, and minimizes the Critic loss function; the state-behavior value output by the Critic model is obtained by calculating the state-behavior value function generated by the time difference optimization method; The Actor model takes the state-behavior value output by the Critic model as the output target and uses the policy gradient update method to update the parameters of the Actor network model.

9. An overcurrent protection setting system for a rectifier transformer of an electric power system, which is used to implement the overcurrent protection setting method for a rectifier transformer of an electric power system as claimed in any one of claims 1 to 8, characterized in that: It includes an overcurrent judgment module, a scoring calculation module and a protection decision module; wherein each module is electrically connected; The overcurrent judgment module collects the system topology diagram and the power data of any rectifier transformer in real time, and judges whether an overcurrent phenomenon occurs based on the power data; if an overcurrent phenomenon occurs, the system topology diagram and power data are sent to the scoring calculation module and the protection decision module; The scoring calculation module collects the current-time data in real time for each rectifier transformer; obtains the current level grade based on the power data, obtains the duration grade based on the current-time data and the system topology diagram, and obtains the overcurrent score from a preset two-dimensional scoring table for each rectifier transformer based on the current level grade and the duration grade; and sends the overcurrent score to the protection decision module; A protection decision module constructs a state space, action space and reward value function of a protection decision model based on overcurrent scores, power data and duration levels; the protection decision model outputs a protection decision strategy, and the power system performs overcurrent protection according to the protection decision strategy; the protection decision model is trained based on the current state and the next state after overcurrent protection.

10. An electronic device, characterized in that: include: A processor and a memory, wherein: The memory stores a computer program that can be called by the processor; The processor executes the overcurrent protection setting method for the rectifier transformer of the power system as described in any one of claims 1 to 8 in the background by calling the computer program stored in the memory.

11. A computer-readable storage medium, characterized in that: A rewritable computer program is stored thereon; When the computer program is executed on a computer device, the computer device executes the overcurrent protection setting method for a rectifier transformer of a power system as claimed in any one of claims 1 to 8 in the background.

Images