Sectional type intelligent matching setting method for power grid distance protection

Abstract

The embodiment of the invention provides a power grid distance protection sectional type intelligent cooperation setting method, and relates to the technical field of power system relay protection, and the method comprises the steps: responding to the execution starting time t of a periodic task, and collecting and obtaining the state information of a power system at the t moment; wherein the power system state information comprises a power system bus voltage vector state information set, a power system bus current vector state information set and a power system circuit breaker state information set; traversing the value reward experience pool, determining a corresponding decision strategy as a historical optimal strategy according to the optimal value recorded in the value reward experience pool, making a decision according to the historical optimal strategy and a random decision algorithm to obtain a decision strategy of the current period, and issuing the decision strategy to the protection equipment at the corresponding position. The protection equipment controls the circuit breaker to act according to the decision strategy; wherein the decision strategy comprises a sectional distance protection impedance setting value and a sectional distance protection action time limit setting value. According to the power grid distance protection sectional type intelligent cooperation setting method provided by the embodiment of the invention, the protection strategy can be automatically learned and optimized, the requirements of quick response and intelligent decision making are met, different power grid environments and operation conditions are adapted, and the accuracy, reliability and flexibility of power grid distance protection are improved.

Description

Technical Field

[0001] This invention relates to the field of power system relay protection technology, specifically to a segmented intelligent coordination setting method for power grid distance protection. Background Technology

[0002] The power systems of refining and chemical enterprises require efficient protection measures due to their complexity. These systems typically contain a large number of devices and lines and operate in constantly changing environments, thus necessitating reliable protection technologies to ensure the safe and stable operation of the power grid.

[0003] In refining and chemical enterprises, distance protection is a crucial technology used to quickly and accurately diagnose and locate faults in the power grid, thereby preventing the fault from escalating and even causing accidents. Distance protection technology measures the distance from the fault point to the protection device and makes judgments based on power grid parameters, enabling it to quickly disconnect the fault area and ensure the safety and reliability of power grid operation.

[0004] However, distance protection also presents some challenges in its application. For example, complex power grid structures and diverse operating conditions can affect the accuracy and reliability of the protection system. Furthermore, traditional distance protection technologies may not meet the demands for rapid response and intelligent decision-making, exhibiting limited flexibility. Summary of the Invention

[0005] To address the shortcomings of existing technologies, embodiments of the present invention provide a segmented intelligent coordination setting method for power grid distance protection.

[0006] This invention provides a segmented intelligent coordination setting method for power grid distance protection, comprising: a first state information acquisition step: in response to the arrival of the execution start time t of a periodic task, acquiring and obtaining power system state information at time t; wherein, the power system state information includes a power system bus voltage vector state information set, a power system bus current vector state information set, and a power system circuit breaker state information set; a decision-making step: traversing a value reward experience pool, determining the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, performing a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and distributing the decision strategy to the protection equipment at the corresponding location so that the protection equipment can control the circuit breaker to operate according to the decision strategy; wherein, the decision strategy includes segmented distance protection impedance setting values ​​and segmented distance protection operation time limit setting values.

[0007] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided. After the decision-making step, the method further includes: a second state information acquisition step: in response to the execution start time t+1 of the next cycle, acquiring and obtaining power system state information at time t+1; a reward value calculation step: acquiring the distance protection segmented impedance setting value coordination reward value, the distance protection segmented action time setting value coordination reward value, and the circuit breaker state reward value for the current cycle, and calculating the reward value for the current cycle based on the distance protection segmented impedance setting value coordination reward value, the distance protection segmented action time setting value coordination reward value, and the circuit breaker state reward value; a value calculation step: calculating the value of the current cycle after executing the decision strategy based on the power system state information at time t; and a storage step: storing the power system state information at time t, the decision strategy for the current cycle, the power system state information at time t+1, the reward value for the current cycle, and the updated value for the current cycle into the value reward experience pool.

[0008] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided. Before the first state information acquisition step, the method further includes a training step, which includes executing the first state information acquisition step, the decision step, the second state information acquisition step, the reward value calculation step, and the value calculation step a preset number of times.

[0009] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided. The step of calculating the value of the decision strategy after the current cycle update based on the power system state information at time t includes: calculating the value after the current cycle update based on the power system state information at time t based on the value before the current cycle update, the reward value of the historical best decision based on the power system state information at time t+1, the reward value of the current cycle, the autonomous learning rate, and the reward decay rate.

[0010] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the power system bus voltage vector state information set is represented as follows:

[0011]

[0012] in, Let be the set of bus voltage vector state information of the power system at time t. Let be the voltage vector value across the circuit breaker numbered j on the busbar numbered i;

[0013] The power system bus current vector state information set is represented as follows:

[0014]

[0015] in, Let be the set of bus current vector state information of the power system at time t. Let J be the vector value of the current passing through the circuit breaker numbered j on bus i.

[0016] The power system circuit breaker status information set is represented as follows:

[0017]

[0018] Among them, S (t) Let t be the set of power system circuit breaker status information. These are the operation status information of the first, second, and third stage protection of the distance protection of the circuit breaker numbered j on busbar numbered i.

[0019] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the segmented distance protection impedance setting value is expressed as:

[0020]

[0021] Among them, Z (t) Let be the segmented distance protection impedance setting value of the circuit breaker on each branch of each busbar at time t. These are the impedance setting values ​​for the first, second, and third stages of the distance protection of the circuit breaker numbered j on busbar numbered i, respectively.

[0022] The segmented distance protection action time limit setting value is expressed as follows:

[0023]

[0024] Among them, T (t) Let t be the segmented distance protection operating time setting value for the circuit breakers on each branch of each busbar at time t. These are the time limit settings for the first, second, and third stages of the distance protection of the circuit breaker numbered j on busbar numbered i.

[0025] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the reward value of the period to which time t belongs is expressed as:

[0026] R (t) =R Z(t) +R T(t) +R bre(t)

[0027] Among them, R (t) R is the reward value for the period to which time t belongs. Z(t)R is the distance protection segmented impedance setting value and bonus value for the period to which time t belongs. T(t) R is the distance protection segmented action time limit setting value for the period to which time t belongs, combined with the reward value. bre(t) The circuit breaker state reward value for the period to which time t belongs.

[0028] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the segmented impedance setting value of the distance protection at time t is coordinated with the bonus value R. Z(t) Represented as:

[0029]

[0030] in, This indicates the reward for the distance protection impedance setting value of the first stage of the cycle at time t. This indicates the reward for the second-stage distance protection impedance setting value within the cycle at time t. This indicates the reward for the three-stage distance protection impedance setting value of the cycle at time t;

[0031]

[0032] Among them, a Ⅰ b Ⅰ L is the first-stage distance protection reward coefficient. i Let z be the length of the busbar numbered i. i Let i be the positive sequence impedance per unit length of the busbar numbered i;

[0033]

[0034] Among them, a Ⅱ b Ⅱ c Ⅱ d Ⅱ This refers to the second-stage distance protection reward coefficient. The impedance setting value that is coordinated with the second-stage impedance setting value on the next busbar numbered i;

[0035]

[0036] Among them, a Ⅲ b Ⅲ c Ⅲ d Ⅲ This refers to the second-stage distance protection reward coefficient. This is the impedance setting value that coordinates with the three-stage impedance setting value on the next busbar numbered i.

[0037] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the segmented action time setting value of the distance protection at time t is coordinated with the bonus value R.T(t) Represented as:

[0038]

[0039] Where k1, k2, k3, k4, and k5 are the reward coefficients for the three-stage distance protection time limit. Δt represents the three-stage operating time limit setting value of the distance protection for circuit breaker j on the next busbar numbered i, and Δt represents the three-stage operating time limit coordination range of the distance protection.

[0040] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the state reward of the circuit breaker in the cycle at time t is R. bre(t) When the circuit breaker at this level trips, and when the trip signal is sent by the first-stage distance protection, R bre(t) R is a positive value when the trip signal is sent by a two-stage or three-stage distance protection. bre(t) The value is negative, and the trip signal is generated by R when the second-stage distance protection is activated. bre(t) The value is greater than the trip signal transmitted by the three-stage distance protection R. bre(t) Value; When the over-current circuit breaker trips, and the trip signal is sent by a one-stage, two-stage, or three-stage distance protection, R bre(t) The value is negative, and the trip signal is generated by R when the first-stage distance protection is activated. bre(t) The value is greater than the trip signal transmitted by the two-stage distance protection R. bre(t) The trip signal is generated by R during the second-stage distance protection signal transmission. bre(t) The value is greater than the trip signal transmitted by the three-stage distance protection R. bre(t) value.

[0041] This invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the segmented intelligent coordination setting method for power grid distance protection as described above.

[0042] This invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the segmented intelligent coordination setting method for power grid distance protection as described above.

[0043] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above-described segmented intelligent coordination setting methods for power grid distance protection.

[0044] The segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention collects and obtains the power system state information at time t in response to the start time t of the periodic task execution, traverses the value reward experience pool, determines the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, makes a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and distributes the decision strategy to the corresponding protection equipment so that the protection equipment can control the circuit breaker to operate according to the decision strategy. It can automatically learn and optimize the protection strategy, meet the needs of rapid response and intelligent decision-making, adapt to different power grid environments and operating conditions, and improve the accuracy, reliability and flexibility of power grid distance protection. Attached Figure Description

[0045] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is one of the flowcharts of the segmented intelligent coordination setting method for power grid distance protection provided in the embodiments of the present invention;

[0047] Figure 2 This is the second flowchart of the segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention;

[0048] Figure 3 This is a schematic diagram of the power grid structure in the segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention;

[0049] Figure 4 This is a schematic diagram of the structure of the segmented intelligent coordination setting device for power grid distance protection provided in an embodiment of the present invention;

[0050] Figure 5 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0052] Figure 1This is one of the flowcharts illustrating the segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention. For example... Figure 1 As shown, the method includes:

[0053] Step S1: In response to the arrival of the execution start time t of the periodic task, collect and obtain the power system status information at time t; wherein, the power system status information includes the power system bus voltage vector status information set, the power system bus current vector status information set, and the power system circuit breaker status information set.

[0054] Before collecting status information, it is necessary to model the power system. Power system modeling is a crucial step in the distance protection optimization scheme, involving the modeling and description of various parameters and states of the refining enterprise's power system. Parameters include those of various nodes in the power grid, such as the resistance, reactance, and capacitance of transformers, circuit breakers, and disconnectors. The description mainly involves numbering the connections between devices at various nodes in the power grid. The numbering rule starts from the initial busbar of the power grid system and proceeds step-by-step from the beginning to the end; circuit breakers on each branch of each numbered busbar are also numbered.

[0055] Step S1 is also known as the first state information acquisition step. In this step, in response to the arrival of the execution start time t of the periodic task, the power system state information at time t is acquired and obtained. This power system state information includes the power system bus voltage vector state information set, the power system bus current vector state information set, and the power system circuit breaker state information set. The power system bus voltage vector state information set and the power system bus current vector state information set include the voltage and current vector values ​​on each line and at key nodes in the system. The power system circuit breaker state information set includes the open / closed status information of the circuit breakers on each branch of each bus.

[0056] Step S2: Traverse the value reward experience pool, determine the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, make a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and send the decision strategy to the protection device at the corresponding location so that the protection device can control the circuit breaker to operate according to the decision strategy; wherein, the decision strategy includes the segmented distance protection impedance setting value and the segmented distance protection operation time setting value.

[0057] The power system of refining and chemical enterprises employs segmented protection for distance protection, with mutual protection and backup to achieve full coverage of the protection range. For power systems using segmented distance protection, circuit breakers on each branch of each busbar are set with segmented distance protection impedance settings. Since the distance protection uses under-limit protection, reflecting the distance from the fault point to the installation location of the protection equipment, when the short-circuit distance (the impedance from the short-circuit point to the busbar) is less than the impedance setting value set by the operating unit, the protection equipment is activated, isolating the fault point by controlling the circuit breaker operation.

[0058] Step S2 is also known as the decision-making step. In this step, the value reward experience pool is traversed, and the corresponding decision strategy is determined as the historical optimal strategy based on the optimal value recorded in the pool. This corresponding decision strategy is the one obtained during the same period of operation corresponding to the optimal value. The decision strategy for the current period is obtained by using the historical optimal strategy and a stochastic decision-making algorithm. This decision strategy is then distributed to the corresponding protection devices, allowing them to control the circuit breakers according to the strategy. Each protection device can control one or more different circuit breakers based on its location.

[0059] When obtaining a decision strategy, to prevent the system from consistently choosing the best historical decision action from the value reward experience pool and getting trapped in a local optimum, a decision strategy for the current period is obtained by combining the historical best strategy with a stochastic decision-making algorithm. Stochastic decision-making algorithms can employ strategies such as the ε-greedy strategy, as well as greedy strategies, Gaussian strategies, and Boltzmann distributions. When obtaining a decision strategy, there is a certain probability of choosing the historical best strategy, and a certain probability of randomly selecting a strategy. The advantage of random selection is that, on the one hand, it can explore the data space beyond the historical best data space, preventing over-reliance on historical best data and getting trapped in a local optimum; on the other hand, it can obtain value rewards better than the historical best data through random trial and error, thereby expanding the historical reward value database.

[0060] The ε-greedy strategy is represented as:

[0061]

[0062] Among them, O(Z) (t) |S (t) S is represented by time t. (t) Under the condition, the segmented distance protection impedance setting value Z is executed. (t) The strategy is ε, where ε is the greed coefficient of the ε-greedy strategy, and R is the number of strategies. O (Z (t) |S (t) ) represents time t, S (t) Under the condition, the segmented distance protection impedance setting value Z is executed.(t) The historically optimal strategy at that time.

[0063] The segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention collects and obtains the power system state information at time t in response to the start time t of the periodic task execution, traverses the value reward experience pool, determines the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, makes a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and distributes the decision strategy to the corresponding protection equipment so that the protection equipment can control the circuit breaker to operate according to the decision strategy. It can automatically learn and optimize the protection strategy, meet the needs of rapid response and intelligent decision-making, adapt to different power grid environments and operating conditions, and improve the accuracy, reliability and flexibility of power grid distance protection.

[0064] Figure 2 This is the second flowchart illustrating the segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention. Figure 2 As shown, after the decision-making step S2, the method further includes:

[0065] Step S3: In response to the arrival of the execution start time t+1 of the next cycle, collect and obtain the power system status information at time t+1.

[0066] Step S3 is also known as the second state information acquisition step. In this step, in response to the arrival of the execution start time t+1 of the next cycle, the power system state information at time t+1 is acquired, that is, the power system bus voltage vector state information set, the power system bus current vector state information set, and the power system circuit breaker state information set at time t+1 are acquired.

[0067] Step S4: Obtain the distance protection segmented impedance setting value with reward value, the distance protection segmented operation time setting value with reward value, and the circuit breaker status reward value for the current cycle. Calculate the reward value for the current cycle based on the distance protection segmented impedance setting value with reward value, the distance protection segmented operation time setting value with reward value, and the circuit breaker status reward value.

[0068] Step S4 is also known as the reward value calculation step. The reward value for the current period is related to the reward value for the distance protection segmented impedance setting, the reward value for the distance protection segmented operating time setting, and the reward value for the circuit breaker status. Therefore, it is necessary to obtain the reward value for the distance protection segmented impedance setting, the reward value for the distance protection segmented operating time setting, and the reward value for the circuit breaker status for the current period, and calculate the reward value for the current period based on these values.

[0069] Step S5: Calculate the value of the decision strategy after the current period update based on the power system state information at time t.

[0070] Step S5 is also known as the value calculation step. In this step, the value of the decision-making strategy executed based on the power system state information at time t after the current period update is calculated. In other words, it is used to evaluate the value of executing the corresponding decision-making strategy under the power system state information at time t.

[0071] Step S6: Store the power system state information at time t, the decision strategy for the current period, the power system state information at time t+1, the reward value for the current period, and the updated value for the current period into the value reward experience pool.

[0072] Step S6 is also known as the storage step. To accumulate data for the value reward experience pool, the power system state information at time t, the decision strategy for the current period, the power system state information at time t+1, the reward value for the current period, and the updated value for the current period are stored in the value reward experience pool. This data belongs to the data obtained within one operating cycle and can be stored as a single data entry.

[0073] Understandably, various data can be stored in the value reward experience pool after the corresponding data is obtained, and the data obtained later can be stored in the same way as the data obtained earlier, without necessarily storing it all in step S6.

[0074] The segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention collects and obtains the power system status information at time t+1 in response to the start time t+1 of the next cycle, calculates the reward value of the current cycle, calculates the updated value of the decision-making strategy based on the power system status information at time t, and stores the obtained data in the value reward experience pool. This realizes the gradual accumulation of data in the value reward experience pool, which is conducive to obtaining more optimized decision results.

[0075] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided. Before the first state information acquisition step, the method further includes a training step, which includes executing the first state information acquisition step, the decision step, the second state information acquisition step, the reward value calculation step, and the value calculation step a preset number of times.

[0076] The above embodiments illustrate the execution process during actual power system protection. Before actual power system protection, the model can be trained multiple times through training steps. These training steps include executing a first state information acquisition step, a decision-making step, a second state information acquisition step, a reward value calculation step, and a value calculation step a preset number of times. This results in a relatively optimized value reward experience pool, facilitating better decision-making. During the initial training, the values ​​of the power system under different states can be initialized, and a maximum value can be set for the value under the optimal state of the power system. This maximum value can be adjusted based on the actual situation and the final training results of the model.

[0077] The execution cycles during training and application can be the same or different.

[0078] The segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention trains the model by performing the first state information acquisition step, decision-making step, second state information acquisition step, reward value calculation step, and value calculation step a preset number of times before the first state information acquisition step. This achieves the acquisition of an optimized value reward experience pool, which is beneficial for making better decisions.

[0079] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided. The step of calculating the value of the decision strategy after the current cycle update based on the power system state information at time t includes: calculating the value after the current cycle update based on the power system state information at time t based on the value before the current cycle update, the reward value of the historical best decision based on the power system state information at time t+1, the reward value of the current cycle, the autonomous learning rate, and the reward decay rate.

[0080] The value of the decision-making strategy updated in the current period based on the power system state information at time t is calculated and expressed as:

[0081] Q new (S t A t )←Q old (S t A t )+α(R t +γ·Q(S t+1 ,A*)-Q old (S t A t ))

[0082] Among them, Q new (S t A t Q represents the value of the decision-making strategy executed based on the power system state information at time t, updated in the current period. old(S t A t R represents the value before the current cycle update, α represents the self-learning rate, and R0 represents the value before the current cycle update. t Q(S) represents the reward value for the current period, γ represents the reward decay rate, and Q(S) represents the reward value for the current period. t+1 A*) represents the power system state information S based on time t+1. t+1 The reward value for executing the historically best decision A*, with the arrow indicating the assignment operation.

[0083] The segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention calculates the value of the decision-making strategy executed based on the power system state information at time t after the current cycle update by calculating the value before the current cycle update, the reward value of executing the historical best decision based on the power system state information at time t+1, the reward value of the current cycle, the autonomous learning rate, and the reward decay rate. This achieves accurate value calculation.

[0084] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the power system bus voltage vector state information set is represented as follows:

[0085]

[0086] in, Let be the set of bus voltage vector state information of the power system at time t. Let be the voltage vector value across the circuit breaker numbered j on the busbar numbered i;

[0087] The power system bus current vector state information set is represented as follows:

[0088]

[0089] in, Let be the set of bus current vector state information of the power system at time t. Let J be the vector value of the current passing through the circuit breaker numbered j on bus i.

[0090] The power system circuit breaker status information set is represented as follows:

[0091]

[0092] Among them, S (t) Let t be the set of power system circuit breaker status information. These are the operation status information for the first, second, and third stages of distance protection for the circuit breaker numbered j on busbar numbered i. The operation status information for the first, second, and third stages of distance protection for the circuit breaker numbered j on busbar numbered i can be 0 for closed and 1 for open.

[0093] The segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention facilitates the collection of power system status information by giving expressions for the power system bus voltage vector state information set, the power system bus current vector state information set, and the power system circuit breaker state information set.

[0094] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the segmented distance protection impedance setting value is expressed as:

[0095]

[0096] Among them, Z (t) Let be the segmented distance protection impedance setting value of the circuit breaker on each branch of each busbar at time t. These are the impedance setting values ​​for the first, second, and third stages of the distance protection of the circuit breaker numbered j on busbar numbered i, respectively.

[0097] The segmented distance protection action time limit setting value is expressed as follows:

[0098]

[0099] Among them, T (t) Let t be the segmented distance protection operating time setting value for the circuit breakers on each branch of each busbar at time t. These are the time limit settings for the first, second, and third stages of the distance protection of the circuit breaker numbered j on busbar numbered i.

[0100] The segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention provides a basis for determining the segmented distance protection impedance setting value and the segmented distance protection operation time setting value by giving expressions for the segmented distance protection impedance setting value and the segmented distance protection operation time setting value.

[0101] According to an embodiment of the present invention, a segmented intelligent coordination setting method for power grid distance protection is provided, wherein the reward value of the period to which time t belongs is expressed as:

[0102] R (t) =R Z(t) +R T(t) +R bre(t)

[0103] Among them, R (t) R is the reward value for the period to which time t belongs. Z(t) R is the distance protection segmented impedance setting value and bonus value for the period to which time t belongs. T(t) R is the distance protection segmented action time limit setting value for the period to which time t belongs, combined with the reward value. bre(t)The circuit breaker state reward value for the period to which time t belongs.

[0104]

[0105] in, This indicates the reward for the distance protection impedance setting value of the first stage of the cycle at time t. This indicates the reward for the second-stage distance protection impedance setting value within the cycle at time t. This indicates the reward for the distance protection impedance setting value of the three-stage cycle at time t.

[0106]

[0107] Among them, a Ⅰ b Ⅰ L is the first-stage distance protection reward coefficient. i Let z be the length of the busbar numbered i. i Let i be the positive sequence impedance per unit length of the busbar numbered i.

[0108]

[0109] Among them, a Ⅱ b Ⅱ c Ⅱ d Ⅱ This refers to the second-stage distance protection reward coefficient. The impedance setting value for the next-level bus, numbered i, is coordinated with the second-stage impedance setting value and is related to the topology of the next-level line, as shown in the following formula:

[0110]

[0111] When the downstream line is a pure line, the impedance setting value needs to be the same as the first-stage impedance setting value Z of the downstream line distance protection. Ⅰ (i+1)j In coordination; when the downstream line contains a transformer, the impedance setting value needs to be coordinated with the downstream fast protection impedance setting value Z. T.(i+1) The impedance setting is ultimately selected based on the minimum impedance value under two different conditions. T.(i+1) The subscript T indicates a transformer.

[0112]

[0113] Among them, a Ⅲ b Ⅲ c Ⅲ d Ⅲ This refers to the second-stage distance protection reward coefficient. The impedance setting value for the next-level bus, numbered i, is coordinated with the three-stage impedance setting value and is related to the topology of the next-level line, as shown in the following formula:

[0114]

[0115] When the downstream line is a pure line, the impedance setting value needs to be the same as the second-stage impedance setting value Z of the downstream line's distance protection. Ⅱ (i+1)j Matching or with the three-stage impedance setting Z Ⅲ (i+1)j In addition, the impedance setting must also avoid the minimum load impedance during normal operation, when the maximum load on the line is... And the lowest bus voltage is When the load impedance is at its minimum. and Choose the smallest value among the three and determine the outcome.

[0116] For segmented distance protection settings, not only the impedance setting value but also the operating time limit is included. The segmented operating time limit setting value of the distance protection within the cycle to which time t belongs is combined with the bonus value R. T(t) Represented as:

[0117]

[0118] Where k1, k2, k3, k4, and k5 are the reward coefficients for the three-stage distance protection time limit. Δt is the three-stage operating time setting value of the distance protection of circuit breaker j on the next busbar numbered i, and Δt is the three-stage operating time coordination range of the distance protection, which is generally greater than 0.2s.

[0119] In addition, the value reward also includes the circuit breaker state reward for the period at time t, which is R. bre(t) Define the fault point as d ij When the fault occurs on the branch circuit of the busbar numbered i, where the circuit breaker numbered j is located, the fault code S... ij The tripping of a circuit breaker with the specified number constitutes a tripping action at this level; any tripping action of any other circuit breaker with the specified number constitutes an out-of-level or erroneous tripping action.

[0120] The reward for the circuit breaker state in the cycle at time t is R. bre(t) When the circuit breaker at this level trips, and when the trip signal is sent by the first-stage distance protection, R bre(t) R is a positive value when the trip signal is sent by a two-stage or three-stage distance protection. bre(t) The value is negative, and the trip signal is generated by R when the second-stage distance protection is activated. bre(t) The value is greater than the trip signal transmitted by the three-stage distance protection R. bre(t) value;

[0121] When a cascading circuit breaker trips, and the trip signal is transmitted by a one-stage, two-stage, or three-stage distance protection, R bre(t) The value is negative, and the trip signal is generated by R when the first-stage distance protection is activated. bre(t) The value is greater than the trip signal transmitted by the two-stage distance protection R. bre(t) The trip signal is generated by R during the second-stage distance protection signal transmission. bre(t) The value is greater than the trip signal transmitted by the three-stage distance protection R. bre(t) value.

[0122] Therefore, the reward method can be defined as follows:

[0123] When the circuit breaker at this level trips, the information acquisition unit can determine which protection mode among the three-stage settings caused the circuit breaker to open:

[0124] ① When the trip signal is sent by the first-stage distance protection, the circuit breaker status reward R bre(t) It is 10;

[0125] ② When the trip signal is sent by the two-stage distance protection, the circuit breaker status reward R bre(t) -10;

[0126] ③ When the trip signal is sent by the three-stage distance protection, the circuit breaker status reward R bre(t) It is -20.

[0127] Similarly, when a circuit breaker trips due to an over-limit circuit breaker:

[0128] ① When the trip signal is sent by the first-stage distance protection, the circuit breaker status reward R bre(t) -100;

[0129] ② When the trip signal is sent by the two-stage distance protection, the circuit breaker status reward R bre(t) -200;

[0130] ③ When the trip signal is sent by the three-stage distance protection, the circuit breaker status reward R bre(t) It is -300.

[0131] The segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention, based on artificial intelligence algorithms, establishes a reward mechanism applicable to the distance protection principle and the actual power grid, which can automatically learn and optimize protection strategies, realize intelligent decision-making, adapt to different power grid environments and operating conditions, and improve the accuracy, reliability and flexibility of distance protection.

[0132] Figure 3 This is a schematic diagram of the power grid structure in the segmented intelligent coordination setting method for power grid distance protection provided in this embodiment of the invention. Figure 3As shown, the power system consists of two 110kV lines AB and BC, and one line DE containing a 110kV to a 35kV transformer. The maximum load current I of lines AB, BC, and DE is... D.max The impedance Z per kilometer of each line is 350A, 150A, and 200A respectively. ij =0.4Ω / km, line length L A-B =30km, L B-C =60km, L D-E =80km, the equivalent impedance of the transformer (fast protection impedance setting) is 44.1Ω.

[0133] In this embodiment, the self-learning rate α = 0.01, the reward decay rate γ = 0.1, the greediness ε = 0.9, and the number of training rounds is 100.

[0134] After multiple rounds of iterative convergence, the optimal strategy was finally obtained for the power system in the implementation example. The optimal action setting values ​​(including the segmented distance protection impedance setting value and the segmented distance protection action time setting value) are shown in the table below.

[0135]

[0136] It should be noted that the preferred embodiments given in this example can be freely combined, provided that there is no logical or structural conflict between them, and the present invention does not limit them.

[0137] The following describes the segmented intelligent coordination setting device for power grid distance protection provided in the embodiments of the present invention. The segmented intelligent coordination setting device for power grid distance protection described below and the segmented intelligent coordination setting method for power grid distance protection described above can be referred to in correspondence with each other.

[0138] Figure 4 This is a schematic diagram of the structure of the segmented intelligent coordination setting device for power grid distance protection provided in an embodiment of the present invention, as shown below. Figure 4 As shown, the device includes:

[0139] The first state information acquisition module 10 is used to: in response to the arrival of the execution start time t of the periodic task, acquire and obtain the power system state information at time t; wherein, the power system state information includes a power system bus voltage vector state information set, a power system bus current vector state information set, and a power system circuit breaker state information set.

[0140] The decision module 20 is used to: traverse the value reward experience pool, determine the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, make a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and send the decision strategy to the protection device at the corresponding location so that the protection device can control the circuit breaker to operate according to the decision strategy; wherein, the decision strategy includes the segmented distance protection impedance setting value and the segmented distance protection operation time setting value.

[0141] The segmented intelligent coordination setting device for power grid distance protection provided in this embodiment of the invention collects and obtains the power system status information at time t in response to the start time t of the periodic task execution, traverses the value reward experience pool, determines the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, makes a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and distributes the decision strategy to the protection equipment at the corresponding location so that the protection equipment can control the circuit breaker to operate according to the decision strategy. It can automatically learn and optimize the protection strategy, meet the needs of rapid response and intelligent decision-making, adapt to different power grid environments and operating conditions, and improve the accuracy, reliability and flexibility of power grid distance protection.

[0142] Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530, and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call logic instructions in the memory 530 to execute a segmented intelligent coordination setting method for power grid distance protection. This method includes: in response to the arrival of the execution start time t of a periodic task, collecting and acquiring power system state information at time t; wherein the power system state information includes a power system bus voltage vector state information set, a power system bus current vector state information set, and a power system circuit breaker state information set; traversing a value reward experience pool, determining the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, making a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and distributing the decision strategy to the protection equipment at the corresponding location so that the protection equipment can control the circuit breaker operation according to the decision strategy; wherein the decision strategy includes a segmented distance protection impedance setting value and a segmented distance protection operation time limit setting value.

[0143] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0144] On the other hand, embodiments of the present invention also provide a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the segmented intelligent coordination setting method for power grid distance protection provided by the above methods. The method includes: in response to the arrival of the execution start time t of a periodic task, collecting and acquiring power system state information at time t; wherein, the power system state information includes a power system bus voltage vector state information set, a power system bus current vector state information set, and a power system circuit breaker state information set; traversing a value reward experience pool, determining the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, making a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and distributing the decision strategy to the protection device at the corresponding location so that the protection device can control the circuit breaker to operate according to the decision strategy; wherein, the decision strategy includes a segmented distance protection impedance setting value and a segmented distance protection operation time limit setting value.

[0145] In another aspect, embodiments of the present invention also provide a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the segmented intelligent coordination setting method for power grid distance protection provided by the above methods. The method includes: in response to the arrival of the execution start time t of a periodic task, collecting and acquiring power system state information at time t; wherein the power system state information includes a power system bus voltage vector state information set, a power system bus current vector state information set, and a power system circuit breaker state information set; traversing a value reward experience pool, determining the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, making a decision based on the historical optimal strategy and a random decision algorithm to obtain the decision strategy for the current period, and distributing the decision strategy to the protection device at the corresponding location so that the protection device can control the circuit breaker to operate according to the decision strategy; wherein the decision strategy includes a segmented distance protection impedance setting value and a segmented distance protection operation time limit setting value.

[0146] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0147] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A segmented intelligent coordination setting method for power grid distance protection, characterized in that, include: First state information acquisition step: In response to the arrival of the execution start time t of the periodic task, the power system state information at time t is acquired and obtained; wherein, the power system state information includes the power system bus voltage vector state information set, the power system bus current vector state information set, and the power system circuit breaker state information set; Decision-making steps: Traverse the value reward experience pool, determine the corresponding decision strategy as the historical optimal strategy based on the optimal value recorded in the value reward experience pool, make a decision based on the historical optimal strategy and a random decision-making algorithm to obtain the decision strategy for the current period, and send the decision strategy to the protection device at the corresponding location so that the protection device can control the circuit breaker to operate according to the decision strategy; wherein, the decision strategy includes the segmented distance protection impedance setting value and the segmented distance protection operation time setting value.

2. The segmented intelligent coordination setting method for power grid distance protection according to claim 1, characterized in that, Following the decision-making step, the method further includes: The second state information acquisition step: In response to the arrival of the execution start time t+1 of the next cycle, the power system state information at time t+1 is acquired and obtained; Reward value calculation steps: Obtain the reward value of the current period's distance protection segmented impedance setting, the reward value of the distance protection segmented operating time setting, and the reward value of the circuit breaker status; calculate the reward value of the current period based on the reward value of the distance protection segmented impedance setting, the reward value of the distance protection segmented operating time setting, and the reward value of the circuit breaker status. Value calculation steps: Calculate the value of executing the decision strategy after the current period update based on the power system state information at time t; Storage steps: Store the power system state information at time t, the decision strategy for the current period, the power system state information at time t+1, the reward value for the current period, and the updated value for the current period into the value reward experience pool.

3. The segmented intelligent coordination setting method for power grid distance protection according to claim 2, characterized in that, Before the first state information acquisition step, the method further includes a training step, which includes executing the first state information acquisition step, the decision-making step, the second state information acquisition step, the reward value calculation step, and the value calculation step a preset number of times.

4. The segmented intelligent coordination setting method for power grid distance protection according to claim 1, characterized in that, The calculation of the value of the decision strategy after the current period update based on the power system state information at time t includes: The value of executing the decision strategy based on the power system state information at time t after the current cycle update is calculated based on the value before the current cycle update, the reward value of executing the historical best decision based on the power system state information at time t+1, the reward value of the current cycle, the self-learning rate, and the reward decay rate.

5. The segmented intelligent coordination setting method for power grid distance protection according to claim 1, characterized in that, The power system bus voltage vector state information set is represented as follows: in, Let be the set of bus voltage vector state information of the power system at time t. Let be the voltage vector value across the circuit breaker numbered j on the busbar numbered i; The power system bus current vector state information set is represented as follows: in, Let be the set of bus current vector state information of the power system at time t. Let J be the vector value of the current passing through the circuit breaker numbered j on bus i. The power system circuit breaker status information set is represented as follows: Among them, S (t) Let t be the set of power system circuit breaker status information. These are the operation status information of the first, second, and third stage protection of the distance protection of the circuit breaker numbered j on busbar numbered i.

6. The segmented intelligent coordination setting method for power grid distance protection according to claim 1, characterized in that, The segmented distance protection impedance setting value is expressed as follows: Among them, Z (t) Let be the segmented distance protection impedance setting value of the circuit breaker on each branch of each busbar at time t. These are the impedance setting values ​​for the first, second, and third stages of the distance protection of the circuit breaker numbered j on busbar numbered i, respectively. The segmented distance protection action time limit setting value is expressed as follows: Among them, T (t) Let t be the segmented distance protection operating time setting value for the circuit breakers on each branch of each busbar at time t. These are the time limit settings for the first, second, and third stages of the distance protection of the circuit breaker numbered j on busbar numbered i.

7. The segmented intelligent coordination setting method for power grid distance protection according to claim 6, characterized in that, The reward value for the period to which time t belongs is expressed as: R (t) =R Z(t) +R T(t) +R bre(t) Among them, R (t) R is the reward value for the period to which time t belongs. Z(t) R is the distance protection segmented impedance setting value and bonus value for the period to which time t belongs. T(t) R is the distance protection segmented action time limit setting value for the period to which time t belongs, combined with the reward value. bre(t) The circuit breaker state reward value for the period to which time t belongs.

8. The segmented intelligent coordination setting method for power grid distance protection according to claim 7, characterized in that, The distance protection segmented impedance setting value of the period to which time t belongs, combined with the bonus value R Z(t) Represented as: in, This indicates the reward for the distance protection impedance setting value of the first stage of the cycle at time t. This indicates the reward for the second-stage distance protection impedance setting value within the cycle at time t. This indicates the reward for the three-stage distance protection impedance setting value of the cycle at time t; Among them, a Ⅰ b Ⅰ L is the first-stage distance protection reward coefficient. i Let z be the length of the busbar numbered i. i Let i be the positive sequence impedance per unit length of the busbar numbered i; Among them, a Ⅱ b Ⅱ c Ⅱ d Ⅱ This refers to the second-stage distance protection reward coefficient. The impedance setting value that is coordinated with the second-stage impedance setting value on the next busbar numbered i; Among them, a Ⅲ b Ⅲ c Ⅲ d Ⅲ This refers to the second-stage distance protection reward coefficient. This is the impedance setting value that coordinates with the three-stage impedance setting value on the next busbar numbered i.

9. The segmented intelligent coordination setting method for power grid distance protection according to claim 7, characterized in that, The distance protection segmented action time limit setting value of the period to which time t belongs, combined with the bonus value R. T(t) Represented as: Where k1, k2, k3, k4, and k5 are the reward coefficients for the three-stage distance protection time limit. Δt represents the three-stage operating time limit setting value of the distance protection for circuit breaker j on the next busbar numbered i, and Δt represents the three-stage operating time limit coordination range of the distance protection.

10. The segmented intelligent coordination setting method for power grid distance protection according to claim 7, characterized in that, The reward for the circuit breaker state in the cycle at time t is R. bre(t) When the circuit breaker at this level trips, and when the trip signal is sent by the first-stage distance protection, R bre(t) R is a positive value when the trip signal is sent by a two-stage or three-stage distance protection. bre(t) The value is negative, and the trip signal is generated by R when the second-stage distance protection is activated. bre(t) The value is greater than the trip signal transmitted by the three-stage distance protection R. bre(t) value; When a cascading circuit breaker trips, and the trip signal is transmitted by a one-stage, two-stage, or three-stage distance protection, R bre(t) The value is negative, and the trip signal is generated by R when the first-stage distance protection is activated. bre(t) The value is greater than the trip signal transmitted by the two-stage distance protection R. bre(t) The trip signal is generated by R during the second-stage distance protection signal transmission. bre(t) The value is greater than the trip signal transmitted by the three-stage distance protection R. bre(t) value.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the segmented intelligent coordination setting method for power grid distance protection as described in any one of claims 1 to 10.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the segmented intelligent coordination setting method for power grid distance protection as described in any one of claims 1 to 10.

13. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the segmented intelligent coordination setting method for power grid distance protection as described in any one of claims 1 to 10.

Images