Method for formulating power flow early warning limit of power transmission channel of power system

Abstract

A power system power transmission channel power flow early warning limit making method belongs to the technical field of power system dispatching. The method solves the problems of strong experience dependence, difficulty in control of randomness and low early warning accuracy in formulation of a power transmission channel power flow early warning limit of an existing power system, divides discrete margin states according to a continuous out-of-limit risk threshold value of a power transmission channel section, collects historical operation data of an actual power transmission channel section, and improves the early warning accuracy. Counting the occurrence frequency of each margin state and the transition frequency between the states, calculating a one-step transition probability and constructing a one-step state transition matrix; determining the state probability vector of the section margin at the current moment, and calculating the probability of each margin state in the future preset time based on the one-step state transition matrix; and screening the alternative probabilities meeting the risk control requirements, and calculating the section power flow early warning limit by using the minimum value in the margin levels corresponding to the screened alternative probabilities. The method is suitable for power system power transmission channel tide early warning.

Description

Technical Field

[0001] This invention belongs to the field of power system dispatching technology. Background Technology

[0002] The safety and stability of power grid operating sections are crucial to the entire power grid system. With the increasing proportion of new energy sources in the power grid, their intermittency, randomness, and volatility pose numerous challenges to grid operation, making the assessment and control of the safety margin of power grid operating sections more complex. Currently, traditional methods for assessing the safety margin of power grid operating sections have significant shortcomings:

[0003] The limitations of empirical methods: Existing methods mostly rely on experience to reserve safety margins for power grid operating sections, depending on dispatchers' manual experience and determining the safety margin based on statistically analyzed power flow overload capacity of power grid operating sections. However, differences in the professional capabilities and experience of different dispatchers lead to a lack of consistency in control results, making it difficult to reflect the overall trend of changing demand for safety margins at power grid operating sections. Moreover, the selection of existing key influencing factors is largely based on manual experience, lacking reliable standards, making safety margin assessments unreliable. In addition, this assessment method is a static assessment, failing to fully consider changes in power grid operating characteristics under different scenarios, and is prone to the problem of mismatch between safety margin reservations and power grid operating conditions.

[0004] Existing methods for assessing the safety threshold of power transmission sections in power grid systems also have problems. For example, calculation methods based on power limit modeling of power transmission sections ignore the dynamic characteristics of the power grid system and cannot reflect the impact of various nonlinear and dynamic factors in actual operation; methods based on power flow calculations involve huge computational loads and are not suitable for real-time calculations and rapid decision-making. In UHV AC / DC hybrid receiving-end power grids, although existing frequency emergency coordination control decisions can eliminate errors caused by power frequency mismatch to a certain extent, they cannot effectively suppress under-control or over-control at other constraint levels caused by uncertainties, which can easily induce power limit exceeding problems at transmission sections. Summary of the Invention

[0005] This invention aims to address the problems of strong reliance on experience, difficulty in controlling randomness, and low accuracy in setting power flow warning limits for transmission channels in existing power systems. A method for setting power flow warning limits for transmission channels in power systems is provided.

[0006] The method for setting power flow early warning limits for power system transmission channels according to the present invention includes:

[0007] Step 1: Based on the continuous over-limit risk threshold of the transmission channel section under different margin levels, divide the margin states into discrete states, including at least a safe state, a warning state, and an over-limit state.

[0008] Step 2: Collect historical operating data of the actual transmission channel section, count the number of occurrences of each margin state and the number of transitions between states, calculate the one-step transition probability and construct the one-step state transition matrix;

[0009] Step 3: Determine the state probability vectors at different margin levels at the current time, and based on the one-step state transition matrix, obtain the state probabilities at different margin levels within a future preset time period;

[0010] Step 4: From the state probabilities under different margin levels within a preset future time, within the preset margin level range, take the state probability of exceeding the limit for n consecutive minutes under each margin level as the candidate probability, screen the candidate probabilities that meet the risk control requirements, and use the minimum value of the margin level corresponding to the screened candidate probabilities to calculate and obtain the cross-sectional power flow warning limit.

[0011] When the power system is mainly powered by thermal power, n=8; when the power system is mainly powered by wind power, n=5.

[0012] Furthermore, in this invention, the specific method for dividing the discrete margin states in step one based on the continuous over-limit risk threshold of the transmission channel section is as follows:

[0013] Set the sustained exceedance limit to 5 minutes, within a given margin level. Below, the corresponding warning limit At that time, the margin state is divided into 8 states, among which, The stable operating limits are as follows:

[0014] state : Safe condition, the power flow value of the cross-section is greater than 0 and less than 0. ;

[0015] state Warning status: The power flow value at the cross-section is greater than or equal to... Less than ;

[0016] state : Over-limit state, first over-limit, the power flow value of the cross section is greater than or equal to Lasts for 1 minute;

[0017] state Two consecutive instances of exceeding the limit for two consecutive minutes, with the power flow value at the cross-section being greater than or equal to... Lasting 2 minutes;

[0018] state Three consecutive instances of exceeding the limit for three consecutive minutes, with the power flow value at the cross-section being greater than or equal to... Lasting 3 minutes;

[0019] state Four consecutive instances of exceeding the limit for four consecutive minutes, with the power flow value at the cross-section being greater than or equal to... Lasting 4 minutes;

[0020] state Five consecutive instances of exceeding the limit for five consecutive minutes, with the power flow value at the cross-section being greater than or equal to... Lasting 5 minutes;

[0021] state The state of exceeding the limit more than five times consecutively, with the power flow value of the cross section being greater than or equal to... This should last for more than 5 minutes.

[0022] Furthermore, in this invention, in step two, the one-step state transition matrix is: :

[0023]

[0024] In the formula, the 0 elements in the matrix represent a state transition probability of 0, or in other words, no transition will occur. , , , , , , , and Representing states respectively to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state Transition probabilities and states to state The transition probability; , , , , , , , , , , , , , , and Representing states respectively to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state Transition probabilities and states to state The transition probability.

[0025] Furthermore, in this invention, in step three, the expression for the probability of different margin levels and states within a preset future time period is as follows:

[0026]

[0027]

[0028]

[0029] in, This represents the state probability vector at the initial moment. Indicates the margin level The state transition matrix after time T is the margin level. Next step state transition matrix power of T, Indicates the margin level Down After time P, the state is P i The probability is given by n, where n represents the total number of states. Indicates the margin level The probability of the cross-section exceeding the limit after time T. Indicates the margin level State probability after time T The i-th value.

[0030] Furthermore, in this invention, in step four, the calculation formula for obtaining the cross-sectional tidal flow early warning limit is as follows:

[0031]

[0032] in, This represents the limit value for stable operation. This indicates the tidal current warning limit for cross-sections. The minimum margin level corresponding to the alternative probability to meet risk control requirements.

[0033] This invention utilizes a Markov chain prediction model to address the issue of highly random power loads and the susceptibility to cross-limit violations. It proposes a method for obtaining critical cross-limit violation warning values ​​under different probabilities and operating modes. This value can alert dispatchers that when a cross-section reaches a certain value under the current operating mode, a limit-over-limit accident is likely to occur, requiring advance preparation for unit control. It can be used for dispatching auxiliary decision-making and safety early warning. Attached Figure Description

[0034] Figure 1 This is a flowchart of the method described in this invention;

[0035] Figure 2 This is a schematic diagram of state transitions for 8 state sections;

[0036] Figure 3 This is a probability distribution chart showing the duration of exceeding the 5-minute limit within a 15-minute timeframe in the first quarter.

[0037] Figure 4 This is a schematic diagram of state transitions for 5 state sections. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0039] Specific implementation method one: Refer to Figure 1 and Figure 2 This embodiment specifically describes the method for setting power system transmission channel power flow early warning limits, which includes:

[0040] Step 1: Based on the continuous over-limit risk threshold of the transmission channel section under different margin levels, divide the margin states into discrete states, including at least a safe state, a warning state, and an over-limit state.

[0041] Step 2: Collect historical operating data of the actual transmission channel section, count the number of occurrences of each margin state and the number of transitions between states, calculate the one-step transition probability and construct the one-step state transition matrix;

[0042] Step 3: Determine the state probability vectors at different margin levels at the current time, and based on the one-step state transition matrix, obtain the state probabilities at different margin levels within a future preset time period;

[0043] Step 4: From the state probabilities under different margin levels within a preset future time, within the preset margin level range, take the state probability of exceeding the limit for n consecutive minutes under each margin level as the candidate probability, screen the candidate probabilities that meet the risk control requirements, and use the minimum value of the margin level corresponding to the screened candidate probabilities to calculate and obtain the cross-sectional power flow warning limit.

[0044] When the power system is mainly powered by thermal power, n=8; when the power system is mainly powered by wind power, n=5.

[0045] Furthermore, in this embodiment, the specific method for dividing the discrete margin states in step one based on the continuous over-limit risk threshold of the transmission channel section is as follows:

[0046] Set the sustained exceedance limit to 5 minutes, within a given margin level. Below, the corresponding warning limit At that time, the margin state is divided into 8 states, among which, The stable operating limits are as follows:

[0047] state : Safe condition, the power flow value of the cross-section is greater than 0 and less than 0. ;

[0048] state Warning status: The power flow value at the cross-section is greater than or equal to... Less than ;

[0049] state : Over-limit state, first over-limit, the power flow value of the cross section is greater than or equal to Lasts for 1 minute;

[0050] state Two consecutive instances of exceeding the limit for two consecutive minutes, with the power flow value at the cross-section being greater than or equal to... Lasting 2 minutes;

[0051] state Three consecutive instances of exceeding the limit for three consecutive minutes, with the power flow value at the cross-section being greater than or equal to... Lasting 3 minutes;

[0052] state Four consecutive instances of exceeding the limit for four consecutive minutes, with the power flow value at the cross-section being greater than or equal to... Lasting 4 minutes;

[0053] state Five consecutive instances of exceeding the limit for five consecutive minutes, with the power flow value at the cross-section being greater than or equal to... Lasting 5 minutes;

[0054] state The state of exceeding the limit more than five times consecutively, with the power flow value of the cross section being greater than or equal to... This should last for more than 5 minutes.

[0055] Furthermore, in this embodiment, in step two, the one-step state transition matrix is: :

[0056]

[0057] In the formula, the 0 elements in the matrix represent a state transition probability of 0, or in other words, no transition will occur. , , , , , , , and Representing states respectively to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state Transition probabilities and states to state The transition probability; , , , , , , , , , , , , , , and Representing states respectively to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state The transition probability, state to state Transition probabilities and states to state The transition probability.

[0058] Furthermore, in this embodiment, and more specifically, in step three of this invention, the expression for the probability of different margin levels and states within a preset future time period is as follows:

[0059]

[0060]

[0061]

[0062] in, This represents the state probability vector at the initial moment. Indicates the margin level The state transition matrix after time T is the margin level. Next step state transition matrix power of T, Indicates the margin level Down After time P, the state is P i The probability is given by n, where n represents the total number of states. Indicates the margin level The probability of the cross-section exceeding the limit after time T. Indicates the margin level State probability after time T The i-th value.

[0063] Furthermore, in this embodiment, in step four, the calculation formula for obtaining the cross-sectional tidal flow warning limit is as follows:

[0064]

[0065] in, This represents the limit value for stable operation. This indicates the tidal current warning limit for cross-sections. The minimum margin level corresponding to the alternative probability to meet risk control requirements.

[0066] Because the output and load of new energy generator units vary with time periods, the actual power output of a section also exhibits a degree of randomness, and consequently, the section margin, which is closely related to this, also contains randomness. Setting a safety margin can, to some extent, characterize the probability that the current section power state will enter different states over a future period, primarily reflecting the probability of the section operating beyond its stability limits and the probability that such operation will persist for a certain period. Therefore, the entire section margin range can be divided into... a range (or (Class), and if each interval's margin range is considered as a margin state, then there are a total of A certain level of sufficiency.

[0067] At each time step, the margin is in a certain margin state. The margin state at the next time step depends only on the margin state at the current time step and is not related to the margin state at past time steps.

[0068] The one-step transition probability of a margin state depends only on the time difference and is independent of the starting time.

[0069] Based on the above three points, the cross-sectional margin at different times is a stochastic process, satisfying the three conditions required for a Markov chain prediction model: randomness, lack of aftereffect, and stability of the transition matrix. Considering the changing pattern of the cross-sectional margin, the setting of the safety margin should also follow a similar pattern. Therefore, we consider using a Markov chain model to predict the cross-sectional margin at the next time step.

[0070] To provide a reasonable safety margin (safety limit) for cross-section operation, a probabilistic prediction of the margin state at the next moment is first given. Then, based on the current state of the cross-section margin, the state of the cross-section margin in the future period is determined, thereby supporting the dispatching department to make adjustments.

[0071] The basic principle of Markov chain prediction model is to use the state probability vector and state transition matrix at a certain point in the past to infer the probability of the predicted object being in different states at a certain point in the future.

[0072] remember If the cross-sectional margin state is a stochastic process that varies with different times, then its state at the t-th moment... The value at each moment It is a one-dimensional discrete random variable, and all its possible values ​​are . The type, that is, the cross-sectional margin state, may reach Any one of the states, respectively .remember The distribution is ; It is a length of Let be the probability vector. The component is Then it means Take the first The probability of each margin state value. Clearly, for any time... , It should meet the following requirements:

[0073] (1)

[0074] (2)

[0075] Let matrix The one-step probability transition matrix has the following form:

[0076] (3)

[0077] If we denote its elements as , .but This indicates that the margin state is at the current moment. Under the condition that, the margin state at the next time step is The probability of this. This is a conditional probability, that is... Clearly, based on the properties of conditional probability, the probability transition matrix... It should meet the following requirements:

[0078] (4)

[0079] (5)

[0080] Based on this formula, the first... The system predicts the probability of a process at each time step being in various margin states. The system can then perform corresponding actions based on the margin state corresponding to the maximum predicted probability.

[0081] According to the Markov chain prediction model, to predict the... The probability of a margin state at a given moment needs to be calculated based on historical data, either at the current moment or at a specific past moment. probability distribution of margin state One-step state transition matrix .

[0082] According to the definition of a one-step transition probability matrix, each element in the matrix... It is a conditional probability. (For calculation...) First, calculate the margin status over all past time periods. Number of times Secondly, statistics are based on the margin status. Transition to margin state Number of times ,but:

[0083] (6)

[0084] Using a certain moment in the past Let the margin state be the initial state. =[0,……,0,1,0,……,0] ,Right now Only the first The element in one position is 1, and the elements in the other positions are all 0.

[0085] It can be calculated that:

[0086] (7)

[0087] The limit time for considering cross-sectional over-limit is At a given margin level Below, the corresponding warning limit At that time, the margin state is divided into N states, which are:

[0088] : Safe status This refers to being within a sufficient margin of safety, i.e., an interval. All the margin values ​​within this range belong to this state;

[0089] Warning status This refers to a margin value that is relatively small but not yet exhausted, i.e., a range. All the margin values ​​within this range belong to this state;

[0090] The "over-limit state" refers to the margin value being exhausted. This can be further divided into two cases: First, the margin value in the two preceding and following minutes is not exceeded, but only in the middle minute is it exceeded. In this case, the state of the middle minute must be... Secondly, if there was no violation in the previous minute but a violation occurred in the next minute, this is equivalent to two consecutive violations, or two consecutive minutes. However, this "consecutive" refers to the next minute; for the intermediate minute, it's the first violation. Therefore, the state of the intermediate minute is... ;

[0091] The count for two consecutive violations within two minutes is calculated from the first violation in the previous minute; that is, this state always starts from... I've recovered from this state;

[0092] And so on, :continuous Subsequent, consecutive The minute limit exceeded state occurs when there have been consecutive minutes exceeding the limit. It's calculated based on exceeding the minute limit, meaning this state must originate from... I've recovered from this state;

[0093] : continuously exceeding The state of exceeding the limit in minutes.

[0094] Based on the above definition For each state, based on the margin range corresponding to the safe state and the warning state, the margin is assigned to the corresponding state value. For each state, the cross-sectional margin at each time step can be assigned a corresponding state value using the difference method. , indicating continuity First, identify all instances of exceeding the limit. Then, the first-order difference method is used. Time and First-order difference Equal to:

[0095] (8)

[0096] like If the value equals 1, it indicates that these are two consecutive moments of exceeding the limit. Therefore, if... The state at any given moment is ,but The state at any given moment should be .

[0097] Once all states are determined, the value of each transition probability in the probability matrix can be calculated.

[0098] Set the sustained exceedance limit to 5 minutes, within a given margin level. Below, the corresponding warning limit At that time, the margin state is divided into 8 states, namely:

[0099] : Safe status This indicates that the margin value is sufficient, and the tidal current value of the cross-section is greater than 0 and less than 0. ;

[0100] Warning status The tidal current value of the cross section is greater than Less than This indicates a small margin value, but it has not yet been exhausted.

[0101] The "over-limit" state refers to the margin value being exhausted. This can be further divided into two cases: First, the margin value in the two preceding and following minutes is not exceeded, but only the margin value in the middle minute exceeds the limit. In this case, the state of the middle minute must be... Secondly, if there was no violation in the previous minute but a violation occurred in the next minute, this is equivalent to two consecutive violations, or two consecutive minutes. However, this "consecutive" refers to the next minute; for the intermediate minute, it's the first violation. Therefore, the state of the intermediate minute is... The tidal current value of the cross section is greater than or equal to Lasts for 1 minute;

[0102] The count for two consecutive violations within two minutes is calculated from the first violation in the previous minute; that is, this state always starts from... The current flow value at the cross-section is greater than or equal to the current flow value at the cross-section. Lasting 2 minutes;

[0103] The three consecutive violations, or three consecutive minutes of exceeding the limit, are calculated based on a previous two consecutive minutes of exceeding the limit. In other words, this state is definitely triggered from... The current flow value at the cross-section is greater than or equal to the current flow value at the cross-section. Lasting 3 minutes;

[0104] The four consecutive instances of exceeding the limit for four consecutive minutes are calculated based on a previous three consecutive minutes of exceeding the limit. In other words, this state is definitely triggered from... The current flow value at the cross-section is greater than or equal to the current flow value at the cross-section. Lasting 4 minutes;

[0105] The five consecutive instances of exceeding the limit for five consecutive minutes are calculated based on a previous four consecutive minutes of exceeding the limit. In other words, this state is definitely triggered from... The current flow value at the cross-section is greater than or equal to the current flow value at the cross-section. Lasting 5 minutes;

[0106] The state of exceeding the limit more than five times consecutively, with the power flow value of the cross section being greater than or equal to... This should last for more than 5 minutes.

[0107] Based on the eight states defined above, the probability transition diagram between states is shown in Figure 2. Based on this state diagram, the elements of the probability transition matrix are as follows:

[0108] (9)

[0109] The 0 element in the matrix represents a state transition probability of 0, or in other words, no transition will occur.

[0110] Based on the classification of states, it can be seen that if the current moment is in a safe state... That is, the margin value is sufficient and is in a state of... If so, the margin value at the next time step may still be in a safe state. It's also possible that due to randomness, it jumped to the warning state. This means the margin value is small, but it can still operate safely. It's also possible that it will directly jump to an over-limit state. That is, it exceeds the safety limits and is impossible to occur. Yes, because These states are This is an accumulation based on the current state. If the current state is a warning state... If it does not pass through, the next moment may still be in a warning state, or it may be in an over-limit state, or it may return to a safe state.

[0111] If the current state is out of bounds, then the next moment has several possibilities, such as being in a state of... The current state indicates that the limit was exceeded just one minute ago. Therefore, the next possible state is continued exceeding of the limit, i.e., the current state... It may also revert to a warning state. It is also possible to jump directly to a safe state. However, if the limit is continuously exceeded, it is actually... And it will not reappear Therefore, the transition probability .

[0112] If the current time is in a state of exceeding the limit by two minutes, that is If the limit is exceeded again in the next moment, then jump to the next state. It is also possible that after adjustments, it will revert to a warning state. It is also possible to return to a safe state. But it is impossible to remain in that state. It is also impossible to be in Etc.

[0113] Similarly, the probabilities of other state transitions can be analyzed, so All are 0. But Different, because The status indicates a serious timeout exceeding limits. The next possible outcome is a return to a safe state after adjustments, a warning state, or continued timeout status. Not equal to 0. State assignment and transition probability calculation utilize the transition probability matrix and the current state probability vector. Calculate the state probability PAI at a future time.

[0114] Based on the eight states defined above, and according to the cross-sectional margin intervals corresponding to the safe and warning states, it is easy to assign these cross-sectional margins to their corresponding state values. For the remaining eight states, the cross-sectional margin value at each time step can be assigned to its corresponding state value using a difference method. Taking the state as an example, to determine two consecutive violations, we first need to find all the times when the limit is exceeded. Then, the first-order difference method is used. Time and First-order difference Equal to:

[0115] (10)

[0116] like If the value equals 1, it indicates that these are two consecutive moments of exceeding the limit. Therefore, if... The state at any given moment is ,but The state at any given moment should be In this way, all states can be determined and the value of each transition probability in the probability matrix can be calculated.

[0117] The data for each year is divided into seasons, and Markov chains are used to predict the cross-sectional data of the same quarter for all years. The probability of being in various states after minutes. Two probabilities are of particular interest:

[0118] First, if the current situation is under alert. Then the state probability vector at this moment Take as ,So After minutes, it will be in an over-limit state (including) ~ The probability of multiple out-of-limit states The calculation formula is:

[0119] (11)

[0120] Since there are six possible out-of-limit states, from state P3 to state P8, the probability of being in an out-of-limit state is the sum of the probabilities of being in any of these six out-of-limit states. If the current timeframe is in a warning state... , The probability vector of a state that is out of bounds at a given time and has persisted for more than 5 minutes. Take as .So If the condition exceeds the limit after 5 minutes and continues for more than 5 minutes, it is considered to be in a state of exceeding the limit. probability of a state The calculation formula is:

[0121] (12)

[0122] Safety margin analysis for Province A in the fourth quarter; safety margin analysis for cross-sections in the first quarter. Figure 3 This represents the probability distribution of a cross-section remaining in an over-limit state for more than 5 minutes within a certain period of time, under different margin levels in the first quarter. The probability of being over-limit at any given moment and lasting for more than 5 minutes should be less than 0.05 (statistical analysis of actual data shows that this ratio is higher than 0.05, therefore, using 0.05 as the standard for selecting a safety margin is more reasonable). The safety margin for the first quarter should be 6%, which can ensure that the probability of being over-limit at any given moment within 60 minutes and lasting for more than 5 minutes is less than 0.05, while also ensuring maximum utilization of the cross-section.

[0123] Similarly, it can be predicted that the probability of exceeding the limit and lasting for more than 5 minutes in the second, third and fourth quarters is less than 0.05.

[0124] For the Qunfanglin section in Heilongjiang Province, the stable operation limit of the section. The safety limits for each quarter are as follows: MW.

[0125] Table 1 Safety Limits for the Fourth Quarter

[0126]

[0127] Considering the safety margin setting for new energy vehicles:

[0128] Considering the characteristics of renewable energy generation and its highly sensitive regulation, when renewable energy generation is substantial, the sustained exceedance limit should be set to be less than that mentioned above; here, it is set to 2 minutes. Therefore, within a given margin level... Below, the cross-section is adjusted to five states according to time.

[0129] Based on the classification of states, it can be seen that if the current moment is in a safe state... If the margin is sufficient, then the margin may still be in a safe state at the next time step. It's also possible that due to randomness, it jumped to the warning state. This means the margin value is small, but it can still operate safely. It's also possible that it will directly jump to an over-limit state. That is, it exceeds the safety limits and is impossible to occur. Yes, because These states are This is an accumulation based on the current state. If the current state is a warning state... If it does not pass through, the next moment may still be in a warning state, or it may be in an over-limit state, or it may return to a safe state.

[0130] If the current state is out of bounds, then the next moment has several possibilities, such as being in a state of... The current state indicates that the limit was exceeded just one minute ago. Therefore, the next possible state is continued exceeding of the limit, i.e., the current state... It may also revert to a warning state. It is also possible to jump directly to a safe state. However, if the limit is continuously exceeded, it is actually... And it will not reappear Therefore, the transition probability .

[0131] If the current time is in a state of exceeding the limit by two minutes, that is If the limit is exceeded again in the next moment, then jump to the next state. It is also possible that after adjustments, it will revert to a warning state. It is also possible to return to a safe state. But it is impossible to remain in that state. It is also impossible to be in state.

[0132] Similarly, the probabilities of other state transitions can be analyzed, so All are 0. But Different, because The status indicates a serious timeout exceeding limits. The next possible outcome is a return to a safe state after adjustments, a warning state, or continued timeout status. Not equal to 0.

[0133] State assignment and transition probability calculation:

[0134] Based on the five states defined above, and according to the cross-sectional margin intervals corresponding to the safe and warning states, it is also easy to attach the corresponding state values ​​to these cross-sectional margins. The difference method can be used to attach the corresponding state value to the cross-sectional margin value at each time step.

[0135] The data for each year is divided into seasons, and Markov chains are used to predict the cross-sectional data of the same quarter for all years. The probability of being in various states after minutes. Two probabilities are of particular interest:

[0136] First, if the current situation is under alert. Then the state probability vector at this moment Take as ,So After minutes, it will be in an over-limit state (including) ~ Multiple out-of-limit states, such as Figure 4 The probability shown) The calculation method is as follows:

[0137] (13)

[0138] in, This represents the state probability vector at the initial moment. Indicates the margin level The state transition matrix after time T is the margin level. Next step state transition matrix power of T, Indicates the margin level The probability of the cross-section exceeding the limit after time T. Indicates the margin level State probability after time T The i-th value.

[0139] Second, if the current time is under warning status , The probability that the state is in an out-of-limit state for more than 2 minutes at the specified time, and the probability vector of the state at this moment. Take as .So If the state exceeds the limit after 2 minutes and continues for more than 2 minutes, it is considered to be in a state of exceeding the limit. probability of a state The calculation formula is:

[0140] (14)

[0141] If only the output of the wind turbine is controlled to achieve cross-sectional power adjustment, then the stable operating limit of the Qunfanglin cross-section, i.e. =3500MW, and the safety limits for each quarter are as shown in the table below.

[0142] Table 2 Safety Limits for the Fourth Quarter under Large-Scale Wind Power Generation

[0143]

[0144] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A method for setting a pre-alarm limit of power flow in a power system transmission channel, characterized in that, Comprising: Comprising: Step one, according to the risk threshold of the continuous over-limit of the section of the power transmission channel under different margin levels, the discrete margin state is divided, and the margin state at least includes the safe state, the warning state and the over-limit state; Step two, the actual power transmission channel section historical operation data is collected, the occurrence times of each margin state and the transfer times between states are counted, the one-step transfer probability is calculated and the one-step state transition matrix is constructed; Step three, the state probability vector under different margin levels at the current time is determined, and the state probability under different margin levels in the future preset time is obtained based on the one-step state transition matrix; Step four, from the state probability under different margin levels in the future preset time, in the preset margin level range, the state probability of continuous n minutes over-limit under each margin level is taken as the candidate probability, the candidate probability meeting the risk control requirement is screened, and the minimum value in the corresponding margin level of the screened candidate probability is used to calculate and obtain the section flow early warning limit; Wherein, when the power system is mainly based on thermal power generation, n=8, and when the power system is mainly based on wind power generation, n=5.

2. The method of claim 1, wherein, In step one, the specific method for dividing the discrete margin state according to the continuous over-limit risk threshold of the section of the power transmission channel is: Set the duration out-of-limit to 5 minutes, at the given margin level Next, the corresponding pre-warning limit The margin state is divided into 8 states, where, Indicates the stable operation limit value, respectively: State : safe state, the flow value of the section is greater than 0 and less than ; State : warning state, the cross-sectional tidal current value is greater than or equal to less than ; State : Out of limit state, first time out of limit, the tidal flow value of the section is greater than or equal to , for 1 minute; State : two consecutive, two-minute out-of-limit state, the tidal current value of the section is greater than or equal to , for 2 minutes; State : three consecutive, three-minute out-of-limit state, the tidal current value of the section is greater than or equal to , for 3 minutes; State : four consecutive, four minute over limit state, the tidal current value of the section is greater than or equal to , for 4 minutes; State : five consecutive times, five consecutive minutes of limit state, the value of the tidal current is greater than or equal to , for 5 minutes; State : State of continuous over-limit for more than five times, the tidal current value of the section is greater than or equal to , lasting for more than 5 minutes.

3. The method of claim 2, wherein the method further comprises: In step two, the one-step state transition matrix is : In the formula, the 0 element in the matrix indicates that the state transition probability is 0, or in other words, no transition, , , , , , , , and respectively represent the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state and the transition probability of state to state ; , , , , , , , , , , , , , , and respectively represent the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state , the transition probability of state to state transition probabilities, state to state transition probabilities, state to state transition probabilities, state to state transition probabilities, state to state transition probabilities, state to state transition probabilities and state to state transition probabilities.

4. The method of claim 3, wherein the power system transmission channel flow warning limit setting method is characterized by, In step three, the expression of the state probability under different margin levels in the future preset time is: wherein, denotes the state probability vector at initial time, denotes the margin level the state transition matrix after T time, for margin level the next step state transition matrix T times of denotes the margin level after the probability of being in state P i after T time, n denotes the total number of states divided into n states, denotes the margin level the probability of exceeding the section after T time, denotes the margin level the i-th value of the state probability after T time.

5. The method of claim 4, wherein, In step four, the calculation formula for calculating the section flow early warning limit is: wherein, represents a stable operating limit value, represents a sectional current flow warning limit, is the minimum value of the margin level corresponding to the alternative probability to meet the risk control requirement.

Images