Modeling and risk assessment method and system for cascading failure of power system
By constructing a cascading failure simulation model and introducing a weighted random disturbance mechanism, the impact of network attacks on the power system is analyzed. This solves the problem of low assessment efficiency in existing technologies, enables efficient assessment and rapid response to power grid risks, and enhances the power grid's resilience.
Patent Information
- Application Number
- CN202410835193.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-26
- Publication Date
- 2024-11-15
AI Technical Summary
Existing methods for modeling and risk assessment of cascading failures in power systems have limited research on the impact of extreme events such as cyberattacks, and most methods employ electromechanical or electromagnetic transient simulations, resulting in low assessment efficiency.
A cascading failure simulation model is constructed, and network attacks are simulated by penetrating the IED system to determine line power flow overload, calculate node load failure rate, evaluate system load failure rate, and analyze the impact of different attack combinations on the system. A weighted random disturbance mechanism is introduced to quantitatively analyze power grid risks and propose multi-level collaborative protection measures.
It significantly improves the power grid's resilience and reliable operation, enabling rapid response to large-scale cascading failures, ensuring power supply to critical loads, reducing grid pressure, and enhancing system stability.
Smart Images

Figure CN118970840A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power system risk assessment, and in particular to a modeling and risk assessment method and system for power system cascading failures. Background Art
[0002] As the global demand and attention for renewable energy continue to increase, wind energy is developing rapidly as a clean and renewable energy form. Large-scale wind power access has become an important strategy for many countries to achieve energy transformation and reduce carbon emissions. However, the uncontrollability and volatility of wind energy have put forward new challenges to the reliability of the power system.
[0003] A cascading failure in a power system refers to a situation where a fault event spreads and expands rapidly in the power grid, causing multiple devices or lines to fail. In a series of major blackouts that have occurred internationally in recent years, the destruction of information systems has become an important factor and a new development trend, leading to large-scale power outages in the power grid. Key equipment such as highly automated and information-based control and protection in the power grid is widely used, and is highly vulnerable to disasters and cyber attacks, which may lead to equipment unavailability, control, and other consequences, which in turn affects the physical power grid. In extreme cases, the risk of system operation increases dramatically.
[0004] The failure forms of power systems under cyber attacks are fundamentally different from traditional equipment failures and human errors, and the evolution process of cascading failures is more complicated. It is necessary to carry out research on security assessment theories and methods for unconventional security incidents in power systems, build a simulation model of cascading failures in power grids, and propose simulation methods for cascading failures in power grids to provide strong support for ensuring the safe operation of power grids.
[0005] Current research focuses on transient, frequency, and voltage instability problems caused by conventional power grid faults, and less research has been conducted on the impact of extreme events such as cyber attacks on the power grid; electromechanical or electromagnetic transient simulation is mostly used, and the evaluation efficiency is low. The present invention proposes a modeling and risk assessment method for cascading failures in power systems, taking the control system load loss rate as the main control target, establishing a key factor model that causes system load loss, and proposing a risk assessment method for cascading failures in power grids, providing an important analysis method for power grids to cope with extreme risks. Summary of the invention
[0006] In view of the problems existing in the existing modeling and risk assessment methods of cascading failures in power systems, the present invention is proposed.
[0007] Therefore, the problem to be solved by the present invention is that there are few studies on the impact of extreme events such as cyber attacks on the power grid, and electromechanical or electromagnetic transient simulation is mostly used, which has low evaluation efficiency.
[0008] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0009] In the first aspect, an embodiment of the present invention provides a modeling and risk assessment method for cascading failures in a power system, which includes constructing a cascading failure simulation model, determining the initial fault conditions, judging whether there is a line flow overload, and calculating the system flow and voltage; calculating the node load loss rate based on the node voltage and line flow, and evaluating the system load loss rate; analyzing the impact of different attack combinations on the system load loss rate, and evaluating the risk of cascading failures.
[0010] As a preferred scheme of the modeling and risk assessment method of cascading failures of the power system described in the present invention, wherein: the initial fault condition is to attack the cascading failure simulation model constructed by collecting the complete topological structure data of the target power system, penetrate the IED system, implant malicious code, artificially trigger the line protection action to cause tripping, or inject erroneous line current / voltage data to deceive the protection algorithm to misjudge the line fault, or cause the IED to lose response through a denial of service attack, thereby failing to detect and protect normally, forming an initial fault.
[0011] As a preferred solution of the modeling and risk assessment method of power system cascading failures of the present invention, the judgment of line flow overload includes:
[0012] According to the topological structure and parameter calculation of the power grid, the node admittance matrix [B'] and [B"] are constructed as follows: for any two nodes i and j, if they are directly connected by a line, then:
[0013] Y ij =Y ji =-y ij
[0014] Y ii =Y jj =Σy ik (k!=i,j)
[0015] where y ij is the branch admittance, and all Y ij The elements form a matrix [Y]
[0016] [B'] = real part ([Y])
[0017] [B”] = imaginary part ([Y])
[0018] Furthermore, the power flow of each line is calculated according to the AC power flow calculation equation:
[0019] [P ij ]=[B ij ']*[θ i -θ j ]
[0020] [Q ij ]=[-B ij ”]*[V i 2 -V i *V j *cos(θ i -θ j )]
[0021] Among them: [P ij ] is the active power from node i to j, [Q ij ] is the reactive power from node i to j, θ is the node phase angle, and V is the node voltage amplitude; determine the power flow P of each line cur Whether it exceeds its thermal stability limit P max , P max is the thermal stability limit transmission power of the line, which is determined by the line parameters, P cur is the actual transmission power of the line, if P cur >P max , the line is considered to be an overloaded line.
[0022] As a preferred solution of the modeling and risk assessment method of power system cascading failures of the present invention, the tripping probability is calculated based on the overloaded line and is expressed as:
[0023]
[0024] Among them, P trip is the probability of line tripping per unit time, Δt is the simulation time step, T r is the line thermal constant, indicating the time required for the line to trip from overload; when P cur >P max When the line temperature rise rate is cur -P max ) is proportional to the temperature. When the temperature rises to a certain level, the line will be tripped by the protection action. trip It reflects the probability that the line will reach the tripping temperature from overload within the time Δt. According to the attack situation, the probability distribution function f(x) of the occurrence of hidden faults is set. For each device, a probability value p is randomly generated according to f(x) to determine whether p exceeds the set threshold. If so, it is considered that the device has a hidden fault and needs to be removed. According to the calculation results, the system topology is updated, and the node admittance matrix [B'] and [B”] are reconstructed. The AC power flow calculation equation is used to calculate the new line power flow and node voltage. Check whether there is an instability situation such as line power overload and node voltage exceeding the limit in the system. If not, it is considered that the chain fault is over and the node load loss rate is evaluated. If so, it returns to recalculate the line power flow.
[0025] As a preferred solution of the modeling and risk assessment method of the power system cascading failure described in the present invention, the calculation of the node load loss rate includes: obtaining the working state of the system after the cascading failure ends, further judging whether there is a Mankou power loss area, setting the node voltage allowable range, and judging each node voltage Vi: if Vi is within the allowable range, the node i belongs to the power supply area; if Vi exceeds the limit or is 0, the node i belongs to the power loss area; if the node i is completely isolated from the system, it is also regarded as a power loss; the adjacent power loss area is defined as the Mankou power loss area, and for each Mankou power loss area j, the load of all nodes therein is summed up:
[0026] L dropj =∑(PL i +jQL i )
[0027] Among them, L dropj represents the total load in the jth power failure area, PL i represents the active load of the ith node, QL i represents the reactive load of the i-th node, i belongs to area j, and the summation range is all nodes i in the power-off area j; the total system load is further calculated as:
[0028] L total =Σ(PL i +jQL i )
[0029] Among them, L totalj Indicates the total load of the entire system, PL i represents the active load of the ith node, QL i represents the reactive load of the i-th node, and the summation range is all nodes i in the system; then the node load loss rate is calculated, which is expressed as:
[0030] LOLR=(∑L dropj ) / L total *100%
[0031] Set the maximum allowable load loss rate threshold LOLR max ;
[0032] If LOLR≤LOLR max , it means that the impact is within the controllable range, and the existing grid operation mode should be maintained without taking emergency measures. Close attention should be paid to the dynamic changes of the system, and appropriate limited topological adjustments should be made to unload the line according to the power flow and voltage conditions; if LOLR>LOLR max, immediately activate the emergency control mechanism, suspend non-emergency operation and maintenance work, give priority to ensuring the power supply of important loads and key users, implement load scheduling for the power outage area, temporarily cut off some non-important loads according to priority, adjust the system operation mode, temporarily switch to partition operation or island operation, promptly inform superiors, request support, and make preparations for manpower and material resources.
[0033] As a preferred solution of the modeling and risk assessment method of power system cascading failures of the present invention, the analysis of the different attack combinations includes: constructing an attack scenario matrix, and dividing the attack scenarios into sub-matrices according to the attack targets, including: a set of attack scenarios for the line, expressed as:
[0034] A line =[a l1 ,a l2 ,...,a ln ]
[0035] The set of attack scenarios for substations is expressed as:
[0036] A sub =[a s1 ,a s2 ,...,a sn ]
[0037] The set of attack scenarios against generator sets is expressed as:
[0038] A gen =[a g1 ,a g2 ,...,a gn ]
[0039] Furthermore, we introduce random perturbation of attack intensity, and for each scenario a i , add Gaussian noise perturbation, assuming that each scene a i It is composed of n independent elements, namely:
[0040] a i =[x1,x2,...,x n ]
[0041] Then for each element x j , calculate x j The expected value of , as the mean of the Gaussian distribution, is expressed as:
[0042] μ j =E(x j )
[0043] Set the hyperparameters to control the noise intensity, expressed as:
[0044] σ j =δj *μ j
[0045] Among them, δ j Represents a hyperparameter; and then j Scrambled Gaussian noise, expressed as:
[0046]
[0047] Then we need to add the noise disturbance scene a i ' is expressed as:
[0048] a i '=[x1',x2',...,x n ']
[0049] For each a i 'Assign weight w i , w i = imp(a i '), where imp(a i ') means in a i The maximum potential load loss under the scenario of ', imp(a i ')=Σ(load j *p j ), where load j represents the normal load of node j, p j Indicates in a i The load ratio that node j may lose in the scenario of ';
[0050] Then for all a i 'Calculate weight w i , construct the weighted matrix, expressed as:
[0051] A=Σ(w i *a i ')
[0052] Furthermore, we analyze the impact of different attack combinations on the system and construct a load loss rate calculation model f(LOLR). i ∈A, calculate the corresponding load loss rate, and sort the LOLR from large to small. High LOLR values correspond to high-risk scenarios. Based on the size of LOLR, the K-Means clustering algorithm is used to cluster the scenarios. The impact of similar scenarios is similar.
[0053] As a preferred solution of the modeling and risk assessment method of power system cascading failures of the present invention, the risk of cascading failures is assessed based on the different attack combinations, including:
[0054] Set the critical point Nc of the tripping line number N, Nmax as the maximum consideration, and perform multiple Monte Carlo simulations for each N value to obtain the corresponding load loss rate average μ(N) and standard deviation σ(N); calculate LOLRα(N) at the confidence level α, expressed as:
[0055] LOLRα(N)=μ(N)+z*σ(N)
[0056] Where z is the α quantile; draw the LOLRα(N)~N curve and determine the critical point Nc. When N>Nc, the power system is in an unstable state. At this time, the fault state is judged, including: if a large-scale cascading failure occurs, resulting in a large-scale power outage in a large area, the power grid is urgently divided to form several relatively independent small areas. In the small area, the emergency reserve generator set is started, and the power supply is gradually restored according to the priority of important loads. The fault line is quickly repaired, the source of the fault is eliminated, and the normal power grid operation topology is gradually restored; if a voltage sag occurs and the power supply voltage in some areas is lower than the standard, some non-important loads are cut off to reduce the pressure on the power grid, and capacitors and reactive power compensation devices are put into use. If the voltage level is still insufficient, the regional DC distribution plan will be initiated. If a large-scale overload occurs and multiple lines are overloaded for a long time, some non-important branches will be immediately cut off to relieve the overload of the lines, and the power flow will be reconfigured to distribute the load to other lines appropriately. If there are not enough backup lines, some loads must be cut off, and the temporarily idle overloaded lines will be repaired and restored as soon as possible. If frequency fluctuations occur and a large number of units are disconnected from the grid, resulting in drastic frequency fluctuations, the virtual synchronous machine will be automatically put into use to stabilize the system frequency, connect to the nearest energy storage power station, quickly support the frequency, check and reconnect the offline units, and dispatch other units with sufficient margin for support.
[0057] In the second aspect, an embodiment of the present invention provides a modeling and risk assessment system for cascading failures in a power system, which includes: a construction module, used to construct a cascading failure simulation model, determine the initial fault conditions, determine whether there is a line flow overload, and calculate the system flow and voltage; a calculation module, used to calculate the node load loss rate based on the node voltage and line flow, and evaluate the system load loss rate; an evaluation module, used to analyze the impact of different attack combinations on the system load loss rate and evaluate the risk of cascading failures.
[0058] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, it implements any step of the above-mentioned power system cascading failure modeling and risk assessment method.
[0059] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned method for modeling and risk assessment of cascading failures in a power system is implemented.
[0060] The beneficial effects of the present invention are that, in response to the problem of simultaneous tripping of transmission lines in the case of chain failures, the present invention establishes a cascading failure model, introduces a weighted random disturbance mechanism, quantitatively analyzes and evaluates the impact of different attack scenarios on the power grid, obtains the system critical point, and proposes multi-level coordinated protection and fault-tolerant measures, which significantly enhances the risk resistance and reliable operation capabilities of the power grid, and is of great significance to improving the safe and stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS
[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:
[0062] Figure 1 It is a flow chart of the present invention.
[0063] Figure 2 is the load loss rate caused by a combined line-breaking attack that randomly disconnects three transmission lines.
[0064] Figure 3 is the load loss rate caused by a combined line-breaking attack that randomly disconnects four transmission lines.
[0065] Figure 4 is the load loss rate caused by a combined line-breaking attack that randomly disconnects five transmission lines.
[0066] Figure 5 is the load loss rate caused by a combined line-breaking attack that randomly disconnects six transmission lines. DETAILED DESCRIPTION
[0067] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.
[0068] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0069] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.
[0070] Example 1
[0071] Reference Figure 1 , which is the first embodiment of the present invention, provides a modeling and risk assessment method for power system cascading failures, including:
[0072] S1: Build a cascading fault simulation model, determine the initial fault conditions, determine whether there is a line flow overload, and calculate the system flow and voltage.
[0073] Collect the complete topological structure data of the target power system, including the number, location, and electrical parameters of substations, generator sets, and transmission lines, and collect the model specifications and technical data of the control principles of related equipment to build a cascading failure simulation model.
[0074] The constructed simulation model is attacked by cyber attacks, which penetrate the IED system, implant malicious code, artificially trigger the line protection action to cause tripping, or inject erroneous line current / voltage data to deceive the protection algorithm to misjudge the line fault, or cause the IED to lose response through denial of service attacks, making it unable to detect and protect normally, resulting in initial faults.
[0075] Furthermore, the fault is judged to determine whether a power flow overload occurs, specifically:
[0076] According to the topological structure and parameters of the power grid, the node admittance matrices [B'] and [B"] are constructed, which reflect the circuit characteristics of the power grid;
[0077] The construction method is:
[0078] For any two nodes i and j, if they are directly connected by a line, then:
[0079] Y ij =Y ji =-y ij
[0080] Y ii =Yjj =Σy ik (k!=i,j)
[0081] where y ij is the branch admittance, and all Y ij The elements form a matrix [Y]
[0082] [B'] = real part ([Y])
[0083] [B”] = imaginary part ([Y])
[0084] Furthermore,
[0085] Calculate the power flow of each line according to the AC power flow calculation equation:
[0086] [P ij ]=[B ij ']*[θ i -θ j ]
[0087] [Q ij ]=[-B ij ”]*[V i 2 -V i *V j *cos(θ i -θ j )]
[0088] Among them: [P ij ] is the active power from node i to j, [Q ij ] is the reactive power from node i to j, θ is the node phase angle, and V is the node voltage amplitude;
[0089] Determine the power flow P of each line cur Whether it exceeds its thermal stability limit P max , P max is the thermal stability limit transmission power of the line, which is determined by the line parameters, P cur is the actual transmission power of the line,
[0090] If P cur >P max , the line is considered to be in an overloaded state;
[0091] For overloaded lines, the circuit will trip with the following probability:
[0092]
[0093] Among them, P trip is the probability of line tripping per unit time, Δt is the simulation time step, T r It is the thermal constant of the line, reflecting the time required for the line to trip from overload.
[0094] It can be seen that when P cur >P max When the line temperature rise rate is cur -P max ) is proportional to the temperature. When the temperature rises to a certain level, the line will be tripped by the protection action. trip It reflects the probability that the line reaches the tripping temperature due to overload within the time Δt.
[0095] According to the attack situation, the probability distribution function f(x) of the occurrence of hidden faults is set. For each device, a probability value p is randomly generated according to f(x) to determine whether p exceeds the set threshold. If so, it is considered that the device has a hidden fault and needs to be removed.
[0096] According to the above calculation results, the system topology is updated, the node admittance matrices [B'] and [B"] are reconstructed, and the AC power flow calculation equation is used to calculate the new line power flow and node voltage.
[0097] Check whether there is instability due to line power overload or node voltage exceeding the limit in the system. If not, it is considered that the cascading failure has ended and the node load loss rate is evaluated;
[0098] If it exists, return to recalculate the line flow.
[0099] S2: Obtain node voltage and line power flow based on power flow calculation, calculate node load loss rate, and evaluate system load loss rate.
[0100] After simulating the evolution of cascading failures, the final system working status is obtained, including:
[0101] Generator status (output voltage, terminal voltage), line status (power flow status, tripping), node voltage, power failure load area and size;
[0102] Further determine whether there is a power failure area at Mankou, set the node voltage allowable range, and calculate the voltage V i Make a judgment:
[0103] If V i If it is within the allowed range, then the node i belongs to the area with power supply;
[0104] If V i If it exceeds the limit or is 0, the node i belongs to the power failure area;
[0105] If node i is completely isolated from the system (island operation), it is also considered to be power-off;
[0106] Among them, the mathematical model of load loss caused by node exceeding the limit is:
[0107]
[0108] Among them, g(P d ) represents the load loss rate of the node, V m and They represent the maximum and minimum values allowed for the node voltage amplitude respectively. When the node voltage amplitude deviates from the allowable value, part of the load will be disconnected from the grid because the voltage does not meet the requirements. Then, a piecewise linear model is used to model the node load loss rate.
[0109] The adjacent power outage area is defined as the Mankou power outage area. For each Mankou power outage area j, the loads of all nodes in it are summed up:
[0110]
[0111] in, represents the total load in the jth power failure area, PL i represents the active load of the ith node, QL i It represents the reactive load of the i-th node, i belongs to area j, and the summation range is all nodes i in the power-off area j.
[0112] The total system load is further calculated as:
[0113] L total =∑(PL i +jQL i )
[0114] in, Indicates the total load of the entire system, PL i represents the active load of the ith node, QL i Represents the reactive load of the i-th node, and the summation range is all nodes i in the system.
[0115] Then the node load loss rate is calculated, which is expressed as:
[0116]
[0117] Set the maximum allowable load loss rate threshold LOLR max , if LOLR≤LOLR max , it means that the impact is within the controllable range, and the existing grid operation mode should be maintained without taking emergency measures. The dynamic changes of the system should be closely monitored, and limited topological adjustments should be made to unload the lines according to the power flow and voltage conditions;
[0118] If LOLR>LOLR max, immediately activate the emergency control mechanism, suspend non-emergency operation and maintenance work, give priority to ensuring the power supply of important loads and key users, implement load scheduling for the power outage area, temporarily cut off some non-important loads according to priority, adjust the system operation mode, temporarily switch to partition operation or island operation, promptly inform superiors, request support, and make preparations for manpower and material resources.
[0119] S3: Analyze the impact of different attack combinations on the system load loss rate and evaluate the risk of cascading failures.
[0120] Construct an attack scenario matrix and divide the attack scenario into sub-matrices according to the attack target, including:
[0121] The set of attack scenarios for the line is expressed as:
[0122] A line =[a l1 ,a l2 ,...,a ln ]
[0123] The set of attack scenarios for substations is expressed as:
[0124] A sub =[a s1 ,a s2 ,...,a sn ]
[0125] The set of attack scenarios against generator sets is expressed as:
[0126] A gen =[a g1 ,a g2 ,...,a gn ]
[0127] Furthermore, we introduce random perturbation of attack intensity, and for each scenario a i , add Gaussian noise perturbation, assuming that each scene a i It is composed of n independent elements, namely:
[0128] a i =[x1,x2,...,x n ]
[0129] Then for each element x j , calculate x j The expected value of , as the mean of the Gaussian distribution, is expressed as:
[0130] μ j =E(x j )
[0131] Set the hyperparameters to control the noise intensity, expressed as:
[0132] σ j =δ j *μ j
[0133] Among them, δ j represents a hyperparameter;
[0134] Then for x j Scrambled Gaussian noise, expressed as:
[0135]
[0136] Then we need to add the noise disturbance scene a i ' is expressed as:
[0137] a i '=[x1',x2',...,x n ']
[0138] For each a i 'Assign weight w i , w i = imp(a i '), where imp(a i ') means in a i The maximum potential load loss under the scenario of ', imp(a i ')=Σ(load j *p j );
[0139] Among them, load j represents the normal load of node j, p j Indicates in a i The load ratio that node j may lose in the scenario of '.
[0140] Then for all a i 'Calculate weight w i , construct the weighted matrix, expressed as:
[0141] A=Σ(w i *a i ')
[0142] Furthermore, we analyze the impact of different attack combinations on the system and construct a load loss rate calculation model f(LOLR). i ∈A, calculate the corresponding load loss rate, and sort the LOLR from large to small. High LOLR values correspond to high-risk scenarios. Based on the size of LOLR, the K-Means clustering algorithm is used to cluster the scenarios. The impact of similar scenarios is similar.
[0143] Set the critical point Nc of the tripped lines N, Nmax as the maximum consideration, and conduct multiple Monte Carlo simulations for each N value to obtain the corresponding average value μ(N) and standard deviation σ(N) of the load loss rate;
[0144] Under the confidence level α, LOLRα(N) is calculated and expressed as:
[0145] LOLRα(N)=μ(N)+z*σ(N)
[0146] Among them, z is the α quantile;
[0147] Draw the LOLRα(N)~N curve and determine the critical point Nc. When N>Nc, the power system is in an unstable state. At this time, the fault state is judged, including:
[0148] If a large-scale cascading failure occurs, resulting in power outages in large areas, the power grid will be urgently divided to form several relatively independent small cells. Within the cells, emergency reserve generators will be started, and power supply will be gradually restored according to the priority of important loads. The faulty lines will be quickly repaired, the source of the fault will be eliminated, and the normal power grid operation topology will be gradually restored.
[0149] If voltage sag occurs and the power supply voltage in some areas is lower than the standard, some non-important loads will be cut off to reduce the pressure on the power grid, capacitors and reactive power compensation devices will be put into use, and nearby generators will be dispatched to increase the regional voltage level. If it still cannot be met, the regional DC distribution plan will be initiated.
[0150] If a large-scale overload occurs and multiple lines are overloaded for a long time, some non-important branches will be immediately cut off to relieve the line overload, reconfigure the power flow, and distribute the load appropriately to other lines; if there are not enough backup lines, part of the load must be cut off, the temporarily idle overloaded lines must be repaired, and restored as soon as possible.
[0151] If frequency fluctuations occur and a large number of units are disconnected from the grid, causing drastic frequency fluctuations, the virtual synchronous machine will be automatically put into use to stabilize the system frequency, connect to the nearest energy storage power station, quickly support the frequency, check and reconnect the offline units to the grid, and dispatch other units with sufficient margin for support.
[0152] In summary, the present invention aims at the problem of simultaneous tripping of transmission lines in the case of chain failures, establishes a cascading failure model, introduces a weighted random disturbance mechanism, quantitatively analyzes and evaluates the impact of different attack scenarios on the power grid, obtains the system critical point, and proposes multi-level coordinated protection and fault-tolerant measures, which significantly enhances the risk resistance and reliable operation capabilities of the power grid, and is of great significance to improving the safe and stable operation of the power system.
[0153] Example 2
[0154] Reference Figure 1Based on the first embodiment, this embodiment further provides a modeling and risk assessment system for cascading failures in a power system, including:
[0155] A construction module is used to construct a cascading fault simulation model, determine the initial fault condition, determine whether there is a line flow overload, and calculate the system flow and voltage;
[0156] A calculation module is used to calculate the node load loss rate and evaluate the system load loss rate based on the node voltage and line flow;
[0157] The evaluation module is used to analyze the impact of different attack combinations on the system load loss rate and evaluate the risk of cascading failures.
[0158] This embodiment also provides a computer device, which is suitable for the modeling and risk assessment method of cascading failures in a power system, and includes a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the modeling and risk assessment method of cascading failures in a power system as proposed in the above embodiment.
[0159] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.
[0160] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for modeling and risk assessment of cascading failures of a power system as proposed in the above embodiment is implemented.
[0161] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.
[0162] Example 3
[0163] Reference Figure 2 to Figure 5 This is the third embodiment of the present invention. Based on the previous two embodiments, this embodiment provides a modeling and risk assessment method for cascading failures in a power system. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.
[0164] For a provincial power grid model, it is assumed that the attacker launches a man-in-the-middle attack on the power system and intercepts the tripping command issued by the control center, causing four transmission lines to trip at the same time at a certain moment, so as to simulate and analyze its impact on the cascading failure of the power grid. In order to analyze the destructive effect of the attack, this section randomly selects three to six transmission lines to launch a combined disconnection attack, repeats the experiment one hundred times for each number of disconnection attack combinations, and counts the load loss rate caused by each disconnection attack combination. The simulation results are shown in Figure 2. Figure 2-Figure 5 shown.
[0165] From the above simulation results, it can be seen that as the number of combined attack disconnections increases, the probability of high load loss in the system increases significantly, which shows that the severity of the initial fault has a significant impact on the results of cascading faults. The average load loss rate under different combined attack disconnection numbers is shown in Table 1:
[0166] Table 1 Average load loss rate caused by different number of disconnections
[0167] Number of combined attack disconnections 3 4 5 6 Average load loss rate 9.52% 11.15% 12.75% 16.61%
[0168] According to the trend of the average load loss rate, it can be found that when the number of disconnected lines is 3 to 5, the average load loss rate will increase linearly at a rate of 15% with the increase in the number of disconnected lines. However, when the number of attacked transmission lines reaches six, the load loss rate increases by more than 30%. The above simulation results show that the power system has a certain robustness against disconnection attacks, but when the number of disconnected lines exceeds a certain value, the robustness of the system will decrease and the load loss will increase rapidly. Therefore, power dispatchers should focus on preventing large-scale disconnection combination attacks.
[0169] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. A modeling and risk assessment method for cascading failures in a power system, characterized by: include, Build a cascading fault simulation model, determine the initial fault conditions, determine whether there is a line flow overload, and calculate system flow and voltage; Based on node voltage and line flow, calculate node load loss rate and evaluate system load loss rate; Analyze the impact of different attack combinations on the system load loss rate and evaluate the risk of cascading failures.
2. The method for modeling and risk assessment of cascading failures in a power system according to claim 1, characterized in that: The initial fault condition is achieved by attacking a cascading fault simulation model constructed by collecting complete topological structure data of the target power system, penetrating the IED system, implanting malicious code, artificially triggering line protection action to cause tripping, or injecting erroneous line current / voltage data to deceive the protection algorithm into misjudging the line fault, or causing the IED to lose response through a denial of service attack, thereby failing to detect and protect normally, resulting in an initial fault.
3. The method for modeling and risk assessment of cascading failures in a power system according to claim 2, characterized in that: The judgment of the line power flow overload includes: According to the topological structure and parameter calculation of the power grid, the node admittance matrix [B'] and [B"] are constructed by: For any two nodes i and j, if they are directly connected by a line, then: AND ij =And ji =-y ij Y ii =Y jj =Σy ik (k!=i,j) where y ij is the branch admittance, and all Y ij The elements form a matrix [Y] [B'] = real part ([Y]) [B”] = imaginary part ([Y]) Furthermore, Calculate the power flow of each line according to the AC power flow calculation equation: [P ij ]=[B ij ']*[θ i -θ j ] [Q ij ]=[-B ij ”]*[V i 2 -V i *V j *cos(θ i -θ j )] Among them: [P ij ] is the active power from node i to j, [Q ij ] is the reactive power from node i to j, θ is the node phase angle, and V is the node voltage amplitude; Determine the power flow P of each line cur Whether it exceeds its thermal stability limit P max , P max is the thermal stability limit transmission power of the line, which is determined by the line parameters, P cur is the actual transmission power of the line, If P cur >P max , the line is considered to be an overloaded line.
4. The method for modeling and risk assessment of cascading failures in a power system according to claim 3, characterized in that: The trip probability is calculated based on the overload line and is expressed as: Among them, P trip is the probability of line tripping per unit time, Δt is the simulation time step, T r is the line thermal constant, which indicates the time required for the line to trip from overload; When P cur >P max When the line temperature rise rate is cur -P max ) is proportional to the temperature. When the temperature rises to a certain level, the line will be tripped by the protection action. trip It reflects the probability that the line reaches the tripping temperature due to overload within the time Δt. According to the attack situation, the probability distribution function f(x) of the occurrence of hidden faults is set. For each device, a probability value p is randomly generated according to f(x). It is determined whether p exceeds the set threshold. If so, it is considered that the device has a hidden fault and needs to be removed. According to the calculation results, the system topology is updated, the node admittance matrices [B'] and [B"] are reconstructed, and the AC power flow calculation equation is used to calculate the new line power flow and node voltage; Check whether there is instability due to line power overload or node voltage exceeding the limit in the system. If not, it is considered that the cascading failure has ended and the node load loss rate is evaluated; If it exists, return to recalculate the line flow.
5. The method for modeling and risk assessment of cascading failures in a power system according to claim 4, characterized in that: The calculation of the node off-load rate includes: Get the working status of the system after the chain failure ends, further determine whether there is a power failure area, set the node voltage allowable range, and judge each node voltage Vi: If Vi is within the allowed range, the node i belongs to the area with power supply; If Vi exceeds the limit or is 0, the node i belongs to the power failure area; If node i is completely isolated from the system, it is also considered to be power-off; The adjacent power outage area is defined as the Mankou power outage area. For each Mankou power outage area j, the loads of all nodes in it are summed up: in, represents the total load in the jth power failure area, PL i represents the active load of the ith node, QL i represents the reactive load of the i-th node, i belongs to area j, and the summation range is all nodes i in the power-off area j; The total system load is further calculated as: L total =Σ(PL i +jQL i ) in, Indicates the total load of the entire system, PL i represents the active load of the ith node, QL i represents the reactive load of the i-th node, and the summation range is all nodes i in the system; Then the node load loss rate is calculated, which is expressed as: Set the maximum allowable load loss rate threshold LOLR max ; If LOLR≤LOLR max , it means that the impact is within the controllable range, and the existing grid operation mode should be maintained without taking emergency measures. The dynamic changes of the system should be closely monitored, and limited topological adjustments should be made to unload the lines according to the power flow and voltage conditions; If LOLR>LOLR max , immediately activate the emergency control mechanism, suspend non-emergency operation and maintenance work, give priority to ensuring the power supply of important loads and key users, implement load scheduling for the power outage area, temporarily cut off some non-important loads according to priority, adjust the system operation mode, temporarily switch to partition operation or island operation, promptly inform superiors, request support, and make preparations for manpower and material resources.
6. The method for modeling and risk assessment of cascading failures in a power system according to claim 5, characterized in that: Analysis of the different attack combinations includes: Construct an attack scenario matrix and divide the attack scenario into sub-matrices according to the attack target, including: The set of attack scenarios for the line is expressed as: A line =[a l1 ,a l2 ,...,a ln ] The set of attack scenarios for substations is expressed as: A sub =[a s1 ,a s2 ,...,a sn ] The set of attack scenarios against generator sets is expressed as: A gen =[a g1 ,a g2 ,...,a gn ] Furthermore, we introduce random perturbation of attack intensity, and for each scenario a i , add Gaussian noise perturbation, assuming that each scene a i It is composed of n independent elements, namely: a i =[x1,x2,...,x n ] Then for each element x j , calculate x j The expected value of , as the mean of the Gaussian distribution, is expressed as: m j =E(x j ) Set the hyperparameters to control the noise intensity, expressed as: s j =d j *m j Among them, δ j represents a hyperparameter; Then for x j Scrambled Gaussian noise, expressed as: Then we need to add the noise disturbance scene a i ' is expressed as: a i '=[x1',x2',...,x n '] For each a i 'Assign weight w i , w i = imp(a i '), where imp(a i ') means in a i The maximum potential load loss under the scenario of ', imp(a i ')=Σ(load j *p j ); Among them, load j represents the normal load of node j, p j Indicates in a i The load ratio that node j may lose in the scenario of '; Then for all a i 'Calculate weight w i , construct the weighted matrix, expressed as: A=Σ(w i *a i ') Furthermore, we analyze the impact of different attack combinations on the system and construct a load loss rate calculation model f(LOLR). i ∈A, calculate the corresponding load loss rate, and sort the LOLR from large to small. High LOLR values correspond to high-risk scenarios. Based on the size of LOLR, the K-Means clustering algorithm is used to cluster the scenarios. The impact of similar scenarios is similar.
7. The method for modeling and risk assessment of cascading failures in a power system according to claim 6, characterized in that: Assess the risk of cascading failures based on the different attack combinations described, including: Set the critical point Nc of the tripped lines N, Nmax as the maximum consideration, and conduct multiple Monte Carlo simulations for each N value to obtain the corresponding average value μ(N) and standard deviation σ(N) of the load loss rate; Under the confidence level α, LOLRα(N) is calculated and expressed as: LOLRα(N)=μ(N)+z*σ(N) Among them, z is the α quantile; Draw the LOLRα(N)~N curve and determine the critical point Nc. When N>Nc, the power system is in an unstable state. At this time, the fault state is judged, including: If a large-scale cascading failure occurs, resulting in a large-scale power outage in a large area, the power grid will be urgently divided to form several relatively independent small areas. In the small area, the emergency reserve generator set will be started, and the power supply will be gradually restored according to the priority of important loads. The faulty line will be quickly repaired to eliminate the source of the fault and gradually restore the normal power grid operation topology; If voltage sag occurs and the power supply voltage in some areas is lower than the standard, some non-important loads will be cut off to reduce the pressure on the power grid, capacitors and reactive power compensation devices will be put into use, and nearby generators will be dispatched to increase the regional voltage level. If the voltage level still cannot be met, the regional DC distribution plan will be initiated; If a large-scale overload occurs and multiple lines are overloaded for a long time, some non-important branches should be immediately cut off to relieve the overload of the lines, reconfigure the flow, and distribute the load to other lines appropriately; if there are not enough backup lines, some loads must be cut off, the temporarily idle overloaded lines must be repaired, and the overloaded lines must be restored as soon as possible; If frequency fluctuations occur and a large number of units are disconnected from the grid, causing drastic frequency fluctuations, the virtual synchronous machine will be automatically put into use to stabilize the system frequency, connect to the nearest energy storage power station, quickly support the frequency, check and reconnect the offline units to the grid, and dispatch other units with sufficient margin for support.
8. A modeling and risk assessment system for cascading failures in a power system, based on the modeling and risk assessment method for cascading failures in a power system according to any one of claims 1 to 7, characterized in that: include, A construction module is used to construct a cascading fault simulation model, determine the initial fault condition, determine whether there is a line flow overload, and calculate the system flow and voltage; A calculation module is used to calculate the node load loss rate and evaluate the system load loss rate based on the node voltage and line flow; The evaluation module is used to analyze the impact of different attack combinations on the system load loss rate and evaluate the risk of cascading failures.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for modeling and risk assessment of cascading failures in a power system according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for modeling and risk assessment of cascading failures in a power system according to any one of claims 1 to 7 are implemented.
Citation Information
Cited By
Main and distribution network cooperation multi-time scale cooperation capability evaluation and analysis device based on cloud edge cooperation architecture
CN119886917A
Power balancing method and system for multiple parallel lines of new energy power grid
CN120262465A
A power balancing method and system for multiple parallel lines of a new energy power grid
CN120262465B
Power grid stability risk boundary generation method based on large model
CN120582236A
Method, device, equipment and medium for power grid load loss analysis
CN121192693A
