Virtual power plant multi-stage peak regulation task reliability optimization method and device, medium and product

Abstract

The invention discloses a virtual power plant multi-stage peak regulation task reliability optimization method and device, a medium and a product, and relates to the field of virtual power plant peak regulation, and the method comprises the steps: constructing a virtual power plant information channel availability rate model and a virtual power plant information physical device availability rate model; establishing a virtual power plant full-stage task binary decision diagram model; according to the above model, determining the reliability index of the full-stage task of the virtual power plant; and establishing a reliability optimization model considering the uncertainty of the real-time peak regulation service demand and a reliability optimization model considering the uncertainty of the main scheme equipment state, solving, optimizing the reliability index of the full-stage task of the virtual power plant, and completing the optimization of the reliability of the multi-stage peak regulation task of the virtual power plant. The method can consider the influence of information physical interaction on the expressive force of the virtual power plant in the process of participating in the electricity market, and achieves the optimization of the reliability of the multi-stage peak regulation task of the virtual power plant.

Description

Technical Field

[0001] The present application relates to the field of virtual power plant peak shaving, and particularly to a method, device, medium and product for optimizing the reliability of multi-stage peak shaving tasks of a virtual power plant. Background Art

[0002] The traditional reliability assessment objects are mainly physical devices such as power transmission and distribution systems, distribution transformers, and energy storage devices. At present, there is still a lack of a reliability assessment method for virtual power plants. As a new energy management entity that includes distributed new energy devices, energy storage, and flexible loads, a virtual power plant obtains benefits by participating in the electricity ancillary service market transactions. The internal operation control of a virtual power plant is a complex process of cyber-physical interaction, and it is urgent to consider the impact of cyber-physical interaction on the performance of the virtual power plant in the process of participating in the electricity market. Summary of the Invention

[0003] The purpose of the present application is to provide a method, device, medium and product for optimizing the reliability of multi-stage peak shaving tasks of a virtual power plant, which can consider the impact of cyber-physical interaction on the performance of the virtual power plant in the process of participating in the electricity market, and realize the optimization of the reliability of multi-stage peak shaving tasks of the virtual power plant.

[0004] To achieve the above purpose, the present application provides the following solutions:

[0005] In the first aspect, the present application provides a method for optimizing the reliability of multi-stage peak shaving tasks of a virtual power plant, including:

[0006] Constructing a normal-fault state transition model of cyber-physical devices;

[0007] Establishing an availability model of the virtual power plant information channel; the availability model of the virtual power plant information channel is determined based on the packet loss, transmission delay, and error code availability of the transmission channel;

[0008] Based on the normal state probability of the cyber-physical devices in the normal-fault state transition model and the actual capacity of the cyber-physical devices, establishing an availability model of the virtual power plant cyber-physical devices;

[0009] Establishing a binary decision diagram model for the full-stage tasks of the virtual power plant; the binary decision diagram model for the full-stage tasks of the virtual power plant is established based on the binary decision diagram model for single-stage tasks of the virtual power plant; the binary decision diagram model for single-stage tasks of the virtual power plant is established based on the fault tree of the single-stage tasks of the virtual power plant and using Shannon's decomposition law;

[0010] According to the availability model of the virtual power plant information channel, the availability model of the virtual power plant cyber-physical devices, and the binary decision diagram model for the full-stage tasks of the virtual power plant, determining the reliability index of the full-stage tasks of the virtual power plant;

[0011] Build a reliability optimization model considering the uncertainty of real-time peak shaving service demand and a reliability optimization model considering the uncertainty of the equipment status of the main plan; the reliability optimization model considering the uncertainty of real-time peak shaving service demand includes a first objective function and a first set of constraint conditions; the first objective function aims to minimize the operation coefficient of the virtual power plant operator in the day-ahead peak shaving ancillary service market and minimize the operation coefficient of the real-time peak shaving ancillary service market; the first set of constraint conditions includes the active power balance constraint of the virtual power plant, the output constraint of distributed new energy equipment, the power quantity constraint of energy storage equipment, the upper and lower adjustment constraints of flexible loads, and the constraint of the output of each device of the virtual power plant in the real-time peak shaving ancillary service market by the output in the day-ahead peak shaving ancillary service market; the reliability optimization model considering the uncertainty of the equipment status of the main plan includes a second objective function and a second set of constraint conditions; the second objective function aims to minimize the operation coefficient of the virtual power plant operator in the peak shaving ancillary service market.

[0012] Solve the reliability optimization model considering the uncertainty of real-time peak shaving service demand and the reliability optimization model considering the uncertainty of the equipment status of the main plan respectively to obtain optimization parameters; the optimization parameters include the output of distributed new energy equipment, the charging power of energy storage equipment, the discharging power of energy storage equipment, the output of distributed new energy equipment in the real-time peak shaving ancillary service market, the charging quantity of energy storage equipment in the real-time peak shaving ancillary service market, the discharging quantity of energy storage equipment in the real-time peak shaving ancillary service market, the downward adjustment quantity of flexible loads in the real-time peak shaving ancillary service market, and the upward adjustment quantity of flexible loads in the real-time peak shaving ancillary service market.

[0013] Optimize the reliability index of the virtual power plant's full-stage tasks based on the optimization parameters to complete the optimization of the reliability of the virtual power plant's multi-stage peak shaving tasks.

[0014] In a second aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the virtual power plant multi-stage peak shaving task reliability optimization method described in any one of the above.

[0015] In a third aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the virtual power plant multi-stage peak shaving task reliability optimization method described in any one of the above.

[0016] In a fourth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the virtual power plant multi-stage peak shaving task reliability optimization method described in any one of the above.

[0017] According to the specific embodiments provided by the present application, the present application has the following technical effects:

[0018] The present application provides a reliability optimization method, device, medium and product for multi-stage peak shaving tasks of a virtual power plant, constructs a normal-fault state transition model of cyber-physical devices; establishes an availability model of the information channels of the virtual power plant; based on the normal state probability of the cyber-physical devices in the normal-fault state transition model and the actual capacity of the cyber-physical devices, establishes an availability model of the cyber-physical devices of the virtual power plant; establishes a binary decision diagram model for the full-stage tasks of the virtual power plant; determines the reliability index of the full-stage tasks of the virtual power plant according to the availability model of the information channels of the virtual power plant, the availability model of the cyber-physical devices of the virtual power plant and the binary decision diagram model for the full-stage tasks of the virtual power plant; establishes a reliability optimization model considering the uncertainty of real-time peak shaving service demand and a reliability optimization model considering the uncertainty of the equipment state of the main plan and solves them to obtain optimization parameters; optimizes the reliability index of the full-stage tasks of the virtual power plant based on the optimization parameters to complete the optimization of the reliability of the multi-stage peak shaving tasks of the virtual power plant. The present application can consider the impact of cyber-physical interaction on the performance of the virtual power plant in the process of participating in the power market and realize the optimization of the reliability of the multi-stage peak shaving tasks of the virtual power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0020] Figure 1 It is a schematic flow chart of a reliability optimization method for multi-stage peak shaving tasks of a virtual power plant provided by an embodiment of the present application;

[0021] Figure 2 It is a schematic diagram of the two-state conversion model of the device provided by an embodiment of the present application;

[0022] Figure 3 It is a schematic diagram of the fault tree of a single-stage task of a virtual power plant provided by an embodiment of the present application;

[0023] Figure 4 It is a schematic diagram of the basic graphical representation of the operation rules provided by an embodiment of the present application;

[0024] Figure 5 It is a schematic diagram of the binary decision diagram model of a single-stage task of a virtual power plant provided by an embodiment of the present application;

[0025] Figure 6Schematic diagram of the fault tree model for the full-stage tasks of a virtual power plant provided by an embodiment of the present application;

[0026] Figure 7 Simplified single-stage binary decision diagram provided by an embodiment of the present application;

[0027] Figure 8 Binary decision diagram for the peak shaving tasks in the full stage of a virtual power plant provided by an embodiment of the present application;

[0028] Figure 9 Reliability optimization flowchart for the multi-stage peak shaving tasks of a virtual power plant provided by an embodiment of the present application;

[0029] Figure 10 Schematic diagram of the structure of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0030] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0031] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below in conjunction with the drawings and specific implementation manners.

[0032] In an exemplary embodiment, as Figure 1 shown, a reliability optimization method for the multi-stage peak shaving tasks of a virtual power plant is provided, including the following steps:

[0033] S1: Construct a normal-fault state transition model for cyber-physical devices.

[0034] In practical applications, the basic "normal-fault" two-state transition model of the device based on the Markov process is as Figure 2 shown.

[0035] Among them, N i,t represents that device i is in the normal state; F i,t represents that device i is in the fault state; λ i represents the failure rate of device i; μ i is the repair rate of device i, and both are historical statistical data; represents the probability that device i is in the normal state at time t; represents the probability that device i is in the fault state at time t. The dynamic transition process of the device between the "normal-fault" states can be represented by formula (1) and formula (2):

[0036]

[0037] The analytical expression formulas (3) and (4) for the probabilities of the device being in normal and faulty states are as follows:

[0038]

[0039] As can be seen from formulas (3) and (4), the probability of the device being in a normal state continuously decreases with the increase of the running time, while the probability of being in a faulty state continuously increases with the increase of the running time; both will eventually converge to a fixed value.

[0040] S2: Establish a virtual power plant information channel availability model; the virtual power plant information channel availability model is determined based on the packet loss, transmission delay, and error code availability of the transmission channel.

[0041] In practical applications, communication transmission methods include two categories: wired transmission and wireless transmission. For the wireless transmission method, the device is only a node device; while for the wired transmission method, the device includes both node devices and line devices, that is, the communication transmission line. The transmission characteristics of node devices include delay, error code, and packet loss, while the transmission characteristics of line devices are only transmission delay. The availability of the data transmission channel is jointly determined by the transmission characteristics of node devices and line devices.

[0042] 1) Communication system routing feature set.

[0043] The communication system routing feature represents the transmission characteristics between node devices and only includes the propagation delay, as shown in formula (5). For wired communication, the propagation delay is only determined by the line length and can be calculated by formula (6):

[0044]

[0045]

[0046] Among them, is the communication system routing feature; DL l (t) is the propagation delay; L l is the length of the communication transmission line l, and ε is the electromagnetic wave propagation rate.

[0047] Different from wired communication, due to the interference of wireless communication, packet retransmission is inevitable, which may increase the propagation delay. The propagation delay is determined by the line length and the number of retransmissions and is expressed as:

[0048]

[0049] where r is the number of data packet retransmissions, and Pr(r) is the probability of the data packet being retransmitted r times, is the upper limit value of the number of data packet retransmissions; p RT is the probability of data packet retransmission, is the combination number.

[0050] 2) Communication node feature set.

[0051] The communication node features represent the transmission characteristics of different node devices, which are jointly defined by packet loss, node switching delay, and error code characteristics, as shown in Equation (9):

[0052]

[0053] where, is the node attribute set; LS j (t) represents node packet loss; DL j (t) represents node switching delay; ER j (t) represents the error code feature.

[0054] The Gilbert-Elliot model based on Markov chain is used to simulate the node packet loss characteristics. First, according to historical data, packet losses are divided into multi-state groups with representative packet loss rates. Then, the probability of each state can be obtained by solving the Kolmogorov equation. Finally, the probability mass function corresponding to the node data passing rate is obtained:

[0055]

[0056] where, represents the node data passing rate of the k-th group; represents the occurrence probability of the node data of the k-th group, and the value ranges of both are from 0 to 100%. For example represents that the probability of 75% of the data passing through node j smoothly is 6%. is the total number of groups of the node data passing rate. For the node error code feature and the node switching delay feature, the Poisson distribution and the normal distribution are used to describe their respective uncertainties, as shown in the following formula:

[0057]

[0058] where s j is the number of error bits per second, γ j is its corresponding expected value. τ j represents the switching delay of node device j; T de represents the switching delay threshold of node device j. represents the mean value of the switching delay; It represents the variance of the switching delay and can be determined through historical statistics.

[0059] 3) The set of transmission channel availability rates.

[0060] The transmission channels of a communication system include the node devices and communication routes through which data passes from the sending point to the receiving point, and can be defined by formula (13):

[0061]

[0062] where, represents the set of node devices included in the communication transmission channel h; represents the set of communication routes included in the communication transmission channel h. The characteristics of the information transmission channel are jointly determined by the corresponding communication node characteristics and routing characteristics, as shown in formulas (14) - (17):

[0063]

[0064] where, is the set of information transmission channel characteristics; LS h (t), DL h (t) and ER h (t) represent the packet loss, delay, and error characteristics of the transmission channel h respectively, and are defined by formulas (15), (16), and (17). LS j (t) is the packet loss of the j-th node device in the transmission channel h; DL j (t) is the switching delay of the j-th node device in the transmission channel h; DL l (t) is the switching delay of the l-th communication route in the transmission channel h; ER j (t) is the error characteristic of the j-th node device in the transmission channel h; It can be seen from formulas (15) and (17) that the packet loss and error of the transmission channel h both depend on the worst-case scenario of the corresponding node characteristics. Formula (16) indicates that the transmission delay of the transmission channel h is the sum of the delays of all node devices and communication routes. Only when the packet loss, transmission delay, and error of the transmission channel h are all within the restricted range can the data be successfully transmitted. Therefore, the availability of the transmission channel is defined by the product of the packet loss, transmission delay, and error availability rates, and the virtual power plant information channel availability rate model is:

[0065]

[0066] where, is the virtual power plant information channel availability rate; is the packet loss of the h-th transmission channel at time t; is the transmission delay of the h-th transmission channel at time t; BER availability of the h-th transmission channel at time t. The calculation method is shown in the following formula:

[0067]

[0068] Wherein, is the probability that the packet loss of transmission channel h is less than or equal to the total number of groups of the node data passing rate; is for node device j * is the probability of being in a normal state, represents that the packet loss rate of device j * is greater than that of other node communication devices; Pr(DL h (t) ≤ T de ) is the probability that the switching delay of transmission channel h is less than or equal to the switching delay threshold of node device j; is the probability that the switching delay of the j-th node device in transmission channel h is less than or equal to the upper limit of the node switching delay; is the upper limit of the switching delay of the j-th node device; is the probability that device j is in a normal state; Pr(DL l (t) ≤ T l DP ) is the probability that the switching delay of the l-th communication route in transmission channel h is less than or equal to the upper limit of the communication route switching delay; T l DP is the upper limit of the switching delay of the l-th communication route; is the probability that device j is in a normal state; is the probability that the bit error probability of the l-th communication transmission channel is less than the upper limit value of the bit error rate; ES h (t) is the bit error rate of the l-th communication transmission channel at time t; is the upper limit of the bit error rate of the communication transmission channel; ES j (t) is the bit error rate of the j-th node communication device at time t; represents the j-th * bit error rate of the node communication device is greater than that of other node devices.

[0069] S3: Based on the normal state probability of the cyber-physical device in the normal-fault state transition model and the actual capacity of the cyber-physical device, establish a cyber-physical device availability model for the virtual power plant.

[0070] In practical applications, in most traditional studies, the availability of physical devices is simplified to a two-state model that includes normal and faulty states. However, restricting the availability of physical devices to the dimension of "normal - faulty" two states is insufficient. The actual capacity of physical devices also has a profound impact on their operational availability. For example, if the actual capacity of a device is too small to support a task, it will be considered to have low availability even though it is in an operating state. Therefore, the availability of physical devices is defined not only by the probability of being in a normal state but also by the actual capacity. The availability model of the cyber-physical devices in the virtual power plant is as follows:

[0071]

[0072] where, is the state availability of the i-th cyber-physical device at time t; Pr i N (t) is the probability that the i-th cyber-physical device is in a normal state at time t; is the actual capacity of the i-th cyber-physical device at time t; DER is a distributed new energy device; ES is an energy storage device; Ω t is the set of physical devices corresponding to time t; T is the time series set of the multi-stage peak shaving tasks of the virtual power plant.

[0073] 1) The actual capacity of distributed new energy devices.

[0074] The prediction process of the power generation of distributed new energy devices ignores the impact of natural disasters, which may affect the judgment of network operators. Therefore, it is urgent to consider the occurrence of meteorological disasters and re-evaluate the actual power generation capacity of distributed new energy devices. Assuming that the occurrence of different disasters is independent, the actual power generation capacity of the i-th distributed new energy device at time t can be expressed by the following formula:

[0075]

[0076] where, is the predicted capacity of distributed new energy device i at time t; Pr(ξ f,t ) is the occurrence probability of the f-th natural disaster at time t, which is a historical statistical value; η f,i,t is the factor reflecting the impact of disaster f on distributed new energy device i at time t, and its value ranges from 0 to 1; Υ dis is the set of natural disasters, including flood disasters, freezing disasters, and typhoon disasters.

[0077] 2) The actual capacity of energy storage devices.

[0078] The capacity decay of the energy storage battery includes two parts: calendar decay and cycle decay. Calendar decay refers to the phenomenon that the capacity of the energy storage battery gradually decreases over time without use; the corresponding capacity loss can be calculated according to Eyring's law, as shown in the following formula:

[0079]

[0080] Where, is the capacity loss of calendar decay of energy storage device i at time t; k B is the Boltzmann constant; T i,b,τ is the ambient temperature of the energy storage battery b in energy storage device i; is the total number of batteries in energy storage device i; A τ is the decay coefficient of the battery, is the activation energy of the energy storage battery b in energy storage device i, and Δτ is the working duration of the battery.

[0081] The cycle decay is the capacity decay during the cyclic charge and discharge of the energy storage battery. The capacity loss is affected by the number of cycles and can be calculated by the following formula:

[0082]

[0083] Where, is the capacity loss of cycle decay of energy storage device i at time t; V 0 and are the standard voltage and initial capacity of the energy storage battery b; is the equivalent charge and discharge cycle number of the battery up to the (τ - 1)th moment; B and c are the cycle aging coefficients of the battery; is the charge / discharge amount of the battery cell b in energy storage device i at the τth moment; kT i,b,τ is a factor reflecting the relationship between temperature and capacity loss and is calculated by the following formula:

[0084]

[0085] Where, is the activation energy of battery cycle aging; T i,b,τ is the temperature of the battery b in energy storage device i at the moment, and T ref is the ambient temperature reference value. Considering the calendar decay and cycle decay, the actual power generation capacity of energy storage device i is expressed by the following formula:

[0086]

[0087] Where, is the actual capacity of energy storage device i at time t, is the initial capacity of the energy storage device without considering power decay.

[0088] S4: Establish a binary decision diagram model for the whole-stage tasks of the virtual power plant; the binary decision diagram model for the whole-stage tasks of the virtual power plant is established based on the binary decision diagram model for single-stage tasks of the virtual power plant; the binary decision diagram model for single-stage tasks of the virtual power plant is established based on the fault tree of single-stage tasks of the virtual power plant using Shannon's decomposition law.

[0089] In practical applications, the single-stage tasks of the virtual power plant can be decomposed into three subtasks, including instruction reception, allocation, and task execution. There are two task execution schemes, namely the main scheme and the auxiliary scheme; if the main scheme or the auxiliary scheme is effectively executed, the task can be successfully completed. When all subtasks are successfully executed, it is determined that the single-stage task of the system is successful. The fault tree model of the single-phase task is as Figure 3 shown.

[0090] According to the above single-stage fault tree model, the single-stage binary decision diagram of the virtual power plant can be obtained by Shannon's decomposition law, as shown in the following formula:

[0091]

[0092] where G is a Boolean function based on the binary logic variable set Z = {z 1 , z 2 , z 3 , z 4}, where z 1 , z 2 , z 3 , z 4 represent the execution status of task A, task B, task C1, and task C2 respectively, and the variable sorting is: z 3 < z 4 < z 2 < z 1 . represents that the m-th task is successfully executed, and z m is the execution status of the m-th task; represents taking the logical opposite of the variable z m , if z m = 1 then vice versa; represents that the m-th task fails to be successfully executed. ite is an abstract operator representing the conditional statement "if-then-else". According to formula (28) and the corresponding variable sorting, the binary decision diagram model for single-stage tasks is constructed from bottom to top by the above rules, and the expression is as follows:

[0093]

[0094] where G and H are two Boolean functions for traversing the fault tree, G 1 , G0 and H 1 、H 2 are the sub - function expressions corresponding to the Boolean functions G and H respectively, as Figure 4 shown. The above rules are recursively applied until one of the sub - expressions becomes a constant expression '0' or '1'.

[0095] Based on the above single - stage task fault tree model and the binary decision diagram construction rules, a single - stage task binary decision diagram model of the virtual power plant is generated, as Figure 5 shown.

[0096] As Figure 5 shown, the proposed single - stage task binary decision diagram model of the virtual power plant is a rooted, directed acyclic graph with two sink nodes and multiple non - sink nodes. The non - sink nodes represent different subtasks, and the sink nodes marked as '0' and '1' represent the failure and success of the single - stage task respectively. Two edges are output from each non - sink node, and the edges marked as '0' and '1' represent the failure and success of the subtask corresponding to the non - sink node respectively. From Figure 5 it can be seen that the success of subtask A and subtask B requires the availability rate of the communication system transmission channel to be maintained at a high level, while the success of subtask C requires the availability rate of the physical system equipment to be maintained at a relatively high level.

[0097] The success of the full - stage peak - shaving task of the virtual power plant highly depends on the successful execution of the single - stage tasks. If each single - stage task of the virtual power plant is successfully executed, the corresponding full - stage peak - shaving task of the virtual power plant can be regarded as completely reliable. Therefore, a full - stage task fault tree model of the virtual power plant is proposed, as Figure 6 shown.

[0098] In addition, to obtain the full - stage task binary decision diagram according to the above full - stage task fault tree model, the single - stage task binary decision diagram needs to be simplified, as Figure 7 shown.

[0099] Applying the combination rule of single - stage subtasks, i.e., formula (29), to the binary decision diagrams simplified in each stage, a full - stage task binary decision diagram composed of single - stage task binary decision diagrams is obtained, as Figure 8 shown.

[0100] S5: Determine the reliability index of the full - stage task of the virtual power plant according to the virtual power plant information channel availability rate model, the virtual power plant cyber - physical equipment availability rate model, and the virtual power plant full - stage task binary decision diagram model.

[0101] As Figure 8As shown, the red dotted line with arrows constitutes the completely reliable main path for the peak shaving tasks in all stages of the virtual power plant; according to this path, the probability reliability index of the peak shaving tasks in all stages determined by the reliability of all single-stage peak shaving tasks is defined as follows:

[0102]

[0103] Wherein, X and respectively represent the success and failure of subtask X (X = A p , B p , C p,1 or C p,2 ); is the coefficient reflecting the importance of the task in time period t p , and is also used to adjust the magnitude of the RE 1 value. N pha is the total number of stages of the multi-stage peak shaving task; T phy＝p is the moment corresponding to stage p; is the probability that tasks A, B, and C1 in time period t p are successfully executed; is the probability that tasks A, B, and C2 in time period t p are successfully executed and task C1 is not successfully executed. Assuming that the tasks in each stage and the subtasks within a single stage are independent of each other, the above formula can be rewritten as follows:

[0104]

[0105] Wherein, Pr(X) is the probability that task X is successfully executed, and the successful completion of the task depends on the status of the devices participating in the task. Unless all devices are available, the successful execution of the task cannot be ensured; therefore, the task success probability is defined as follows:

[0106]

[0107] Wherein, and are the sets of communication transmission channels participating in task A p and task B p ; while and are the sets of physical devices participating in task C p,1 and C p,2 ; is the availability rate of the h-th information channel; Τ pha＝p is the time series set of the peak shaving tasks in stage p of the virtual power plant.

[0108] In addition to the probability reliability of the peak shaving tasks in all stages of the virtual power plant described above, reliability can be further defined by the insufficient demand for peak shaving services, which can be expressed by the following formula:

[0109]

[0110]

[0111] Among them, is the supply-demand deviation of the peak shaving service for subtask C1; is the supply-demand deviation of the peak shaving service for subtask C2; is the service demand for subtask C1; and are the sets of physical devices corresponding to the main plan and the auxiliary plan, respectively; p i (t p ) is the output value of the i-th physical device during the execution of the main plan; p j (t p ) is the output value of the j-th device during the execution of the auxiliary plan; is the correlation coefficient between subtask A and subtask C1 in the p stage; is the correlation coefficient between subtask A and subtask C2 in the p stage, and are the elements in the equipment availability matrix and as shown in formulas (39) and (40). Similarly, is the correlation coefficient between subtask B and subtask C1 in the p stage, is the correlation coefficient between subtask B and subtask C2 in the p stage, and are the elements in the equipment availability matrix and as shown in formulas (41) and (42):

[0112]

[0113] Among them, M C1-A is the coupling matrix of the physical devices participating in subtask C1 and the communication channels participating in task A; M C2-A is the coupling matrix of the physical devices participating in subtask C2 and the communication channels participating in task A; M C1-B is the coupling matrix of the physical devices participating in subtask C1 and the communication channels participating in task B; M C2-B is the coupling matrix of the physical devices participating in subtask C2 and the communication channels participating in task B; is the availability matrix of the communication transmission channels participating in subtask A; is the availability matrix of the communication transmission channels participating in subtask B, as shown in the following formula:

[0114]

[0115] Among them, and are the total number of physical devices participating in subtasks C1 and C2; is the availability rate of the h-th information channel; and are respectively the total number of communication transmission channels participating in tasks A and B.

[0116] S6: Establish a reliability optimization model considering the uncertainty of real-time peak shaving service demand and a reliability optimization model considering the uncertainty of the equipment status of the main plan; the reliability optimization model considering the uncertainty of real-time peak shaving service demand includes a first objective function and a first set of constraint conditions; the first objective function aims to minimize the operation coefficient of the virtual power plant operator in the day-ahead peak shaving ancillary service market and minimize the operation coefficient of the real-time peak shaving ancillary service market; the first set of constraint conditions includes the active power balance constraint of the virtual power plant, the output constraint of distributed new energy equipment, the power quantity constraint of energy storage equipment, the upper and lower adjustment constraints of flexible loads, and the constraint of the output of each device of the virtual power plant in the real-time peak shaving ancillary service market by the day-ahead peak shaving ancillary service market; the reliability optimization model considering the uncertainty of the equipment status of the main plan includes a second objective function and a second set of constraint conditions; the second objective function aims to minimize the operation coefficient of the virtual power plant operator in the peak shaving ancillary service market.

[0117] In practical applications, the deviation between the supply and demand of peak shaving services for subtasks C1 and C2 will have a greater impact on the reliability index RE 2 of the virtual power plant's full-stage peak shaving service; the supply-demand deviation of subtasks C1 and C2 is defined by the following formula:

[0118]

[0119] Among them, and respectively represent the service demand of subtask C1 and the set of physical devices participating in subtask C1; and respectively represent the service demand of subtask C2 and the set of physical devices participating in subtask C2. In fact, after the peak shaving ancillary service price volume in the day-ahead market is determined, the virtual power plant operator will allocate power demand to each entity according to the contract volume. However, in order to ensure operation stability and reliability, the total peak shaving service demand of the virtual power plant operator will be adjusted by the system operator in the real-time market; therefore, the demand for subtask C1 is uncertain.

[0120] To improve the reliability of virtual power plants in the face of uncertainties in real-time peak shaving services, this application proposes a two-stage robust optimization model for the day-ahead market and the real-time market. Among them, the goal of the virtual power plant operator is to minimize its total coefficient in the two markets, and the first objective function is:

[0121]

[0122] Among them, and are the operation coefficients of the operator in the day-ahead market and the real-time market respectively. The coefficient of the day-ahead market is shown in the following formula:

[0123]

[0124] Among them, is the operation coefficient of the virtual power plant operator in the day-ahead peak shaving ancillary service market; the first item represents the power generation coefficient of distributed new energy equipment, is the output of the i-th distributed new energy equipment at time t p ; α RG is the cost coefficient of distributed new energy equipment; the second item is the charge and discharge coefficient of energy storage equipment, is the charging power of the i-th energy storage equipment at time t p ; is the discharge power of the i-th energy storage equipment at time t p ; α ES is the cost coefficient of energy storage equipment; Τ pha＝p is the time series set of the peak shaving tasks of the virtual power plant in the p stage. The constraints corresponding to the day-ahead market cost optimization problem are shown as follows.

[0125] The active power balance constraint of the virtual power plant is:

[0126]

[0127] Among them, is the set of distributed new energy equipment implementing the main plan; is the set of energy storage equipment implementing the main plan; is the set of flexible loads implementing the main plan; is the upward regulation power of the i-th flexible load at time t p ; is the downward regulation power of the i-th flexible load at time t p ; is the day-ahead peak shaving service demand of the virtual power plant at time t p .

[0128] The output constraint of the distributed new energy equipment is:

[0129]

[0130] The power constraint of the energy storage device is as follows:

[0131]

[0132] Equations (52) and (53) are the charge-discharge constraints corresponding to energy storage device i, and Equation (54) indicates that energy storage device i cannot be in both the charging and discharging states simultaneously at the same moment. Equations (55) and (56) represent the capacity constraints of energy storage device i.

[0133] Among them, is the Boolean variable for the charging of the i-th energy storage device at time t p ; is the Boolean variable for the discharging of the i-th energy storage device at time t p ; is the lower limit of the charging power of the i-th energy storage device at time t p ; is the upper limit of the charging power of the i-th energy storage device at time t p ; is the lower limit of the discharging power of the i-th energy storage device at time t p ; is the upper limit of the discharging power of the i-th energy storage device at time t p ; is the stored power of the i-th energy storage device at time t p ; is the stored power of the i-th energy storage device at time t p-1 ; is the discharging coefficient of the energy storage device; is the charging coefficient of the energy storage device; is the actual capacity of the i-th energy storage device at time t p ;

[0134] The up and down adjustment constraints of the flexible load are as follows:

[0135]

[0136] Among them, is the Boolean variable for the upward adjustment of the i-th flexible load at time t p ; is the Boolean variable for the downward adjustment of the i-th flexible load at time t p ; is the upper limit of the upward adjustment power of the i-th flexible load at time t p ; is the lower limit of the upward adjustment power of the i-th flexible load at time t p ; is at time tp The upper limit of the downward regulation power of the i-th flexible load at time t is t p The lower limit of the downward regulation power of the i-th flexible load at time t. As can be seen from formula (59), the flexible load point can only be in the upward or downward regulation state at the same time.

[0137] The operating cost of the real-time market is jointly determined by the output of distributed renewable energy devices, the charging and discharging conditions of energy storage, and the flexible load adjustment amount.

[0138]

[0139] Among them, is the operating coefficient of the virtual power plant operator in the real-time peak regulation ancillary service market; is the output of the i-th distributed new energy device at time t in the real-time peak regulation ancillary service market; β p time in the real-time peak regulation ancillary service market; β RG is the cost coefficient of the distributed new energy device in the real-time peak regulation ancillary service market; is the charging power of the i-th energy storage device at time t in the real-time peak regulation ancillary service market; p time in the real-time peak regulation ancillary service market; is the discharging power of the i-th energy storage device at time t in the real-time peak regulation ancillary service market; p time in the real-time peak regulation ancillary service market; β ES is the cost coefficient of the energy storage device in the real-time peak regulation ancillary service market; is the upward regulation power of the i-th flexible load at time t in the real-time peak regulation ancillary service market; p time in the real-time peak regulation ancillary service market; is the downward regulation power of the i-th flexible load at time t in the real-time peak regulation ancillary service market; p time in the real-time peak regulation ancillary service market; β FL is the cost coefficient of the flexible load in the real-time peak regulation ancillary service market. U is the uncertainty set corresponding to the demand in the real-time peak regulation ancillary service market.

[0140] Further introduce standardized auxiliary variables and to describe the uncertainty of the real-time market demand. The uncertainty set can be further defined as shown in the following formula:

[0141]

[0142] Δd C1 = μ d - 3σ d (61 - 4)

[0143]

[0144] Among them, and Δd C1 are the upper and lower limits of the demand deviation for the real-time market service, which can be calculated from Formulas (61-3) and (61-4). Formula (61-5) is the constraint for the auxiliary variable and . Π is the upper limit value of uncertainty, and the uncertainty intensity of the real-time market can be controlled by adjusting its value size.

[0145] In the real-time peaking auxiliary service market, the constraints on the output of each device of the virtual power plant by the day-ahead peaking auxiliary service market are as follows:

[0146]

[0147]

[0148] Among them, κ RG is the adjustment coefficient of the output of distributed new energy devices; κ ES+ is the adjustment coefficient of the charging power of energy storage devices; κ ES- is the adjustment coefficient of the discharging power of energy storage devices; κ FL+ is the adjustment coefficient of the upward power of flexible loads; κ FL- is the adjustment coefficient of the downward power of flexible loads.

[0149] In practical applications, considering the uncertainty of the "normal-fault" state of the physical devices participating in the main scheme, a reliability improvement reserve scheme model is established. Similar to the main scheme, the objective function of the reserve scheme is also to minimize the operating cost of the virtual power plant, and the second objective function is:

[0150]

[0151] Among them, is the set of distributed new energy devices implementing the auxiliary scheme; is the set of energy storage devices implementing the auxiliary scheme; is the set of flexible loads implementing the auxiliary scheme; is the output of the jth distributed new energy device at time t p ; is the charging power of the jth energy storage device at time t p ; is the discharging power of the jth energy storage device at time t p ; is the upward power of the jth flexible load at time t p ; is the downward power of the jth flexible load at time t p .

[0152] The output constraints of each device in the auxiliary scheme are similar to the constraints in the main scheme. However, the key difference is that the set of devices involved in the auxiliary scheme is different from that in the main scheme, and the specific constraints are expressed as follows:

[0153] The second set of constraint conditions is:

[0154]

[0155] Where, is the actual power generation capacity of the j-th distributed new energy device at time t p moment.

[0156]

[0157] Where, is the Boolean variable for the charging of the j-th energy storage device at time t p moment; is the lower limit of the charging power of the j-th energy storage device at time t p moment; is the upper limit of the charging power of the j-th energy storage device at time t p moment.

[0158]

[0159] Where, is the Boolean variable for the discharging of the j-th energy storage device at time t p moment; is the lower limit of the discharging power of the j-th energy storage device at time t p moment; is the upper limit of the discharging power of the j-th energy storage device at time t p moment.

[0160]

[0161] Where, is the stored electricity of the j-th energy storage device at time t p+1 moment; is the stored electricity of the j-th energy storage device at time t p-1 moment.

[0162]

[0163] Where, is the stored electricity of the j-th energy storage device at time t p moment; is the actual capacity of the i-th energy storage device at time t p moment.

[0164]

[0165] Among them, is the Boolean variable for the upward adjustment of the j-th flexible load at time t; p At time t, the Boolean variable for the upward adjustment of the j-th flexible load; is the upper limit of the upward adjustment power of the j-th flexible load at time t; p At time t, the upper limit of the upward adjustment power of the j-th flexible load; is the lower limit of the upward adjustment power of the j-th flexible load at time t. p At time t, the lower limit of the upward adjustment power of the j-th flexible load.

[0166]

[0167] Among them, is the Boolean variable for the downward adjustment of the j-th flexible load at time t; p At time t, the Boolean variable for the downward adjustment of the j-th flexible load; is the upper limit of the downward adjustment power of the j-th flexible load at time t; p At time t, the upper limit of the downward adjustment power of the j-th flexible load; is the lower limit of the downward adjustment power of the j-th flexible load at time t. p At time t, the lower limit of the downward adjustment power of the j-th flexible load.

[0168]

[0169] In addition, the total demand of the auxiliary plan is jointly affected by the day-ahead market peak shaving demand corresponding to the main plan and the output of distributed new energy equipment and energy storage systems participating in the main plan, as shown in formula (77):

[0170]

[0171] Among them, is the day-ahead peak shaving service demand of the virtual power plant at time t; p At time t, the day-ahead peak shaving service demand of the virtual power plant; is the day-ahead peak shaving service demand of the virtual power plant's main plan at time t; p At time t, the day-ahead peak shaving service demand of the virtual power plant's main plan; is the discrete random variable of the actual output of the i-th distributed new energy equipment at time t; p At time t, the discrete random variable of the actual output of the i-th distributed new energy equipment; is the discrete random variable of the actual output of the i-th energy storage device at time t. It is defined by the probability of its "normal - fault" state and the actual output in the main plan, as shown in formulas (78) - (79). In addition, the supply-demand balance of the auxiliary plan is shown in formula (80). p At time t, the discrete random variable of the actual output of the i-th energy storage device. It is defined by the probability of its "normal - fault" state and the actual output in the main plan, as shown in formulas (78) - (79). In addition, the supply-demand balance of the auxiliary plan is shown in formula (80).

[0172]

[0173] Among them, is the probability distribution that the output of the i-th distributed new energy equipment at time t follows; p At time t, the probability distribution that the output of the i-th distributed new energy equipment follows; is the probability distribution that the output of the i-th distributed new energy equipment at time t follows; pThe discrete random variable of the actual output at a moment is equal to the probability that the i-th distributed new energy device outputs power at t p moment; is the probability that the discrete random variable of the actual output of the i-th distributed new energy device at time t p is equal to 0; is the probability that the i-th device is in a normal state at time t p moment; is the probability that the i-th device is in a faulty state at time t p moment.

[0174]

[0175] Among them, is the probability distribution that the output of the i-th energy storage device at time t p obeys; is the probability that the discrete random variable of the actual output of the i-th energy storage device at time t p is equal to the difference between the discharge power of the i-th energy storage device at time t p and the charging power of the i-th energy storage device at time t p moment; is the probability that the discrete random variable of the actual output of the i-th energy storage device at time t p is equal to 0.

[0176]

[0177] S7: Solve the reliability optimization model considering the uncertainty of real-time peak shaving service demand and the reliability optimization model considering the uncertainty of the main scheme device state respectively to obtain optimization parameters; the optimization parameters include the output of distributed new energy devices, the charging power of energy storage devices, the discharge power of energy storage devices, the output of distributed new energy devices in the real-time peak shaving ancillary service market, the charging amount of energy storage devices in the real-time peak shaving ancillary service market, the discharge amount of energy storage devices in the real-time peak shaving ancillary service market, the downward adjustment amount of flexible load in the real-time peak shaving ancillary service market, and the upward adjustment amount of flexible load in the real-time peak shaving ancillary service market.

[0178] S8: Optimize the reliability index of the virtual power plant's full-stage tasks based on the optimization parameters to complete the optimization of the reliability of the virtual power plant's multi-stage peak shaving tasks.

[0179] The reliability optimization method for the virtual power plant's multi-stage peak shaving tasks in this application is as Figure 9As shown in the figure, first, based on the "normal - fault" state transition model of the cyber - physical device proposed in Step 1, further considering the characteristics of information transmission error codes, delays, and packet losses, the available rate model of the virtual power plant information channel in Step 2 is proposed; combined with the actual capacity of the physical device, the available rate model of the virtual power plant physical device in Step 3 is proposed.

[0180] Next, considering the actual response of the virtual power plant's actual single - stage peak - shaving task, analyzing the process of single - stage peak - shaving service failure, the fault tree and binary decision diagram model of the virtual power plant's single - stage task in Step 4 are proposed; based on the single - stage binary decision diagram model, considering the inter - stage coupling relationship of the virtual power plant's multi - stage peak - shaving tasks, the binary decision diagram model of the virtual power plant's full - stage in Step 5 is proposed.

[0181] Subsequently, based on the available rate model of the virtual power plant information channel, the available rate model of the physical device, and the binary decision diagram model of the virtual power plant's full - stage peak - shaving task, the probability reliability index RE1 of the full - stage peak - shaving task and the reliability index RE2 reflecting the service quality deviation are proposed.

[0182] Finally, to reduce the deviation of the virtual power plant's full - stage peak - shaving service and further improve the reliability of the virtual power plant's tasks, a main reliability improvement plan considering the uncertainty of real - time peak - shaving services in Step 7 and a backup reliability improvement plan considering the uncertainty of the device status of the main plan in Step 8 are proposed.

[0183] (1) This application proposes a "normal - fault" state transition model of the cyber - physical device. Based on the Markov process, combined with the historical statistical data of the failure rate and repair rate of the device's reliability index, the probabilities of the device being in the normal and fault states are obtained, which is the basic step for constructing the available rate model of the information channel and the available rate model of the physical device.

[0184] (2) This application proposes an available rate model of the virtual power plant information channel. Comprehensively considering the propagation delay effect of the communication system routing and the packet loss, switching delay, and error - code characteristics of the communication system node devices, the available rate model of the virtual power plant information channel is defined. In addition, combined with the state probabilities of the communication system line devices and node devices, the considered factors are relatively comprehensive, and using this model can make a relatively objective evaluation of the available rate of the virtual power plant information channel.

[0185] (3) This application proposes a model for the availability rate of physical devices in a virtual power plant. It comprehensively considers the state probability of physical system devices and the impact of the actual capacity of devices on the availability rate. For distributed new energy devices in the virtual power plant, a model for the actual power generation capacity considering the impact of natural disasters is proposed; for the distributed energy storage system in the virtual power plant, a model for the actual storage capacity considering calendar decay and cycle decay is proposed. The evaluation perspective is diverse, fully reflecting the relationship between the availability rate of physical devices and the reliability of the virtual power plant's peak shaving service in all stages.

[0186] (4) This application proposes a fault tree and binary decision diagram model for single-stage tasks in a virtual power plant. According to the actual response of the virtual power plant's peak shaving service, the failure process of single-stage peak shaving service is analyzed, and a fault tree and binary decision diagram model for single-stage tasks are proposed. The binary decision diagram can more vividly describe the relationship between each single-stage peak shaving subtask and the total task of the virtual power plant.

[0187] (5) This application proposes a binary decision diagram model for all stages of a virtual power plant. Based on the fault tree and binary decision diagram model for single-stage tasks in the virtual power plant, the inter-stage coupling relationship of multi-stage peak shaving tasks in the virtual power plant is fully considered. The constructed binary decision diagram model for all stages of the virtual power plant can clearly reflect the impact of each subtask of the cyber-physical system of the virtual power plant on the reliability of the all-stage peak shaving task, providing a new idea for the reliability assessment of the virtual power plant.

[0188] (6) This application proposes a reliability index for all-stage tasks in a virtual power plant. According to the binary decision diagram model of the all-stage peak shaving task in the virtual power plant, a probability reliability index RE1 for the all-stage peak shaving task is proposed; then, on the basis of probability reliability, considering the insufficient demand for peak shaving service, a reliability index RE2 reflecting the deviation of service quality is proposed. These two indexes can fully describe the comprehensive performance ability of the virtual power plant in the process of providing peak shaving service.

[0189] (7) This application proposes a main reliability improvement plan considering the uncertainty of real-time peak shaving service. Considering the uncertainty deviation between the day-ahead service demand and the real-time service demand in the process of peak shaving service, with the minimum operation cost of the virtual power plant as the objective function, a main reliability improvement plan model is proposed. By applying this model, the service demand deviation of the virtual power plant can be effectively reduced and the reliability of the virtual power plant can be improved.

[0190] (8) This application proposes a backup reliability improvement plan considering the uncertainty of the device state in the main plan. Considering the uncertainty of the device state in the main plan, the service demand deviation of the main plan is adjusted through the backup plan, and the reliability of the all-stage tasks of the virtual power plant is further improved from the perspective of the backup plan.

[0191] In an exemplary embodiment, a computer device is provided, which includes a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the above-mentioned reliability optimization method for the multi-stage peak shaving task of the virtual power plant is implemented.

[0192] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the above-mentioned reliability optimization method for the multi-stage peak shaving task of the virtual power plant is implemented.

[0193] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the above-mentioned reliability optimization method for the multi-stage peak shaving task of the virtual power plant is implemented.

[0194] In an exemplary embodiment, a computer device is provided. The computer device can be a server or a terminal, and its internal structure diagram can be as Figure 10 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, 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, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with external terminals through a network connection. When the computer program is executed by the processor, a reliability optimization method for the multi-stage peak shaving task of the virtual power plant is implemented.

[0195] Those skilled in the art can understand that Figure 10 the structure shown in

[0196] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0197] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0198] The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0199] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0200] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The descriptions of the above embodiments are only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A reliability optimization method for multi-stage peak load regulation tasks of a virtual power plant, characterized in that: include: Construct a normal-fault state transition model for cyber-physical devices; Establish a virtual power plant information channel availability model; The virtual power plant information channel availability model is determined based on packet loss, transmission delay and bit error availability of the transmission channel; Based on the normal state probability of the cyber-physical device in the normal-fault state transition model and the actual capacity of the cyber-physical device, a virtual power plant cyber-physical device availability model is established; Establishing a binary decision diagram model for all-stage tasks of a virtual power plant; the binary decision diagram model for all-stage tasks of a virtual power plant is established based on a binary decision diagram model for a single-stage task of a virtual power plant; the binary decision diagram model for a single-stage task of a virtual power plant is established based on a fault tree of a single-stage task of a virtual power plant using Shannon's decomposition law; Determine the reliability index of the virtual power plant's full-stage tasks according to the virtual power plant information channel availability model, the virtual power plant's cyber-physical equipment availability model and the virtual power plant's full-stage tasks binary decision diagram model; Establish a reliability optimization model that considers the uncertainty of real-time peak-shaving service demand and a reliability optimization model that considers the uncertainty of the main scheme equipment state; the reliability optimization model that considers the uncertainty of real-time peak-shaving service demand includes a first objective function and a first set of constraints; the first objective function aims to minimize the virtual power plant operator's operating coefficient in the day-ahead peak-shaving auxiliary service market and minimize the real-time peak-shaving auxiliary service market operating coefficient; the first set of constraints includes the active power balance constraint of the virtual power plant, the output constraint of distributed new energy equipment, the power constraint of energy storage equipment, the up and down adjustment constraint of the flexible load, and the constraint that each device of the virtual power plant in the real-time peak-shaving auxiliary service market is subject to the output of the day-ahead peak-shaving auxiliary service market; the reliability optimization model that considers the uncertainty of the main scheme equipment state includes a second objective function and a second set of constraints; the second objective function aims to minimize the virtual power plant operator's operating coefficient in the peak-shaving auxiliary service market; The reliability optimization model considering the uncertainty of the real-time peak-shaving service demand and the reliability optimization model considering the uncertainty of the main scheme equipment state are solved respectively to obtain optimization parameters; the optimization parameters include the output of distributed new energy equipment, the charging power of energy storage equipment, the discharging power of energy storage equipment, the output of distributed new energy equipment in the real-time peak-shaving auxiliary service market, the charging amount of energy storage equipment in the real-time peak-shaving auxiliary service market, the discharging amount of energy storage equipment in the real-time peak-shaving auxiliary service market, the downward adjustment amount of flexible load in the real-time peak-shaving auxiliary service market, and the upward adjustment amount of flexible load in the real-time peak-shaving auxiliary service market; The reliability index of the virtual power plant's full-stage mission is optimized based on the optimization parameters to complete the optimization of the reliability of the virtual power plant's multi-stage peak-shaving mission.

2. The reliability optimization method for multi-stage peak load regulation tasks of a virtual power plant according to claim 1 is characterized in that: The virtual power plant information channel availability model is: in, is the availability rate of the virtual power plant information channel; is the packet loss of the h-th transmission channel at time t; is the transmission delay of the hth transmission channel at time t; is the bit error availability rate of the hth transmission channel at time t.

3. The reliability optimization method for multi-stage peak load regulation tasks of a virtual power plant according to claim 1 is characterized in that: The availability model of the cyber-physical equipment of the virtual power plant is: in, is the state availability rate of the i-th cyber-physical device at time t; Pr i N (t) is the probability that the i-th cyber-physical device is in a normal state at time t; is the actual capacity of the i-th cyber-physical device at time t; DER is the distributed new energy device; ES is the energy storage device; Ω t is the set of physical equipment corresponding to time t; T is the time series set of multi-stage peak-shaving tasks of the virtual power plant.

4. The reliability optimization method for multi-stage peak load regulation tasks of a virtual power plant according to claim 1 is characterized in that: The first objective function is: in, The operating coefficient of the virtual power plant operator in the day-ahead peak load ancillary service market; The operating coefficients for virtual power plant operators in the real-time peak load ancillary services market; is the number of distributed new energy devices at t p Output at a given moment; α RG is the cost coefficient of distributed new energy equipment; is the energy storage device at t p Charging power at the moment; is the energy storage device at t p The discharge power at the moment; α ES is the cost coefficient of energy storage equipment; pha＝p is the time series set of the peak load-shaving tasks of the virtual power plant in stage p; A set of distributed new energy equipment to execute the main plan; is the set of energy storage devices that execute the main plan; U is the uncertainty set corresponding to the market demand for real-time peak-shaving auxiliary services; is the i-th distributed new energy device t p The contribution of real-time peak load ancillary service market at all times; RG The cost coefficient of distributed new energy equipment in the real-time peak load auxiliary service market; is the i-th energy storage device t p Charging power in the real-time peak load ancillary service market at all times; is the i-th energy storage device t p The discharge power in the real-time peak load auxiliary service market at all times; β ES The cost coefficient of energy storage equipment in the real-time peak load ancillary service market; is the i-th flexibility load t p Always adjust the power in the real-time peak load ancillary service market; is the i-th flexibility load t p The power reduction in the real-time peak load ancillary service market; FL The cost coefficient of the flexible load in the real-time peak load ancillary service market; The flexibility load set for executing the main plan.

5. The reliability optimization method for multi-stage peak load regulation tasks of a virtual power plant according to claim 1 is characterized in that: The active power balance constraint of the virtual power plant is: in, is the number of distributed new energy devices at t p The effort of the moment; is the energy storage device at t p Charging power at the moment; is the energy storage device at t p Discharge power at the moment; A set of distributed new energy equipment to execute the main plan; A set of energy storage devices for executing the main scheme; The flexibility load set for executing the main plan; is the i-th flexibility load t p Increase power at all times; is the i-th flexibility load t p Reduce power at all times; t p The demand for day-ahead peak-shaving services of the virtual power plant at each moment; The output constraints of the distributed new energy equipment are: in, is the number of distributed new energy devices at t p Actual power generation capacity at the moment; The power constraint of the energy storage device is: in, t p A Boolean variable indicating the charging status of the i-th energy storage device at the moment; for p A Boolean variable indicating the discharge of the i-th energy storage device at the moment; t p The lower limit of the charging power of the i-th energy storage device at the moment; t p The upper limit of charging power of the i-th energy storage device at the moment; t p The lower limit of the discharge power of the i-th energy storage device at the moment; t p The upper limit of the discharge power of the i-th energy storage device at the moment; t p The storage capacity of the i-th energy storage device at the moment; t p-1 The storage capacity of the i-th energy storage device at the moment; is the discharge coefficient of the energy storage device; is the charging factor of the energy storage device; t p The actual capacity of the i-th energy storage device at the moment; The upward and downward adjustment constraints of the flexibility load are: in, t p Boolean variable indicating the upward adjustment of the i-th flexibility load at the moment; t p Boolean variable indicating the downward adjustment of the i-th flexibility load at time; t p The upper limit of the power increase of the i-th flexible load at the moment; t p The lower limit of the power increase of the i-th flexible load at the moment; t p The upper limit of the power reduction of the i-th flexible load at the moment; t p The lower limit of the power reduction of the i-th flexible load at the moment; The equipment of the virtual power plant in the real-time peak-shaving auxiliary service market is subject to the output constraints of the day-ahead peak-shaving auxiliary service market: Among them, κ RG is the adjustment coefficient of the output of distributed new energy equipment; κ ES+ is the adjustment coefficient of the charging power of the energy storage device; κ ES- is the adjustment coefficient of the discharge power of the energy storage device; κ FL+ The adjustment factor for increasing the power of the flexible load; κ FL- The adjustment factor for reducing power for flexible loads.

6. The reliability optimization method for multi-stage peak load regulation tasks of a virtual power plant according to claim 1 is characterized in that: The second objective function is: Among them, pha＝p is the time series set of the peak load-shaving tasks of the virtual power plant in stage p; A set of distributed new energy devices to implement auxiliary solutions; A set of energy storage devices for executing auxiliary schemes; A set of flexible loads for implementing auxiliary schemes; is the number of distributed new energy devices at t p The effort of the moment; is the jth energy storage device at t p Charging power at the moment; is the jth energy storage device at t p Discharge power at the moment; is the jth flexibility load t p Increase power at all times; is the jth flexibility load t p Reduce power at all times.

7. The reliability optimization method for multi-stage peak load regulation tasks of a virtual power plant according to claim 1 is characterized in that: The second set of constraints is: in, A set of distributed new energy devices to implement auxiliary solutions; is the number of distributed new energy devices at t p Actual power generation capacity at the moment; is the number of distributed new energy devices at t p The effort of the moment; in, A set of energy storage devices for executing auxiliary schemes; t p Boolean variable indicating the charge of the jth energy storage device at the moment; is the jth energy storage device at t p Charging power at the moment; t p The lower limit of the charging power of the jth energy storage device at the moment; t p The upper limit of charging power of the jth energy storage device at the moment; in, for p Boolean variable indicating the discharge of the jth energy storage device at the moment; is the jth energy storage device at t p Discharge power at the moment; t p The lower limit of the discharge power of the jth energy storage device at the moment; t p The upper limit of the discharge power of the jth energy storage device at the moment; in, t p+1 The storage capacity of the jth energy storage device at the moment; t p-1 The storage capacity of the jth energy storage device at the moment; is the discharge coefficient of the energy storage device; is the charging factor of the energy storage device; in, t p The storage capacity of the jth energy storage device at the moment; t p The actual capacity of the i-th energy storage device at the moment; in, A set of flexible loads for implementing auxiliary schemes; t p Boolean variable indicating the increase in flexibility load at time j; is the jth flexibility load t p Increase power at all times; t p The upper limit of the power increase of the j-th flexible load at the moment; t p The lower limit of the power increase of the jth flexible load at the moment; in, t p Boolean variable indicating the jth flexibility load reduction at time instant; is the jth flexibility load t p Reduce power at all times; t p The upper limit of the power reduction of the j-th flexible load at the moment; t p The lower limit of the power reduction of the jth flexible load at the moment; in, t p The auxiliary solution of the virtual power plant at any time can provide peak load service demand on the day before; t p The day-ahead peak load-shaving service demand of the main scheme of the virtual power plant; A set of distributed new energy equipment to execute the main plan; A set of energy storage devices for executing the main scheme; The flexibility load set for executing the main plan; is the i-th distributed new energy device t p The discrete random variable of actual output at any moment; is the i-th energy storage device t p The discrete random variable of actual output at any moment; is the i-th flexibility load t p Increase power at all times; is the i-th flexibility load t p Reduce power at all times; in, is the i-th distributed new energy device t p The probability distribution of the output at each moment; is the i-th distributed new energy device t p The discrete random variable of actual output at time t is equal to the value of the i-th distributed new energy device at t p The probability of output at a given moment; is the number of distributed new energy devices at t p The effort of the moment; is the i-th distributed new energy device t p The probability that the discrete random variable of actual output at a given moment is equal to 0; Pr i N (t p ) is the i-th device t p The probability of being in a normal state at all times; Pr i F (t p ) is the i-th device t p The probability of being in a fault state at all times; in, is the i-th energy storage device t p The probability distribution of the output at each moment; is the i-th energy storage device t p The discrete random variable of actual output at time t is equal to the value of the ith energy storage device at t p The discharge power at time t is related to the discharge power of the i-th energy storage device at t p The probability of the difference in charging power at the moment; is the energy storage device at t p Discharge power at the moment; is the energy storage device at t p Charging power at the moment; is the i-th energy storage device t p The probability that the discrete random variable of actual output at a given moment is equal to 0; 8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the reliability optimization method for multi-stage peak-shaving tasks of a virtual power plant according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the reliability optimization method for multi-stage peak-shaving tasks of a virtual power plant as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the reliability optimization method for multi-stage peak-shaving tasks of a virtual power plant as described in any one of claims 1-7.

Images