Virtual power plant equipment heterogeneous communication network access optimization method and system considering non-ideal transmission

Abstract

The invention provides a virtual power plant equipment heterogeneous communication network access optimization method and system considering non-ideal transmission, belongs to the technical field of power grid system resource optimization configuration, and is based on constraint conditions of charge and discharge capacity, power output characteristics, network access time delay and network bearing capacity of virtual power plant VPP equipment. A heterogeneous network communication time delay model and a data packet loss model are constructed by combining data packet loss and access time delay of equipment under different network transmission mechanisms, a distributed equipment power output characteristic and charging and discharging limitation are considered by taking benefit loss minimization as a target, an equipment access optimization model is constructed, optimization solution is performed by combining hierarchical solution and a greedy algorithm, and a network access optimization model is constructed. And obtaining an equipment access scheme. According to the invention, the distribution of network resources is optimized, and the purposes of service guarantee and cost minimization are achieved; by matching devices with different priorities and proper network performance, the negative influence of communication uncertainty on VPP income is reduced, and the income tends to be dynamically balanced.

Description

Technical Field

[0001] The present invention relates to the technical field of optimal allocation of power grid system resources, and particularly to an optimization method and system for accessing heterogeneous communication networks of virtual power plant devices considering non-ideal transmission. Background Art

[0002] At present, a large number of distributed new energies are gradually becoming an important part of the power system. A virtual power plant (VPP) can use heterogeneous networks to achieve aggregated scheduling of distributed new energies and improve comprehensive benefits. To meet the requirements of supply-demand matching and flexible operation of the new power system, the VPP forms a virtual power generation entity with elasticity and high efficiency by integrating distributed energies such as solar energy, wind energy, and energy storage distributed in different geographical locations, thereby enhancing the flexible regulation ability of the power system.

[0003] Wireless communication networks are easy to deploy and expand, and device access is flexible, which are widely used in the VPP device access network. However, the open and shared propagation environment, as well as random channel fading and noise, cause transmission delay and data packet loss in wireless networks, resulting in uncertainty in the VPP scheduling operation, thereby reducing the expected benefits. It is necessary to model, analyze, and solve the impact of non-ideal network access on the VPP and its scheduling performance.

[0004] The operation of the VPP requires the cooperation of various distributed energies to achieve the purpose of outputting reliable, flexible, and high-quality electric energy. Most of the existing related research on the operation of the VPP only focuses on the operation scheduling mechanism within the power system. In order to further strengthen the scheduling ability of the VPP for distributed new energies and consider the adverse effects such as network communication data congestion and system capacity fluctuations during the scheduling process, the existing simulation methods are used to analyze and quantify the impact of communication network performance on power grid operation. The above research has been deeply explored in the field of power grid intelligence and has made important contributions to the optimization of power system communication scheduling. However, the current research is not sufficient in considering the communication uncertainty factors during the VPP scheduling process, especially the consideration of the impact of packet loss rate and delay is not comprehensive enough. Summary of the Invention

[0005] The purpose of the present invention is to provide an optimization method and system for accessing heterogeneous communication networks of virtual power plant devices considering non-ideal transmission to solve at least one of the technical problems in the above background art.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions:

[0007] In the first aspect, the present invention provides an optimization method for accessing heterogeneous communication networks of virtual power plant devices considering non-ideal transmission, including:

[0008] Based on the constraint conditions of the charging and discharging capacity, power output characteristics, access network delay, and network carrying capacity of virtual power plant (VPP) devices, combined with data packet loss and device access delay under different network transmission mechanisms, a heterogeneous network communication delay model and a data packet loss model are constructed;

[0009] Based on the heterogeneous network communication delay model and the data packet loss model, with the goal of minimizing revenue loss, considering the power output characteristics and charging and discharging limitations of distributed devices, the impact of communication delay and data packet loss rate on VPP revenue is quantified, and a device access optimization model is constructed;

[0010] Combined with hierarchical solution and greedy algorithm, the device access optimization model is optimized and solved to obtain a device access plan.

[0011] As a further limitation of the first aspect of the present invention, constructing a heterogeneous network communication delay model includes: according to the radius of the cellular network coverage area and the number of distributed devices with access requirements evenly distributed in the area, calculating the collision probability of the devices initiating random access requests in the i-th random access process, and calculating the average access delay of the devices during the access process according to the collision probability.

[0012] As a further limitation of the first aspect of the present invention, the cellular network coverage area is a radius R co , and n distributed devices with access requirements are evenly distributed in the area. In the i-th random access process, the collision probability of the n i devices initiating random access requests is:

[0013]

[0014] In the formula, N q is the number of preamble sequences provided by the access node;

[0015] Then, the average access delay of the device is:

[0016]

[0017] Among them, D s is the average transmission delay, and T′ is the transmission period.

[0018] As a further limitation of the first aspect of the present invention, a heterogeneous network communication data packet loss model is constructed, including: calculating the probability of packet loss and delay generated by a device according to the probability of non-ideal communication occurring during the data interaction process between the device and VPP; the probability of packet loss of a device due to collision increases as the probability of data packet collision increases. Each competing device randomly selects an integer as a counter, and when the counter counts down to 0, transmission is triggered; when the retransmitted packet collides again, retransmission is triggered, the counter is randomly selected again, and the counter counts down; until the number of retransmissions reaches the maximum backoff stage; a non-linear equation regarding the probability of the device transmitting data packets, bandwidth, and collision probability is obtained by solving the steady-state distribution of the Markov chain.

[0019] As a further limitation of the first aspect of the present invention, constructing a device access model includes:

[0020] The optimization objective of the VPP scheduling side is to minimize the revenue loss caused by factors such as delay jitter and data packet loss generated by the access of distributed energy devices. VPP issues power regulation instructions to distributed devices according to the requirements of the distribution operator, obtains rewards from the operator, and pays scheduling fees for the devices participating in the response; the difference between the rewards obtained from the distribution operator and the scheduling payment is the net revenue obtained by VPP.

[0021] According to the virtual power plant revenue model, considering issues such as reliability and effectiveness during the communication process between the device and VPP, the two influencing factors of data packet loss and communication delay are quantified. VPP selects the scheduling instruction with the maximum revenue according to different network access delays and device feedback information. The difference between the revenue of VPP in the case of non-ideal uplink transmission and non-ideal downlink transmission and the revenue obtained by VPP in the case of ideal error-free transmission is the revenue loss caused by non-ideal communication.

[0022] As a further limitation of the first aspect of the present invention, during non-ideal uplink transmission, VPP cannot receive the status information of the device in a timely and accurate manner. At this time, VPP makes a scheduling strategy based on n the historical state distribution and delay probability distribution of the device to maximize its expected revenue.

[0023] During non-ideal downlink transmission, since the device cannot receive the scheduling instruction accurately and in a timely manner, if the scheduling instruction is discarded due to data packet loss or excessive delay, VPP needs to fully bear the revenue loss and determine the expected revenue at this time.

[0024] To measure the revenue risk brought by communication uncertainty to the VPP scheduling operation, considering the power output characteristics of distributed devices and the communication packet loss and delay of the devices, minimizing the revenue loss of VPP device access competition, and determining the objective function.

[0025] Second aspect, the present invention provides a virtual power plant device heterogeneous communication network access optimization system considering non-ideal transmission, including:

[0026] A first construction module, configured to construct a heterogeneous network communication delay model and a data packet loss model based on the constraint conditions of the charging and discharging capacity, power output characteristics, access network delay, and network carrying capacity of the virtual power plant VPP device, in combination with the data packet loss under different network transmission mechanisms and the access delay of the device;

[0027] A second construction module, configured to construct a device access optimization model by quantifying the impact of communication delay and data packet loss rate on the VPP revenue with the goal of minimizing revenue loss, considering the power output characteristics and charging and discharging limitations of distributed devices based on the heterogeneous network communication delay model and the data packet loss model;

[0028] A solving module, configured to optimize and solve the device access optimization model by combining hierarchical solving and a greedy algorithm to obtain a device access scheme.

[0029] Third aspect, the present invention provides a non-transitory computer-readable storage medium, which is used to store computer instructions. When the computer instructions are executed by a processor, the virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission as described in the first aspect is implemented.

[0030] Fourth aspect, the present invention provides a computer device, including a memory and a processor, where the processor and the memory communicate with each other. The memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission as described in the first aspect.

[0031] Fifth aspect, the present invention provides an electronic device, including: a processor, a memory, and a computer program; wherein, the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device runs, the processor executes the computer program stored in the memory so that the electronic device executes the instructions for implementing the virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission as described in the first aspect.

[0032] Advantages of the present invention: Considering the charging and discharging constraints of devices and communication uncertainty factors, a mixed-integer non-linear programming model is constructed, and the device access process is optimized through a heuristic selection algorithm; by flexibly selecting the access network type, the model optimizes the allocation of network resources according to the urgency of device services, achieving the goals of service guarantee and cost minimization; by matching devices with different priorities to appropriate network performances, the negative impact of communication uncertainty on the VPP revenue is reduced, making the revenue tend to dynamic balance; the optimization model starts from the perspective of heterogeneous communication networks, comprehensively considering the dual effects of packet loss and delay on device scheduling, and is closer to the actual situation compared with the traditional model that only considers packet loss.

[0033] Advantages of additional aspects of the present invention will be more clearly given in the following description part, or understood through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0035] Figure 1 Schematic diagram of the VPP heterogeneous network communication scheduling architecture described in the embodiment of the present invention.

[0036] Figure 2 Schematic diagram of the VPP communication network access framework described in the embodiment of the present invention.

[0037] Figure 3 Schematic diagram of the VPP communication interaction process described in the embodiment of the present invention.

[0038] Figure 4 Schematic diagram of the decomposition logic of the model optimization solution algorithm described in the embodiment of the present invention.

[0039] Figure 5 Schematic diagram of the iterative solution process of the model target problem described in the embodiment of the present invention.

[0040] Figure 6 Schematic diagram of the heuristic selection algorithm process described in the embodiment of the present invention.

[0041] Figure 7 Schematic diagram of the fluctuation of the non-ideal communication occurrence probability with the number of access devices described in the embodiment of the present invention. Among them, Figure 7 (a) is the fluctuation under the 5G network, Figure 7 (b) is the fluctuation under the LTE network, Figure 7 (c) is the fluctuation under the WiFi network.

[0042] Figure 8 It is a diagram showing the relationship between network access delay and the number of devices according to the embodiments of the present invention.

[0043] Figure 9 It is a schematic diagram showing the distribution of different devices and APs within the scheduling area according to the embodiments of the present invention.

[0044] Figure 10 It is a schematic diagram showing the iterative results of VPP revenue loss and the probability of non-ideal communication according to the embodiments of the present invention. Among them, Figure 10 (a) is the residential area, Figure 10 (b) is the power generation area, Figure 10 (c) is the energy storage area.

[0045] Figure 11 It is a schematic diagram showing the iterative results of VPP revenue loss and the probability of non-ideal communication (including delay) according to the embodiments of the present invention. Among them, Figure 11 (a) is the residential area, Figure 11 (b) is the power generation area, Figure 11 (c) is the energy storage area.

[0046] Figure 12 It is a schematic diagram showing the change of the probability of non-ideal communication with the number of iterations according to the embodiments of the present invention. Among them Figure 12 (a) is the case without considering delay, Figure 12 (b) is the case considering delay.

[0047] Figure 13 It is a schematic diagram showing the final device access volume according to the embodiments of the present invention. Detailed implementation manners

[0048] The following details the implementation manners of the present invention. Examples of the implementation manners are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The implementation manners described through the drawings are exemplary and are only used to explain the present invention, and cannot be construed as a limitation to the present invention.

[0049] Those skilled in the art of the present technology can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used here have the same meaning as the general understanding of those of ordinary skill in the art in the field to which the present invention belongs.

[0050] It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless defined as here.

[0051] Those skilled in the art can understand that, unless specifically stated otherwise, the singular forms "a", "an", "the" and "said" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the description of the present invention means the presence of the described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements and / or their groups.

[0052] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. Without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0053] For ease of understanding of the present invention, the following will further explain the present invention with specific embodiments in conjunction with the accompanying drawings, and the specific embodiments do not constitute a limitation to the embodiments of the present invention.

[0054] Those skilled in the art should understand that the drawings are only schematic diagrams of the embodiments, and the components in the drawings are not necessarily essential for implementing the present invention.

[0055] Embodiment 1

[0056] In this Embodiment 1, first, an optimization system for accessing heterogeneous communication networks of virtual power plant equipment considering non-ideal transmission is provided. The system includes: a first construction module, which is used to construct a heterogeneous network communication delay model and a data packet loss model based on the constraint conditions of the charging and discharging capacity, power output characteristics, access network delay and network carrying capacity of the virtual power plant VPP equipment, combined with the data packet loss under different network transmission mechanisms and the access delay of the equipment; a second construction module, which is used to construct an equipment access optimization model by quantifying the impact of communication delay and data packet loss rate on the VPP revenue with the goal of minimizing revenue loss, based on the heterogeneous network communication delay model and the data packet loss model, considering the power output characteristics and charging and discharging limitations of distributed equipment; a solving module, which is used to optimize and solve the equipment access optimization model by combining hierarchical solving and greedy algorithm to obtain an equipment access plan.

[0057] In this embodiment, the above system is used to implement an optimization method for accessing heterogeneous communication networks of virtual power plant equipment considering non-ideal transmission, including: using a first construction module, based on the constraints of the charge-discharge capacity, power output characteristics, access network delay, and network carrying capacity of virtual power plant VPP equipment, combined with data packet loss and equipment access delay under different network transmission mechanisms, constructing a heterogeneous network communication delay model and a data packet loss model; using a second construction module, based on the heterogeneous network communication delay model and the data packet loss model, considering the power output characteristics and charge-discharge limitations of distributed equipment with the goal of minimizing revenue loss, quantifying the impact of communication delay and data packet loss rate on VPP revenue, and constructing an equipment access optimization model; finally, using a solution module to optimize and solve the equipment access optimization model by combining hierarchical solution and greedy algorithm to obtain an equipment access plan.

[0058] Specifically, in this embodiment, the implementation process of the above method is described in detail as follows:

[0059] The structure of the VPP communication scheduling network is as Figure 1 shown, mainly including an equipment response layer, a group access layer, and a top-level scheduling layer. Distributed equipment such as energy storage equipment, wind and solar power generation equipment, and electric vehicles in the equipment response layer establish connections with the VPP through heterogeneous wireless access nodes (Access Point, AP) of the group access center to achieve two-way transmission and interaction of uplink equipment collected data and downlink scheduling instructions.

[0060] In the uplink, equipment such as energy storage and electric vehicles upload accurate real-time status information u n ={q, S t , η ch , η dis , T} to the VPP, including: the real-time power load q of the equipment; the real-time charging status S t ; the charge-discharge efficiency η ch and η dis ; and data such as the maximum remaining stay time T of the electric vehicle at the charging station. The VPP executes a scheduling algorithm based on the received equipment status information u n and communication delay t n etc., and obtains the best downlink scheduling action d n to maximize the comprehensive revenue of the VPP.

[0061] Figure 2The up and down data transmission processes between the VPP and the devices are presented. Obviously, scheduling needs to consider the non-ideal transmission of communication data. In the uplink data transmission link, due to network packet loss and transmission delay, the VPP cannot accurately, timely, and effectively collect the real-time status information of the devices, and thus cannot obtain the best decision-making and scheduling instructions. Similarly, in the downlink transmission, if the scheduling instructions from the VPP cannot be accurately and timely received, the devices will not be able to execute their scheduling arrangements as planned.

[0062] During the device scheduling process, the interaction steps among the device side, the power plant side, and the operator side are as Figure 3 shown. The VPP, according to the power scheduling requirements: ① issues instructions for collecting the real-time status information of the devices, ② the energy device data is transmitted back, ③ the VPP scheduling strategy, and ④ the scheduling instructions are issued. Corresponding communication delays and data packet losses occur during this process. The delays include data access delay, data transmission delay, and data processing delay. The energy devices also need to periodically transmit back data.

[0063] For the charging and discharging constraint model of distributed devices, the VPP obtains benefits by aggregating and scheduling distributed new energy devices. To analyze the key influencing factors of scheduling, it is necessary to analyze and model the charging and discharging characteristics and parameters of key devices, so as to perform scheduling under the conditions of meeting their respective charging and discharging constraints.

[0064] The output power of a wind turbine is related to the natural wind speed:

[0065]

[0066] In the formula: q WT is the output power of the wind turbine; v is the natural wind speed; v in is the cut-in wind speed of the wind turbine; is the rated output power of the wind turbine; v rate is the rated wind speed of the wind turbine; v out is the cut-out wind speed of the wind turbine.

[0067] The output power of a photovoltaic power generation device is related to the light intensity:

[0068]

[0069] In the formula: q PV is the output power of the photovoltaic power generation device; is the rated output power of the photovoltaic power generation device; s is the light intensity; s rate is the rated light intensity. According to the internationally accepted standard test conditions, its value is generally 1000 W / m 2 .

[0070] When the VPP cannot obtain the device status information accurately and in a timely manner, the VPP can issue instructions based on the historical state distribution data of the device. Without loss of generality, assume that the historical state of the distributed generation device follows a Weibull distribution, and the probability density function is:

[0071]

[0072] Where: q w,t and q w,e are the output power and rated power at time t, respectively; μ and σ are the mean and variance of the device power prediction error, respectively.

[0073] The prerequisite for the scheduling of distributed devices is that the difference Δq between the real-time power q of the device and the planned scheduling power q′ is within the feasible scheduling region For energy storage devices, it is further restricted by the charging and discharging behavior of the device and the upper and lower limits of the state of charge (SOC) of the device:

[0074] q′ = q′ ch -q′ dis (4)

[0075]

[0076] S min ≤ S t ≤ S max (6)

[0077] Where: q′ ch and q′ dis are the charging and discharging powers; S t and S t+1 are the SOCs at times t and t + 1, respectively; ⊙ is the Hadamard product; η ch and η dis are the charging and discharging efficiencies; E is the energy storage capacity of the device; Δt is the time interval; S min and S max are the minimum and maximum state of charge limits of the device, respectively.

[0078] The scheduling of electric vehicle equipment is also restricted by the charging time at the charging station. If the planned stay time of the electric vehicle at the charging station does not exceed the time interval Δt, then continuous scheduling power cannot be provided during Δt. When the scheduled electric vehicle leaves the charging station, its SOC requirement must be met, and its feasible scheduling region is further restricted:

[0079]

[0080] Δq = (q - q′) ⊙ y (8)

[0081] S exit ≤ η ch ⊙ q max ·(T - Δt) + S t+1 (9)

[0082] Wherein, T is the maximum residence time of the device at the charging station; y is a binary variable: if the residence time of device n at the charging station does not exceed Δt, y = 0, otherwise it is 1. The vehicle immediately charges at the maximum charging power q max after scheduling and should be greater than the exit state of charge S exit .

[0083] For the VPP network delay performance analysis, the two-way interaction delay between VPP and distributed energy devices is related to parameters such as the access algorithm of the communication network, the number of devices, and the coverage distance. The transmission delay is mainly determined by factors such as the packet length, coding efficiency, and channel bandwidth. Assume that the coverage range of the cellular network is a radius R co , and n distributed devices uniformly distributed in the area have access requirements. During the i-th random access process, the collision probability of the n i -th device initiating a random access request is:

[0084] Wherein, N q is the number of preamble sequences provided by the access node.

[0085] Then during this access process, the average access delay of the device can be expressed as:

[0086]

[0087] Wherein: D s is the average transmission delay, which can be theoretically estimated. Assume that it follows a uniform distribution. After the device establishes a session through four-way handshake, the average transmission delay of the message is 4D s , and T' is the transmission period.

[0088] The WiFi network adopts the IEEE802.11n standard, and users in the area follow the CSMA / CA protocol to compete for access to the channel. The average access delay D wi is:

[0089] D wi = D B + D c + D v (12)

[0090] D c = n bp t c (13)

[0091] where: D c is the collision delay; n bp is the number of data packet transmission collisions; t c is the time taken for each collision.

[0092] where: t p is the physical layer preamble and header transmission time; H is the MAC header length; L is the data packet length; C is the channel bit rate; t ELIFS is the duration of a short frame.

[0093] The average backoff delay D B represents the sum of the delays experienced in each backoff before successful transmission. Each backoff delay is equal to the product of the backoff timer b i and the mean value of the slot length t slot :

[0094] where: represents the calculation of the mean value; the backoff timer b i is uniformly distributed in [0, CW i -1]; W i is the contention window determined in the protocol, and the expected value of b i can be expressed as:

[0095]

[0096] where: P NS is the probability that a user successfully transmits among the remaining N - 1 users when the user does not transmit data; t NS , t NC and t E are the times corresponding to the occurrence of events: P NS =(N - 1)τ s (1 - τ s ) N-2 (18)

[0097] where τ s is the probability that a user attempts to transmit data in a random time slot. Assuming it is a constant, the probability P E that the channel is idle can be expressed as: P E =(1 - τ s ) N-1 (19)

[0098] P NC represents the probability of a collision occurring among the remaining N - 1 users: P NC =1 - P NS - P E (20)

[0099] According to the CSMA / CA protocol, the data successful transmission delay D v is directly related to the ACK frame length A, the channel bit rate C, the duration t between short frames SIFS and so on:

[0100]

[0101] For the communication packet loss performance modeling of VPP scheduling, during the interaction between the energy device and the VPP, if there are too many devices connected to the same AP, it will inevitably cause an increase in data transmission delay and the number of collision packet losses. Let p represent the probability of non-ideal communication occurring during the data interaction between the device and the VPP:

[0102] where: C m,n is the element corresponding to the m-th row and n-th column of the connection matrix of the AP; row m corresponds to the AP; column n corresponds to the device. If device n is connected to the AP m for data transmission and reception, then C m,n = 1, otherwise C m,n = 0. ω m is the probability of packet loss and delay generated by the corresponding device.

[0103] For the WiFi network:

[0104] where: ω m,SNR is the packet loss rate generated by the current transmitted data when the signal-to-noise ratio is lower than the interruption threshold; ω m,co is the device packet loss probability due to collision in the WiFi network, which increases with the increase of the packet collision probability α, and α is positively correlated with the number of devices:

[0105] where, i max is the maximum limit of the number of retransmission times. According to the CSMA / CA protocol, each competing device randomly selects an integer from [0, W i as the counter. When the counter counts down to 0, the transmission is triggered. When the retransmitted packet collides again, the retransmission is triggered, and the counter randomly selects from [0, 2 i W i - 1], and the counter counts down. If a collision occurs again, continue to repeat the previous step until i reaches the maximum backoff stage i′ max .

[0106] This process can be regarded as a Markov chain. By solving the steady-state distribution of the Markov chain, the following non-linear equations about the probability τ of the device transmitting data packets, the bandwidth B, and the collision probability α can be obtained:

[0107]

[0108] α = 1 - (1 - τ) N-1 (26)

[0109]

[0110] ω delay represents the probability that the actual transmission and processing delay D(t) of the transmitted service data does not exceed the theoretical delay D co , and this probability should be less than the maximum tolerable error rate ε corresponding to the transmission service:

[0111] ω delay = P{D(t) < D co} ≤ ε (28)

[0112]

[0113] v i = B i log2(1 + S SNR ) (30)

[0114]

[0115] Where: v i is the data transmission rate of the i-th device; B i is the channel bandwidth of the i-th device; S SNR is the signal-to-noise ratio.

[0116] For a cellular network, the access node controls all communication resources and allocates these resources to devices according to certain rules. LTE adopts orthogonal frequency division multiple access (OFDMA), which divides the entire time-frequency resource into smaller micro-slots to provide access for users. According to the 3GPP standard, there are b RB-srf time slots in a sub-frame with a length of 1 ms, R B time slots are reserved for the control channel, and the remaining time slots are used for data transmission. Each R B carries r RB bits of data. Therefore, the total R B required by the device is: b n = [r / r RB (32)

[0117] Where [.] is the ceiling function, and r is the total number of bits to be sent.

[0118] Assume that each device requires b n time slots to transmit data, and the remaining b r time slots are idle, N mIf \(N\) is the maximum number of devices that an AP can serve, the packet loss rate, which cannot meet the service requirements of other devices due to limited transmission resources, can be expressed as:

[0119]

[0120] \(k\) represents that \(k\) users have completed data transmission before the maximum deadline \(t\). d

[0121] The revenue modeling of VPP scheduling is as follows:

[0122] The optimization goal on the VPP scheduling side is to minimize the revenue loss \(O\) caused by factors such as delay jitter and data packet loss resulting from the access of distributed energy devices. The VPP issues power adjustment instructions to distributed devices according to the requirements of the distribution operator, obtains rewards from the operator, and pays scheduling fees to the devices participating in the response. The difference between the rewards obtained from the distribution operator and the scheduling payments is the net revenue obtained by the VPP. The following formula is the virtual power plant revenue model:

[0123]

[0124] In the formula: \(d\) n , \(u\) n are respectively the communication instructions issued by the virtual power plant and the device status information uploaded by the distributed device; \(R\) is the scheduling fee paid by the power supply operator to the virtual power plant; \(\mu\) n is the cost (in yuan / kw) paid by the virtual power plant to distributed device \(n\); \(\Delta q\) n is the scheduling power of the corresponding device; is the feasible scheduling set of the device. When , a quantity of negative infinity, \(-\infty\), is used to represent that the current device is not schedulable.

[0125] Considering issues such as reliability and effectiveness in the communication process between the device and the VPP, the two influencing factors of data packet loss and communication delay are quantified:

[0126]

[0127] In the formula: \(F\) p represents the revenue obtained by the VPP in the case of ideal error-free transmission; the VPP selects the scheduling instruction \(d\) n that maximizes the revenue according to different network access delays \(t\) n and the device feedback information \(u\); \(p\) n , \(p\) up , \(p\) down are respectively the probabilities of non-ideal transmission in the uplink and downlink. Since the probability of simultaneous errors in the uplink and downlink is relatively small, \(p\) up and \(p\) down ​The product is close to 0, so the model is approximately simplified, and the revenue loss F when both the uplink and downlink have errors simultaneously is ignored updown ; F up 、F down are the revenue situations of VPP in the cases of non-ideal uplink transmission and non-ideal downlink transmission respectively. Subtracting them from F p represents the revenue loss caused by non-ideal communication. The device n receives and executes the instruction d n The revenue F p,n The expected value is as follows:

[0128] In the case of non-ideal uplink transmission, VPP cannot receive the device status information in a timely and accurate manner. At this time, VPP makes a scheduling strategy based on the historical status distribution and delay probability distribution of device n to maximize its revenue expectation:

[0129] In the case of non-ideal downlink transmission, since the device cannot receive the scheduling instruction accurately and in a timely manner, if the scheduling instruction is discarded due to data packet loss or excessive delay, VPP has to bear the entire revenue loss. The revenue expectation at this time can be expressed as:

[0130]

[0131] The power characteristics and charge-discharge states of different distributed devices have their own characteristics. For a device to participate in scheduling, not only does the current real-time power of the device need to support the scheduling requirements, but also its own charge-discharge limitations need to be met. To measure the revenue risk brought by communication uncertainty to the VPP scheduling operation, considering the power output characteristics of distributed devices and the communication packet loss and delay of devices, to minimize the revenue loss of VPP device access competition, the objective function can be expressed as:

[0132] This model takes into account the charge-discharge constraints and communication performance constraints of distributed devices during the VPP scheduling process. To reduce the VPP communication scheduling cost and select a suitable access method for distributed devices in different regions, the goal is ultimately to minimize the communication loss O

[0133] To solve the model (40), the optimization model can be decomposed into multiple sub-problems, and then each sub-optimization problem is solved separately. Thus, complex problems are simplified and the solution speed is improved. The algorithm decomposition logic is as Figure 4 shown

[0134] First, the revenue loss O(C l , C -l ) is a function of the device connection matrix C l of the l-th layer network layer and the association matrix C -l of devices in other layers Decompose the problem into multiple ordered sub - problems:

[0135]

[0136] That is, the optimal access of devices in a network layer l is regarded as a sub - problem. When optimizing the devices in the l - th layer, the variables of other layers can be regarded as fixed parameters and solved in sequence.

[0137] After decomposing the target problem, optimize the devices m connected to each network layer l. Considering packet loss and network delay of devices, the AP decides whether to connect the devices within its coverage area. The goal of each AP is as follows:

[0138]

[0139] In the formula, C m is the device connection vector of the m - th AP.

[0140] The decomposition algorithm for the target problem is as Figure 5 shown. First, before optimizing device access, pre - calculate the theoretical expected revenue of each device. Then, use the evaluation results as parameters and pass the parameters to the APs in the current network layer. Each AP solves the decomposed problem separately. The VPP updates the non - ideal communication probability p, where p includes the delay probability and packet loss rate after this iteration, and calculates the revenue reduction O. Continuously iterate until O and p converge.

[0141] After decomposing the problem, it makes the solution of the whole target problem more convenient. However, the sub - problem is still a mixed - integer non - linear programming problem. Adopt a heuristic selection algorithm to optimize and iterate the device access of each layer of the network, as Figure 6 shown in the algorithm flow chart.

[0142] Let n d represent the set of devices connected to the AP. When the current network layer needs to reduce device connections, different from the model that only considers packet loss, combined with the delay parameters of the devices and the selected network layer, select the device n′ in the device set n d with the best performance in terms of delay and packet loss (i.e., the device that causes the least revenue reduction), reduce the connection of one device. The packet loss rate and delay probability of the m - th AP both decrease, that is, the non - ideal communication probability drops from ω m to ω′ m , and the remaining devices connected to the AP (n d -n′) reduce by ω′ m / ω m . Therefore, if the device n′ disconnects, the change in the revenue reduction of the device set n d can be expressed as: can be expressed as:

[0143]

[0144] Similarly, the latency and packet loss performance metrics of the device increase by 1 / ω due to the disconnection. m , and the change in revenue reduction is updated to:

[0145]

[0146] Since ΔO = ΔO + -ΔO - , if ΔO - is greater than ΔO + , then the disconnection of device n′ improves the overall latency and packet loss of the VPP system, and the revenue change O decreases (ΔO < 0). The m-th AP rejects the connection of device n′, and the algorithm repeats until ΔO + exceeds ΔO - .

[0147] When adding a device to a certain network layer, first, select the device n′ with the worst latency and packet loss performance (i.e., the device that causes the maximum reduction in revenue) from the candidate device set n c . After connecting device n′, the packet loss of the m-th AP increases from ω m to ω′ m . The non-ideal communication probability of the device n m originally connected to the m-th AP increases, and the revenue loss O n′ of the selected device n′ decreases. The AP calculates the revenue change of the previously connected device and the revenue change of the selected device (ΔO - = ΔO n′ ). If ΔO - exceeds ΔO + , then accept the connection of device n′. Repeat the iteration until ΔO + exceeds ΔO - .

[0148]

[0149] ΔO- = ΔO n′ = ∑ up,down O n′ × (1 - ω′ m ) (46)

[0150] The original optimization problem is split into sub-problems to be solved sequentially. The device access selection for each network layer is performed through a heuristic selection algorithm. VPP only needs to optimize the device access in a single network layer each time, reducing the complexity of the target problem.

[0151] In this embodiment, in order to explore the impact of latency factors and data packet loss on VPP scheduling during the data transmission between the device and VPP, simulation analysis will be carried out according to the distributed device charging and discharging constraint model and network latency model described above. WiFi can be set according to the IEEE 802.11 standard, with the contention window W0 being 32, and i max and i' max being 6 and 5 respectively, and the packet loss θ m,SNR caused by low signal-to-noise ratio is set to 0.1.

[0152] In the cellular network, t d = 1000 ms, the bandwidth is 5 MHz, b RB-srf = 100, r RB = 168 bit, and r = 256 bit. For the 5G network, r RB = 2016 bit, b RB-srf = 100, r = 256 bit, the bandwidth is 20 MHz, and the 64-QAM (64-Quadrature Amplitude Modulation) method is adopted.

[0153] First, analyze the relationship between the latency probability, packet loss probability, and the number of devices. Figure 7 Shows the fluctuation of the probability p of non-ideal communication (including latency probability and packet loss probability) with the number of access devices in 5G, LTE, and WiFi networks. By comparison, it can be seen that under the same latency and packet loss probabilities, the number of devices accessed by the 5G network is the largest, and the overall network latency fluctuation is also smaller. This is because the 5G network adopts a more flattened design architecture and technologies such as millimeter-wave frequency bands, massive MIMO, and beamforming, which improve network capacity and transmission capabilities and reduce transmission latency and packet loss. The performance of the LTE network is second. LTE uses lower-frequency spectrums. Although it has a wide coverage range, its overall performance is worse than that of 5G. The WiFi network uses unlicensed 2.4 GHz and 5 GHz frequency bands, is more vulnerable to interference from other devices, and uses the CSMA / CA mechanism to manage communication between devices. Devices need to listen to whether the channel is idle before sending data, which increases communication latency. However, WiFi has an obvious cost advantage.

[0154] As Figure 8 shown, they are the network access latencies under different signal-to-noise ratios in 5G, LTE, and WiFi networks respectively. The RB (Resource Block) size of the cellular network is 180 KHz, the number of random access preamble sequences is 54, the random access opportunity interval is 5 ms, the average transmission latency D s is 5 ms, the small-scale fading adopts the Rayleigh model, and the number of devices accessing within the AP coverage range is 100 - 1000. The retransmission limit M of the WiFi network is 6, the PHY (Physical Layer) header length is 192 bits, the header transmission time t p is 1.28 μs, and the device data packet length L is 2 Mbits.

[0155] Comparing the simulation results, with the same number of devices connected, the latency of the 5G network is significantly lower than that of LTE and WiFi; due to spectrum management and access mechanisms, etc., WiFi has the highest latency.

[0156] To compare and analyze the scheduling characteristics of different application scenarios, three scheduling areas, namely residential areas, power generation areas, and energy storage areas, were set up. The number of devices and scheduling parameters are shown in Table 1. Among them, the AP positions are randomly generated uniformly to ensure that there are three different access networks in each area. For each successful scheduling of a device by the VPP, a reward of 0.5 yuan / kw can be obtained from the power supply operator. However, due to the impact of latency cost and packet loss, the response cost of the device is selected from [0.3, 0.6] according to the latency and packet loss situation.

[0157] Table 1 Parameter Settings

[0158]

[0159] Each device may be connected to APs in different network layers. There are a total of 4 access network layers, namely two layers of 5G networks, one layer of LTE, and one layer of WiFi network, two layers of 5G networks. The number of APs in each layer: AP 5G1 = 8; AP 5G2 = 8; AP LTE = 12; AP WiFi = 15.

[0160] Figure 9 Figure is a schematic diagram of the distribution of different devices and APs in the scheduling area. The circles represent APs, and different colors represent APs in different layers. The triangles represent distributed devices.

[0161] The simulation results visualized the device scheduling situation of the VPP by calculating the revenue loss O. When optimizing network access in different areas, if the device can be effectively scheduled by the VPP, corresponding benefits can be obtained. For devices with higher business urgency, networks with better performance will be preferentially matched, while for devices with lower requirements for service quality, networks with lower costs can be selected. By calculating the final revenue loss O, the impact of communication uncertainty factors on the scheduling efficiency of the VPP can be intuitively demonstrated, highlighting the relationship between network performance and scheduling efficiency. As Figure 10 shown, they are the relationships between the total revenue loss O of the VPP and the non-ideal communication probability p after the optimization algorithm is executed in the three areas respectively. Currently, p only considers the data transmission packet loss probability.

[0162] Based on Figure 10 the simulation basis, Figure 11 the impact of latency and packet loss on the scheduling revenue of the VPP was further considered. In Figure 11The non - ideal communication probability p no longer only represents the packet loss rate, but also includes the delay probability. With the iterative optimization of the algorithm for device access, compared with the situation where only packet loss was considered in Figure 10 the result is more convergent and concentrated, the convergence speed is faster, and the objective function O of the VPP revenue loss is also significantly reduced. Figure 11

[0163] In contrast, Figure 10 the iterative results are relatively discrete, showing a higher packet loss rate and a greater revenue loss O. From this comparison, it can be seen that the optimization algorithm considering both delay and packet loss factors can significantly improve the scheduling efficiency and reduce the negative impact of communication uncertainty on revenue. The revenue fluctuation of VPP is reduced, the overall operation is more efficient, and the goal of minimizing revenue loss is finally achieved.

[0164] At the same time, from Figure 10 and Figure 11 it can be seen that in residential areas, due to the relatively small device power q n the revenue fluctuation range of VPP is also relatively small; and there are a large number of access devices in residential areas, indicating that it is more suitable to adopt WiFi and LTE access networks with lower economic costs. While in the power generation area and energy storage area, due to the large amount of power scheduling, it is more suitable to use a heterogeneous communication access network with 5G as the main and LTE and WiFi as the auxiliary.

[0165] Figure 12 The non - ideal communication probability p under two schemes is compared with the change of the number of iterations. Every 50 iterations is used as an index point, and the average values of the optimization results of the two schemes are shown in Table 2. From the results, it can be seen that for the scheme considering the delay factor, after the iteration tends to be stable, the probability of non - ideal communication is relatively low, and the objective function revenue loss O is also significantly reduced. Figure 13 The final device access amounts under the two schemes are shown. It can be clearly seen that Scheme 2 considering both delay and packet loss has significantly improved performance compared to Scheme 1 considering only packet loss, and the device access amount has increased by about 100.

[0166] The above results show that the model and optimization algorithm proposed in this embodiment have excellent performance. Through the optimization and scheduling of distributed device access, not only the device access amount is improved, but also the VPP revenue loss is significantly reduced, effectively suppressing the uncertainty in the scheduling process of the new power system.

[0167] Table 2 Optimization Results

[0168]

[0169] ​In summary, by studying the problem of device access optimization in the VPP heterogeneous access network, in this embodiment, a mixed-integer non-linear programming model is constructed on the basis of considering the charging and discharging constraints of devices and communication uncertainty factors, and the device access process is optimized through a heuristic selection algorithm. The following conclusions are finally obtained: 1) In the device access scenarios of different regions, communication delay and packet loss rate have a significant impact on VPP scheduling. By flexibly selecting the access network type, the model optimizes the allocation of network resources according to the urgency of device services, achieving the goals of service guarantee and cost minimization. 2) As the number of device accesses increases, the limitation of network transmission resources leads to a significant increase in packet loss rate and delay, affecting the overall revenue of VPP. The optimization algorithm proposed in this embodiment reduces the negative impact of communication uncertainty on VPP revenue by matching devices with different priorities with appropriate network performance, making the revenue tend to dynamic balance. 3) Starting from the perspective of heterogeneous communication networks, the optimization model comprehensively considers the dual effects of packet loss and delay on device scheduling, and is closer to the actual situation compared with the traditional model that only considers packet loss.

[0170] Embodiment 2

[0171] This Embodiment 2 provides a non-transitory computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission as described above. The method includes: based on the constraint conditions of the charging and discharging capacity, power output characteristics, access network delay, and network bearing capacity of the virtual power plant VPP devices, combining the data packet loss and device access delay under different network transmission mechanisms, constructing a heterogeneous network communication delay model and a data packet loss model; based on the heterogeneous network communication delay model and the data packet loss model, considering the power output characteristics and charging and discharging limitations of distributed devices with the goal of minimizing revenue loss, quantifying the impact of communication delay and data packet loss rate on VPP revenue, and constructing a device access optimization model; combining hierarchical solution and greedy algorithm to optimize and solve the device access optimization model to obtain a device access scheme.

[0172] Embodiment 3

[0173] Embodiment 3 provides a computer device, including a memory and a processor. The processor and the memory communicate with each other. The memory stores program instructions executable by the processor. The processor calls the program instructions to execute the virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission as described above. The method includes: Based on the constraint conditions of the charge and discharge capacity, power output characteristics, access network delay, and network carrying capacity of the virtual power plant VPP device, combining the data packet loss and device access delay under different network transmission mechanisms, constructing a heterogeneous network communication delay model and a data packet loss model; Based on the heterogeneous network communication delay model and the data packet loss model, considering the power output characteristics and charge and discharge limitations of distributed devices with the goal of minimizing revenue loss, quantifying the impact of communication delay and data packet loss rate on VPP revenue, and constructing a device access optimization model; Combining hierarchical solution and greedy algorithm to optimize and solve the device access optimization model to obtain a device access scheme.

[0174] Embodiment 4

[0175] Embodiment 4 provides an electronic device, including: a processor, a memory, and a computer program; wherein, the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device runs, the processor executes the computer program stored in the memory to enable the electronic device to execute the instructions for implementing the virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission as described above. The method includes: Based on the constraint conditions of the charge and discharge capacity, power output characteristics, access network delay, and network carrying capacity of the virtual power plant VPP device, combining the data packet loss and device access delay under different network transmission mechanisms, constructing a heterogeneous network communication delay model and a data packet loss model; Based on the heterogeneous network communication delay model and the data packet loss model, considering the power output characteristics and charge and discharge limitations of distributed devices with the goal of minimizing revenue loss, quantifying the impact of communication delay and data packet loss rate on VPP revenue, and constructing a device access optimization model; Combining hierarchical solution and greedy algorithm to optimize and solve the device access optimization model to obtain a device access scheme.

[0176] Although the specific implementation manners of the present invention are described above in conjunction with the accompanying drawings, it is not a limitation to the protection scope of the present invention. Those skilled in the art should understand that based on the technical solutions disclosed in the present invention, various modifications or deformations that can be made by those skilled in the art without creative efforts should be covered within the protection scope of the present invention.

Claims

1. A virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission, characterized in that Including: Based on the constraint conditions of the charging and discharging capacity, power output characteristics, access network delay, and network carrying capacity of the virtual power plant (VPP) equipment, combined with the data packet loss and equipment access delay under different network transmission mechanisms, construct a heterogeneous network communication delay model and a data packet loss model; Based on the heterogeneous network communication delay model and the data packet loss model, with the goal of minimizing the revenue loss, consider the power output characteristics and charging and discharging limitations of distributed equipment, quantify the impact of communication delay and data packet loss rate on the VPP revenue, and construct an equipment access optimization model; Combine hierarchical solution and greedy algorithm to optimize and solve the equipment access optimization model to obtain an equipment access plan.

2. The virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission according to claim 1, wherein Construct a heterogeneous network communication delay model, including: According to the radius of the cellular network coverage area and the number of distributed devices with access requirements evenly distributed in the area, calculate the collision probability of the devices initiating random access requests during the i-th random access process, and calculate the average access delay of the devices during the access process based on the collision probability.

3. The virtual power plant device heterogeneous communication network access optimization method considering non-ideal transmission according to claim 2, wherein The coverage range of the cellular network is a radius of R co , and n distributed devices evenly distributed within the area have access requirements. During the i-th random access process, n i The collision probability of the device that initiates a random access request is: where N q is the number of preamble sequences provided for the access node; Then, the average access delay of the device is: Among them, D s is the average transmission delay, and T' is the transmission period.

4. The method for optimizing the access of heterogeneous communication networks of virtual power plant equipment considering non-ideal transmission according to claim 1, characterized in that Construct a heterogeneous network communication data packet loss model, including: Calculate the probability of packet loss and delay generated by the device according to the probability of non-ideal communication occurring during the data interaction process between the device and the VPP; The probability of device packet loss due to collision increases with the increase of the data packet collision probability. Each competing device randomly selects an integer as a counter. When the counter counts down to 0, transmission is triggered; When the retransmitted packet collides again, retransmission is triggered, the counter is randomly selected again, and the counter counts down; Until the number of retransmissions reaches the maximum backoff stage; Obtain a non-linear equation about the probability of the device transmitting data packets, bandwidth, and collision probability by solving the steady-state distribution of the Markov chain.

5. The method for optimizing the access of heterogeneous communication networks of virtual power plant equipment considering non-ideal transmission according to claim 1, wherein, Construct an equipment access model, including: The optimization goal of the VPP scheduling side is to minimize the revenue loss caused by factors such as delay jitter and data packet loss generated by the access of distributed energy equipment; The VPP issues power adjustment instructions to distributed devices according to the requirements of the distribution operator, obtains rewards from the operator, and pays scheduling fees for the devices participating in the response; The difference between the rewards obtained from the distribution operator and the scheduling payment is the net revenue obtained by the VPP; According to the VPP revenue model, considering issues such as reliability and effectiveness during the communication process between the device and the VPP, quantify the two influencing factors of data packet loss and communication delay; The VPP selects the scheduling instruction with the maximum revenue according to different network access delays and the device feedback information; Subtract the revenue of the VPP in the case of non-ideal uplink transmission and non-ideal downlink transmission from the revenue obtained by the VPP during ideal error-free transmission respectively, which is the revenue loss caused by non-ideal communication.

6. The method for optimizing the access of heterogeneous communication networks of virtual power plant equipment considering non-ideal transmission according to claim 5, characterized in that When the uplink transmission is not ideal, VPP cannot receive the device status information in a timely and accurate manner. At this time, VPP makes a scheduling strategy based on the historical state distribution and delay probability distribution of the device n to maximize its expected revenue; During non-ideal downlink transmission, since the device cannot accurately and timely receive the scheduling instruction, if the scheduling instruction is discarded due to data packet loss or excessive delay, the VPP needs to fully bear the revenue loss, and determine the revenue expectation at this time. To measure the revenue risk brought by communication uncertainty to the VPP scheduling operation, considering the power output characteristics of distributed devices and the communication packet loss and delay of devices, minimizing the loss of competitive revenue for VPP device access, the objective function is determined.

7. A virtual power plant device heterogeneous communication network access optimization system considering non-ideal transmission, characterized in that It includes: The first construction module is used to construct a heterogeneous network communication delay model and a data packet loss model based on the constraints of the charging and discharging capacity, power output characteristics, access network delay, and network carrying capacity of virtual power plant (VPP) devices, combined with data packet loss under different network transmission mechanisms and the access delay of devices. The second construction module is used to construct an equipment access optimization model by quantifying the impact of communication delay and data packet loss rate on VPP revenue, considering the power output characteristics and charging and discharging limitations of distributed devices with the goal of minimizing revenue loss, based on the heterogeneous network communication delay model and the data packet loss model. The solution module is used to optimize and solve the equipment access optimization model by combining hierarchical solution and greedy algorithm to obtain an equipment access plan.

8. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the heterogeneous communication network access optimization method for virtual power plant devices considering non-ideal transmission as described in any one of claims 1-6 is implemented.

9. A computer device, characterized in that, It includes a memory and a processor, the processor communicates with the memory, the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the heterogeneous communication network access optimization method for virtual power plant devices considering non-ideal transmission as described in any one of claims 1-6.

10. An electronic device, characterized in that, It includes: A processor, a memory, and a computer program; wherein, the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device runs, the processor executes the computer program stored in the memory so that the electronic device executes instructions to implement the heterogeneous communication network access optimization method for virtual power plant devices considering non-ideal transmission as described in any one of claims 1-6.

Images