A control method, system and computer medium of edge energy router system

Abstract

The present disclosure relates to a control method, system and computer medium of an edge energy router system, wherein the edge energy router system comprises a plurality of nodes, each node comprising an edge energy router integrating both edge computing devices and energy routers and its corresponding computing load, each node having adjacent nodes connected thereto via an energy transmission line and having adjacent nodes connected thereto via an information communication path, and the control method comprising establishing a power balance model for each node and a frequency deviation variation model of the node AC bus. Then a joint solution is performed to determine the control parameters of each node, and the corresponding node is controlled according to the control parameters of each node. The control method of the present disclosure takes into account the influence of various uncertainties, randomness, external interference and delay, and realizes the robust control of the frequency deviation of the AC bus inside each edge energy router.

Description

Technical Field

[0001] The present disclosure relates to the field of edge computing technology, and more specifically, to a control method for an edge energy router system, an edge energy router system, and a non-transient computer medium. Background Art

[0002] In order to cope with energy problems such as global warming and the depletion of fossil energy, the large-scale access of renewable energy to the power grid and the replacement of traditional fossil energy have received great attention from academia, industry and even governments. The Energy Internet adopts a bottom-up decentralized collaborative autonomous management model, with the large power grid as the backbone and several bottom-level autonomous microgrids as local area networks. Among them, each microgrid can access renewable energy power generation equipment, fossil energy power generation equipment and energy storage equipment according to specific needs, and give priority to achieving its own energy supply balance. In order to efficiently and stably achieve the energy supply balance within each microgrid, a large number of control devices, power monitoring equipment, sensors and various electrical equipment are connected to the Energy Internet. Among them, the energy router serves as the core device for information and energy exchange in the Energy Internet.

[0003] To solve the high latency problems of traditional cloud computing, edge computing, as a new computing paradigm, can offload task computing and data storage at the edge of the network close to the user end, thereby avoiding the high latency and unreliability caused by long-distance network transmission. However, as the amount of tasks offloaded to edge devices increases, there are problems with the power supply of edge computing devices, which affects the stability of computing tasks within the microgrid. Summary of the invention

[0004] The invention aims to provide a control method, an edge energy router system and a non-transient computer medium for an edge energy router system, taking into account the influence of various uncertainties, randomness, external interference and delay, to realize the power balance of various application levels (such as but not limited to energy transmission with neighboring nodes via energy transmission lines, task scheduling with neighboring nodes via information communication channels, computing load, etc.) inside the nodes to which each edge energy router belongs, while ensuring the stable and robust control of the AC bus frequency deviation inside the nodes.

[0005] In a first aspect, an embodiment of the present disclosure provides a control method for an edge energy router system, wherein the edge energy router system includes a plurality of nodes, each node including an edge energy router integrating both an edge computing device and an energy router and its corresponding computing load, each node having an adjacent node connected to it via an energy transmission line and having an adjacent node connected to it via an information communication path, the control method comprising, for each node: establishing a first task change model for the computing load of the node, the first task change model being configured to determine the rate of change of the current moment's change of the computing load based on the current moment's change of the task of the computing load, a first input disturbance amount and a first random change rate, while resisting the current moment's change of the task of the computing load; establishing a first power change model for each energy transmission line connected to the node, the first power change model being configured to determine the energy transmission line based on the control information about the energy transmission power change of the energy transmission line at the current moment, while resisting the energy transmission power change of the energy transmission line at a moment before a predetermined time period. The rate of change of the energy transmission power change at the current moment; establishing a second task change model for each information communication path connected to the node, the second task change model being configured to determine the rate of change of the task scheduling change of the information communication path at the current moment based on the control information about the task scheduling change of the information communication path at the current moment, while resisting the task scheduling change of the information communication path at the moment before the predetermined time period; establishing a power balance model for the edge energy router of the node based on the first task change model, the first power change model and the second task change model; establishing a frequency deviation change model of the AC bus for the edge energy router of the node, the frequency deviation change model of the AC bus being configured to determine the rate of change of the frequency deviation at the current moment based on the frequency deviation at the current moment and the power change at the current moment; performing a joint solution based on the first task change model, the first power change model, the second task change model, the power balance model and the frequency deviation change model of each node to determine the control parameters of each node; and the edge energy router of each node performing control according to the control parameters corresponding to the node.

[0006] In a second aspect, an embodiment of the present disclosure further provides an edge energy router system, wherein the edge energy router system comprises a plurality of nodes, each node comprising an edge energy router integrating both an edge computing device and an energy router and its corresponding computing load, each node having an adjacent node connected thereto via an energy transmission line and having an adjacent node connected thereto via an information communication path, the edge energy router system further comprising a controller, the controller being configured to: execute the aforementioned control method for the edge energy router system to determine the control parameters of each node, and transmit them to the edge energy router of each node; the edge energy router of each node is configured to: receive the control parameters of the corresponding node from the controller, and perform control according to the control parameters of the corresponding node.

[0007] In a third aspect, an embodiment of the present disclosure further provides a non-transitory computer medium having computer executable instructions stored thereon, which, when executed by a controller in an edge energy router system, implements the aforementioned control method for an edge energy router system.

[0008] By using a control method for an edge energy router system, an edge energy router system and a computer medium according to various embodiments of the present disclosure, for each node of the edge energy router system, taking into account that it is affected by various uncertainties, randomness, external interference and delays, the power balance of various application levels (for example, but not limited to energy transmission with neighboring nodes via energy transmission lines, task scheduling with neighboring nodes via information communication channels, computing loads, etc.) is integrated within the nodes to which each edge energy router belongs, while ensuring stable and robust control of the AC bus frequency deviation within the node. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In the drawings, which are not necessarily drawn to scale, the same reference numerals may describe similar components in different views. The same reference numerals with letter suffixes or different letter suffixes may represent different instances of similar components. The accompanying drawings generally illustrate various embodiments by way of example and not limitation, and together with the description and claims, serve to illustrate the disclosed embodiments. Such embodiments are illustrative and are not intended to be exhaustive or exclusive embodiments of the present apparatus or method.

[0010] Figure 1 A basic architecture diagram of an edge energy router system according to an embodiment of the present disclosure is shown.

[0011] Figure 2 An example of the structure of each node in the edge energy router system according to an embodiment of the present disclosure is shown.

[0012] Figure 3 A flow chart of a control method according to an embodiment of the present disclosure is shown.

[0013] Figure 4 Another structural example of each node in the edge energy router system according to an embodiment of the present disclosure is shown.

[0014] Figure 5 A flow chart of a control method according to an embodiment of the present disclosure is shown.

[0015] Figure 6 Another structural example of each node in the edge energy router system according to an embodiment of the present disclosure is shown.

[0016] Figure 7 A flow chart of a control method according to an embodiment of the present disclosure is shown.

[0017] Figure 8 Another structural example of each node in the edge energy router system according to an embodiment of the present disclosure is shown.

[0018] Fig. 9 A flow chart of a control method according to an embodiment of the present disclosure is shown.

[0019] Fig.10 A schematic diagram of the joint solution process according to an embodiment of the present disclosure is shown.

[0020] Fig.11 A schematic diagram of an iterative process of a particle swarm algorithm according to an embodiment of the present disclosure is shown.

[0021] Fig.12 A schematic structural diagram of an edge energy router system according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is described in detail below in conjunction with the drawings and specific implementation methods. The embodiments of the present disclosure are further described in detail below in conjunction with the drawings and specific implementation methods, but are not intended to limit the present disclosure.

[0023] The words "first", "second" and similar words used in this disclosure do not indicate any order, quantity or importance, but are only used to distinguish. The words "include" or "comprise" and similar words mean that the elements before the word include the elements listed after the word, and do not exclude the possibility of including other elements.

[0024] The embodiment of the present disclosure provides a control method for an edge energy router system. The edge energy router system may include multiple nodes, such as Figure 1A black dot in the figure represents a node, and each node includes an edge energy router that integrates both edge computing devices and energy routers and its corresponding computing load. Each node has adjacent nodes connected to it via energy transmission lines and adjacent nodes connected to it via information communication paths. For example Figure 1 The physical lines between the nodes can represent the energy transmission line 101, and the dotted lines can represent the information communication path 102 between the nodes. Considering the geographical relationship between adjacent nodes, a pair of adjacent nodes can only transmit energy through the energy transmission line 101, can only communicate through the information communication path 102, can transmit energy through the energy transmission line 101 and communicate through the information communication path 102 at the same time, or can have neither the energy transmission line 101 nor the information communication path 102. However, it should be noted that for any node, it usually has an energy transmission line 101 and / or an information communication path 102 with at least one adjacent node in the edge energy router system, thereby becoming an organic part of the edge energy router system.

[0025] In this example, the edge energy router system can be represented by an undirected graph G(V,ζ), where V represents N different nodes in the system, and ζ is the edge set of the undirected graph, which is used to represent the energy transmission line between the nodes. The edge e(l)=(i,j)=(j,i) indicates that there is energy transmission and exchange between the i-th edge energy router and the j-th edge energy router.

[0026] For each node in the edge energy router system, such as Figure 2 As shown, in this example, the node includes an edge energy router 201 that integrates both edge computing devices and energy routers. The edge energy router 201 can communicate with the terminal device (hereinafter referred to as ED) 202 in the area to receive the computing tasks sent by ED 202. There can be several ED 202, such as ED1, ..., EDn, and the specific number is not limited here. Figure 3As shown, the control method disclosed in the present invention may include: in step S301a, a first task change model is established for the computing load of the node, and the first task change model is configured to determine the rate of change of the current moment's change of the computing load based on the current moment's change of the task of the computing load, the first input disturbance and the first random change rate, while resisting the current moment's change of the task of the computing load. That is, in this example, the first input disturbance and the first random change rate can be designed in the first task change model corresponding to the computing task. In a short period of time, for the node, the rate of change of the change of the computing load caused by external factors is described by the first input disturbance, and the randomness of the change of the task amount of the edge energy router 201 itself is simulated by the first random change rate. In this example, the first task change model established also needs to be used to resist the change of the computing load task at the current moment to achieve system convergence, thereby meeting the robustness requirements of the edge energy router system.

[0027] Further, such as Figure 3 As shown, the edge energy router 201 in the area may have adjacent nodes connected to it via the energy transmission line 101 and adjacent nodes connected to it via the information communication path 102. The number of specific energy transmission lines 101 and information communication paths 102 depends on the number of specific connections, which is not limited here. For example, when the load consumption of the edge energy router 201 is too high, the local energy is not enough to make up for its consumption, then the edge energy router 201 can obtain the abundant electric energy in other connected edge energy routers through the existing energy transmission line 101. In step S301b, a first power change model can be established for each energy transmission line connected to the node, and the first power change model is configured to determine the rate of change of the energy transmission power change of the energy transmission line at the current moment based on the control information of the energy transmission line on the energy transmission power change at the moment before the predetermined time period of the energy transmission line. In this example, it is considered that there is a certain delay between the energy transmission and the control information about the energy transmission power change sent by the edge energy router 201 to the energy transmission line. In order to meet the robustness of the edge energy router system, the first power variation model for each energy transmission line 101 is also used to resist the energy transmission power variation of the energy transmission line at a time before a predetermined time period.

[0028] Similarly, when the amount of tasks in the edge energy router 201 of the node is too large, the local energy is not enough to compensate for its consumption, then the edge energy router 201 of the current node can schedule part of the computing tasks to other nodes through the existing information communication path 102. Therefore, in this example, in step S301c, a second task change model can also be established for each information communication path connected to the node, and the second task change model is configured to determine the change rate of the task scheduling change amount of the information communication path at the current moment based on the control information of the information communication path at the current moment about the task scheduling change amount, while resisting the task scheduling change amount of the information communication path at the moment before the predetermined time period. That is, in this example, it is considered that there is a certain delay between the task scheduling on the information communication path 102 and the control information about the task scheduling change amount sent by the edge energy router 201 to the information communication path 102. In order to meet the robustness of the edge energy router system, the second task change model for each information communication path 102 is also used to resist the task scheduling change amount of the information communication path at the moment before the predetermined time period.

[0029] Then, in step S302, a power balance model can be established for the edge energy router of the node based on the first task change model, the first power change model, and the second task change model. Then, in step S303, a frequency deviation change model of the AC bus can also be established for the edge energy router of the node, and the frequency deviation change model of the AC bus is configured to determine the rate of change of the frequency deviation at the current moment based on the frequency deviation at the current moment and the power change at the current moment. That is, in order to further meet the requirements of the power balance of the node and the robustness of the system, in this example, a power balance model can be established in combination with the first task change model, the first power change model, and the second task change model established above. And according to the established power balance model, the frequency deviation change model of the AC bus inside the edge energy router is correspondingly obtained.

[0030] Next, in step S304, based on the first task change model, the first power change model, the second task change model, the power balance model and the frequency deviation change model, a joint solution (i.e., a joint solution of each node in the system) can be performed to determine the control parameters of each node. For example, the robustness and stability of the AC bus frequency deviation change model can be used as a goal to establish a system differential equation group, and solve the differential equation group to obtain the control parameters of each node.

[0031] In step S305, the edge energy router of each node performs control according to the control parameter corresponding to the node, that is, control can be performed by solving the obtained control parameters of each node.

[0032] By using the control method for an edge energy router system according to an embodiment of the present disclosure, a corresponding model can be established for each node of the edge energy router system, taking into account the combined effects of various uncertainties, randomness, external interference and delays on the power balance and frequency deviation within each node, and control parameters are obtained by solving the model. The edge energy router performs control through the control parameters, thereby effectively solving the control problem of the stability and robustness of the AC frequency deviation within each node.

[0033] In some embodiments, the first task change model includes a first resistance term, a first disturbance term and a first random term. The first resistance term is configured to resist the change in the amount of the task of the computing load at the current moment, and the first disturbance term is determined based on the first input disturbance amount. The first random term simulates the effect of the first random change rate on the change in the amount of the task of the computing load at the current moment. In order to reflect the disturbance caused by external factors and the random disturbance caused by internal factors, the corresponding first disturbance term and the first random term can be set for the first task change model in this example. The first resistance term can be configured to resist the change in the amount of the task of the computing load at the current moment. The first disturbance term can be determined based on the first input disturbance amount. The first random term can be configured according to a specified random process. For example, the Wiener random process w(t) can be used to simulate the randomness of the task volume change.

[0034] For example, use To describe the first input disturbance, the first task change model can be expressed as formula (1) in this example:

[0035]

[0036] in To calculate the rate of change of task variation, To calculate the time constant of the load, To calculate the system factor for the load, It is represented as the change of the computing task at the current moment. As an example of the first resistance term (the symbol "-" realizes resistance), it should be noted that the implementation of the first resistance term is not limited to this. For example, based on formula (1), it can be divided by dt at the same time, so that This can be used as an example of the effect of the first random change rate on the change amount of the task of computing the load at the current moment (using multiplication to achieve the effect), but it should be noted that the implementation method of the effect is not limited to this. The effects of other resistance items and random change rates on the change amount of other parameters in this article are similar to this, and can also be achieved by using similar algorithms, which will not be repeated here.

[0037] In some embodiments, the first power change model includes a second resistance term and an energy control term, wherein the second resistance term is configured to resist the energy transmission power change amount of the energy transmission line at a time before a predetermined time period, and the energy control term is determined based on the control signal of the energy transmission line at the current time about the energy transmission power change amount and the control coefficient of the energy transmission line. Figure 1 As shown, since the energy transmission between nodes depends on the energy transmission line 101 connected to them, the energy transmission has a certain delay compared to the transmission of the control signal. Therefore, the second resistance item in this example is configured to resist the energy transmission power change of the energy transmission line 101 before the predetermined time period. For any energy scheduling control signal, the energy transmission change on the final energy transmission line 101 can be determined based on its corresponding control coefficient. That is, when the load consumption of an edge energy router is too high, the local energy is not enough to make up for its consumption, then the edge energy router can obtain the surplus electric energy in other connected edge energy routers through the existing energy transmission line. In this example, the first power change model of each energy transmission line 101 can be modeled as formula (2):

[0038]

[0039] in is the time constant of the mth energy transmission line, is the control coefficient of the mth energy transmission line, is the control signal for controlling the change in energy transmission on the mth transmission line, and τ(t) is the time delay on the mth energy transmission line.

[0040] In some embodiments, the second task change model includes a third resistance item and a task control item, the third resistance item is configured to resist the task scheduling change amount of the information communication path at a time before a predetermined time period, and the task control item is determined based on the control signal of the information communication path at the current moment about the task scheduling change amount and the control coefficient of the information communication path. Figure 1As shown, since the task scheduling between nodes depends on the information communication path 102 connected to them, compared with the transmission of control signals, task scheduling can also consider a certain delay. Of course, the delay of task scheduling is lower than the delay of energy transmission. Therefore, the delay of task scheduling can also be ignored in the specific modeling process. Considering the delay of task scheduling, the second resistance item in this example is configured to resist the change in energy transmission power at the moment before the predetermined time period of the information communication path 102. For any task scheduling control signal, the task scheduling change on the final information communication path 102 can be determined based on its corresponding control coefficient. That is, when the task volume of the edge node is too large, the local energy is not enough to make up for its consumption, and the edge energy router can schedule part of the computing tasks to other edge energy routers for execution. The second task change model of each information communication path 102 in this example can be modeled as formula (3):

[0041]

[0042] in, is the time constant of the lth information communication path, is the control coefficient of the lth information communication path, is the control signal for controlling the change in the scheduling task on the lth information communication path, and τ(t) is the delay size on the lth information communication path.

[0043] In some embodiments, Figure 4 As shown, the node may also include a traditional power generation device 203 and an energy storage device 204. The traditional power generation device 203 can supply power within the node, and the energy storage device 204 can store energy and supply energy to other devices in the node through the energy storage device in an emergency. Figure 5 As shown, the control method of the present disclosure can also establish a second power change model for the traditional power generation equipment of the node in step S501d, and the second power change model is configured to determine the rate of change of the power generation change of the traditional power generation equipment at the current moment based on the control information of the traditional power generation equipment at the current moment, while resisting the change of the power generation of the traditional power generation equipment at the current moment. The traditional power generation equipment 203 can be controlled by sending a corresponding control signal through the edge energy router. Therefore, the power generation change of the traditional power generation equipment based on the control signal can be described according to the control signal and its corresponding system coefficient. Similarly, in order to ensure the convergence of the system, it is necessary to resist the change of the power generation of the traditional power generation equipment at the current moment.

[0044] Of course, a third power change model can also be established for the energy storage device 204 of the node in step S501e, and the third power change model is configured to resist the frequency deviation at the current moment and the energy storage power change of the energy storage device 204 at the current moment to determine the rate of change of the energy storage power change of the energy storage device 204 at the current moment. In order to ensure the convergence and robustness of the system, the change of the energy storage power of the energy storage device 204 needs to resist the change of the frequency deviation and the energy storage power change of the energy storage device 204 itself at the current moment. Specifically, other modeling and solving processes can refer to the aforementioned process, which will not be repeated here.

[0045] In some embodiments, the second power change model includes a fourth resistance term and a power generation control term, the fourth resistance term is configured to resist the current change in the power generation of the traditional power generation equipment, and the power generation control term is determined based on the control signal of the traditional power generation equipment at the current moment about the power generation change and the control coefficient of the traditional power generation equipment. In this example, it can be determined based on the control signal of the traditional power generation equipment 203 about the power generation change and the control coefficient of the traditional power generation equipment. In this example, the second power change model can be modeled as formula (4):

[0046]

[0047] in is the time constant of traditional power generation equipment, is the control coefficient of traditional power generation equipment.

[0048] In some embodiments, the third power change model of the energy storage device 204 needs to be configured to resist the frequency deviation at the current moment and the energy storage power change amount of the energy storage device 204 at the current moment to determine the change rate of the energy storage power change amount of the energy storage device 204 at the current moment. For example, the third power change model can be modeled as formula (5):

[0049]

[0050] in is the time constant of the energy storage device, Δf k (t) is the AC bus frequency deviation inside node k, is the energy storage power change of node k at the current moment.

[0051] In some embodiments, Figure 6 As shown, the node may also include a renewable energy power generation device 205. The renewable energy power generation device 205 may also be used as a power supply device for the node. The renewable energy power generation device 205 may be a photovoltaic power generation device, a wind turbine generator, or a hydroelectric generator. For such a node, Figure 7As shown, the disclosed control method can also establish a fourth power change model for the renewable energy power generation equipment of the node in step 701f, and the fourth power change model is configured to determine the rate of change of the power generation change of the renewable energy power generation equipment 205 at the current moment based on the change in the power generation of the renewable energy power generation equipment 205 at the current moment, the second input disturbance and the second random change rate, while resisting the change in the power generation of the renewable energy power generation equipment 205 at the current moment. In this example, a photovoltaic power generation equipment is used as an example of the renewable energy power generation equipment 205 for illustration. Considering that the photovoltaic power generation equipment may be subject to external interference, such as a sudden change in light intensity, in order to more realistically describe the photovoltaic power generation equipment, in this example, the second input disturbance is used to describe the rate of change of the photovoltaic power generation power change caused by external factors, and the second random change rate is used to simulate the randomness of the photovoltaic change in the node. In this example, the fourth power change model established also needs to be used to resist the change in the power generation of the photovoltaic power generation equipment itself at the current moment to achieve system convergence, thereby meeting the robustness requirements of the edge energy router system. For other specific modeling and solution processes, please refer to the aforementioned process, which will not be repeated here.

[0052] In some embodiments, the fourth power change model includes a fifth resistance term, a second disturbance term and a second random term, wherein the fifth resistance term is configured to resist the change in the power generation of the renewable energy power generation device at the current moment, the second disturbance term is determined based on the second input disturbance, and the second random term is configured to simulate the effect of the second random change rate on the change in the power generation of the renewable energy power generation device at the current moment. In order to reflect the disturbance caused by external factors and the random disturbance caused by the renewable energy power generation device's own factors, in this example, the corresponding fifth resistance term, second disturbance term and second random term can be set for the fourth power change model. The second random term simulates the effect of the second random change rate on the change in power generation of the renewable energy power generation device at the current moment. The fifth resistance term can be configured to resist the change in the power generation of the renewable energy power generation device at the current moment. The second disturbance term can be determined according to the specified disturbance term of the input. The second random term can be configured according to the specified random process. For example, the Wiener random process w(t) can be used to simulate the randomness of the power generation of the renewable energy power generation device, using To describe the second input disturbance, the fourth power change model in this example can be expressed as formula (6):

[0053]

[0054] in, is the time constant of renewable energy power generation equipment, is the system coefficient of renewable energy power generation equipment, is the disturbance input of renewable energy power generation equipment, It is the change of power generation of photovoltaic power generation equipment at the current moment.

[0055] In some embodiments, Figure 8 As shown, some nodes may also include ordinary loads 206, for example, ordinary loads 206 may be street lamps and other equipment in the area. In the case where the node includes an ordinary load 206, the control method of the present disclosure may also establish a fifth power change model for the ordinary load of the node in step S901g, and the fifth power model is configured to determine the rate of change of the load power change of the ordinary load at the current moment based on the change in load power of the ordinary load at the current moment, the third input disturbance amount and the third random change rate, while resisting the change in load power of the ordinary load at the current moment. In this example, ordinary electrical equipment in the node is taken as an example of ordinary load 206 for illustration. Considering that the load of ordinary electrical equipment may be subject to external interference, such as sudden changes in user electricity consumption, in order to more realistically describe ordinary electrical equipment, in this example, the third input disturbance amount is used to describe the rate of change of the load power of the ordinary load caused by external factors, and the third random change rate is used to simulate the randomness of load changes in the node. In this example, the fifth power change model established also needs to be used to resist the change in the energy consumption of the ordinary load itself at the current moment to achieve system convergence, thereby meeting the robustness requirements of the edge energy router system. Of course, a power balance model can also be established in step S302 based on the first task change model, the first power change model, the fifth power change model, and the second task change model. And in step S303, based on the first task change model, the first power change model, the fifth power change model, the second task change model, the power balance model and the frequency deviation change model, a joint solution is performed to determine the control parameters. Other specific modeling and solution processes can refer to the aforementioned process, which will not be repeated here.

[0056] In some embodiments, the fifth power change model includes a sixth resistance term, a third disturbance term, and a third random term, wherein the sixth resistance term is configured to resist the change in the load power of the ordinary load at the current moment, the third disturbance term is determined based on the third input disturbance, and the third random term is configured to simulate the effect of the third random change rate on the change in the load power of the ordinary load at the current moment. In order to reflect the disturbance caused by external factors and the random disturbance caused by the ordinary load 206 itself, the corresponding sixth resistance term, third disturbance term, and third random term can be set for the fifth power change model in this example. The third random term simulates the effect of the third random change rate on the load power of the ordinary load 206 at the current moment. The sixth resistance term can be configured to resist the change in the load power of the ordinary load 206 at the current moment. The third disturbance term can be determined according to the specified disturbance term input. The third random term can be configured according to a specified random process. For example, the Wiener random process w(t) can be used to simulate the randomness of the power change of the ordinary load 206, using To describe the third input disturbance, the fifth power variation model can be expressed as formula (7) in this example:

[0057]

[0058] in, is the time constant of normal load, is the system factor for normal load, is the disturbance input of normal load, It is the load power change of the common load at the current moment.

[0059] In some embodiments, for the edge energy router of the node, a power balance model is established in step S302 based on the first task change model, the first power change model and the second task change model. Figure 8 As shown, when the nodes include traditional power generation equipment, renewable energy power generation equipment, common loads, computing loads and energy storage equipment, and there is energy scheduling and task scheduling between nodes, the power balance model defined by the following formula (8) is established:

[0060]

[0061] Among them, f(k,l) represents the direction of task scheduling, which takes 1 or -1. -1 means that node k accepts scheduling tasks from other nodes through the lth information communication channel, and vice versa, it means that the task is scheduled to other nodes. Similarly, g(k,m) represents the direction of energy transmission, and m is the mth energy transmission line. L and M respectively represent the number of information communication channels and energy transmission lines in the system. Among them, E represents the ratio between the computing task amount and the energy consumption. Exemplarily, if the node does not contain a certain access element (for example but not limited to renewable energy power generation equipment), the corresponding item is set to 0 to obtain the corresponding power balance model, and so on. Figure 2 , Figure 4 , Figure 6 and Figure 8 Power balance model for nodes with any structure.

[0062] In step S303, a frequency deviation change model of the AC bus of the following formula (9) can also be established for the edge energy router of the node, and the frequency deviation change model of the AC bus is configured to determine the change rate of the frequency deviation at the current moment based on the frequency deviation at the current moment and the power change at the current moment.

[0063]

[0064] Where D k and M k represent the inertia constant and damping coefficient respectively.

[0065] In some embodiments, step S304 joint solution is performed by the following steps:

[0066] S1001 organizes the various models on which the joint solution is based into a total system differential equation.

[0067] For example, the system state vector can be defined for each node as: The control input is The disturbance input vector is

[0068] For the edge energy router of a node, its state vector is defined as n is the total number of nodes, ′ represents transposition, and the control vector is

[0069] For the entire edge energy router system, the state variable of the system is defined as x(t) = [x 1 (t)′,x 2 (t)′,…,x n (t)′,x ER(t)′]′, the control input is u(t)=[u 1 (t)′,u 2 (t)′,…,x n (t)′,u ER (t)′]′, the disturbance input is v(t)=[v 1 (t)′,v 2 (t)′,…,v n (t)′]′.

[0070] The output bus frequency deviation of the system changes to z(t) = [Δf 1 (t),Δf 2 (t),…,Δf n (t)]′.

[0071] By arranging the above modeling equations, we can obtain the following system differential equation (10):

[0072]

[0073] Among them, A, A d , B, C, R, and D are all constant matrices obtained by sorting out formulas (1)-(9), x is the total system state matrix, u is the total control input matrix, and v is the total external disturbance input matrix.

[0074] The control method of the present invention controls the power generation of traditional power generation equipment, energy transmission between edge energy routers, and task scheduling to maintain the internal power balance of each edge energy router node, while ensuring the stability and robustness of the AC bus frequency inside each edge energy router node. In step S1002, the linear inequality matrix can be obtained by the Lyapunov functional method based on the total system differential equation, as shown in formula (11).

[0075]

[0076] Where Ω = AX + XA' + BY + Y'B' + S, S, X, Y are matrix variables in the linear inequality matrix, Υ represents the disturbance factor, I represents the unit matrix, A represents the constant coefficient matrix corresponding to the system state matrix, A d represents the constant coefficient matrix corresponding to energy transmission with time delay and task scheduling, B represents the constant coefficient matrix corresponding to control input, C represents the constant coefficient matrix corresponding to disturbance, D represents the constant coefficient matrix corresponding to the system, R represents the constant coefficient matrix corresponding to the random process, ′ represents the transpose of the matrix, A, A d , B, C, R, and D can be obtained by sorting out formulas (1)-(9), where h represents the derivative of the delay function τ(t).

[0077] Then, in step S1003, it is solved that the symmetric matrix X>0, S>0 and Y satisfies the linear inequality matrix.

[0078] Then, in step S1004, a solution is performed using a preset algorithm. For example, u * =Kx,K=YX -1 , where u * represents the control parameters corresponding to each node, x represents the system state matrix in the overall system differential equation, and K represents the amplification factor matrix.

[0079] In some embodiments, the preset algorithm may also be solved in other ways.

[0080] For example, corresponding constraints can be set for each model of the node.

[0081] For example, since the computing power of each edge energy router is limited, There are upper and lower limits on the value That is, the corresponding constraint condition satisfies formula (12):

[0082]

[0083] Another example is the power change of ordinary loads and the power generation change of renewable energy equipment. and The upper limits are and The lower limit is and That is, the corresponding constraints satisfy formulas (13) and (14):

[0084]

[0085]

[0086] For example, considering the power limit of traditional power generation equipment and preventing excessive control of traditional power generation equipment, the power upper limit of traditional power generation equipment is The power lower limit is Therefore, the constraints of traditional power generation equipment satisfy formula (15):

[0087]

[0088] The charging and discharging constraints of the energy storage battery of the energy storage device satisfy formula (16):

[0089]

[0090] For example, the upper limits of energy and task scheduling transmitted on energy transmission lines and communication lines are and Lower limit and That is, the corresponding constraints satisfy formulas (17) and (18):

[0091]

[0092]

[0093] In some embodiments, solving Y further includes: using the value range of the task variation and / or power variation in each model as a constraint condition, solving Y so that it satisfies the linear inequality matrix and minimizes the infinite norm. In this example, it is further considered that obtaining the control signal only by solving the linear matrix inequality may lead to excessive control of the traditional power generation equipment. Therefore, in this example, the infinite norm ||Y|| ∞ Minimize to solve the problem of over-control, thus we can get ||Y|| ∞ Minimization is an optimization problem of the objective function, that is, formula (19a)-formula (19i):

[0094] minimize||Y|| ∞ Formula (19a)

[0095] Make

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103] In some embodiments, solving Y is performed by a particle swarm optimization algorithm or a reinforcement learning algorithm. Based on the obtained optimization problem formula (19a)-(19i) is a convex optimization problem, the optimization problem can be solved by a particle swarm optimization algorithm (ParticleSwarm Optimization), or the obtained optimization problem, that is, formula (19a)-(19i) can be jointly solved by reinforcement learning to obtain the control parameters of each node of the edge energy router system. By controlling the corresponding nodes through the obtained control parameters of each node, the deviation of each AC bus of the edge energy router system can be ensured to be stable and prevent over-control.

[0104] The calculation steps of the exemplary particle swarm optimization algorithm are as follows: Fig.11 As shown. In step S1101, the particles can be initialized, and then in step S1102, the appropriate values ​​of the particles are calculated. Then in step S1103, the best position gbest experienced by a single particle individual and the best position pbest experienced by the particle population are calculated. Then in step S1104, the speed and position are updated. And in step S1105, it is determined whether the number of iterations reaches the maximum value of the preset number of iterations. If the maximum number of iterations is not reached, the appropriate values ​​of each particle individual are repeatedly calculated. If the maximum number of iterations has been reached, the iterative calculation can be terminated, and the optimized control parameters obtained by the solution are output.

[0105] The edge energy router system considered by the control method of the present disclosure has the self-sufficiency characteristics of a micro-power system, that is, each node can be connected to energy devices such as ordinary loads, traditional power generation equipment, photovoltaic power generation equipment, and energy storage equipment through edge energy routers. In this example, each edge energy router has the ability of edge computing, and task computing is also energy-consuming, so the control method of the present disclosure uses computing load to describe this energy consumption process. In energy devices, ordinary loads, computing loads, photovoltaic power generation equipment, and energy storage equipment are uncontrollable. Considering the reduction of fossil fuels and the protection of the ecological environment, the power generation of traditional power generation equipment that consumes fossil fuels is controlled by the control module in the edge energy router. The controllable process also includes energy transmission and task scheduling between edge energy routers. The control method of the present disclosure can model each node, which can specifically include photovoltaic power generation equipment, traditional power generation equipment, ordinary loads, and computing loads, and model the task scheduling and energy transmission process between edge energy routers, and then obtain the power balance equation inside each node according to the above modeling. Then the robustness and stability of the change value of the AC bus frequency deviation can be used as the goal to jointly solve and determine the control parameters. For each node of the edge energy router system, considering that it is affected by various uncertainties, randomness, external interference and delays, the method disclosed in the present invention can solve the problem of stability and robustness of AC frequency deviation within each node.

[0106] The embodiment of the present disclosure also provides an edge energy router system, such as Fig.12As shown, the edge energy router system includes a plurality of nodes 100, each node includes an edge energy router integrating both edge computing devices and energy routers and its corresponding computing load, each node has a neighboring node connected to it via an energy transmission line 101 and has a neighboring node connected to it via an information communication path 102, and the edge energy router system also includes a controller 1201, and the controller 1201 can be configured to execute the aforementioned control method for the edge energy router system to determine the control parameters of each node 100 and transmit them to the edge energy router of each node. That is, the controller 1201 can communicate with the edge energy router of each node and send the corresponding control parameters to the edge energy router of each node. The edge energy router of each node can be configured to: receive the control parameters of the corresponding node from the controller, and perform control according to the control parameters of the corresponding node. The controller can be set in a node area of ​​a specified edge energy router system, or it can be arranged in the cloud. The specific location can be set according to actual needs, which are not listed here. The controller 1201 can include one or more processors, and the processor can include a processing device of more than one general processing device, such as a microprocessor, a central processing unit (CPU), a graphics processing unit (GPU), etc. More specifically, the processor may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor running other instruction sets, or a processor running a combination of instruction sets. The processor may also be one or more special processing devices, such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), a system on a chip (SoC), etc.

[0107] The embodiment of the present disclosure also provides a non-transient computer medium, on which computer executable instructions are stored, and when the computer executable instructions are executed by the controller in the edge energy router system, the control method for the edge energy router system described above is implemented. In this example, the controller can be arranged in a specified area or in the cloud, and the controller 1201 can include one or more processors, and the processor can include a processing device of more than one general-purpose processing device. The processor can also be communicatively coupled to the non-transient computer medium. The non-transient computer medium can be a memory / storage, such as a read-only memory (ROM), a random access memory (RAM), a phase change random access memory (PRAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), an electrically erasable programmable read-only memory (EEPROM), other types of random access memory (RAM), a flash disk or other forms of flash memory, a cache, a register, a static memory, a compact disk read-only memory (CD-ROM), a digital versatile disk (DVD) or other optical memory, a cassette or other magnetic storage device, or any other possible non-temporary medium used to store information or instructions that can be accessed by a computer device.

[0108] In addition, although exemplary embodiments have been described herein, the scope includes any and all embodiments based on the present disclosure with equivalent elements, modifications, omissions, combinations (e.g., various embodiments intersecting schemes), adaptations or changes. The elements in the claims will be interpreted broadly based on the language adopted in the claims, and are not limited to the examples described in this specification or during the implementation of this application, and the examples will be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered as examples only, and the true scope and spirit are indicated by the following claims and the full scope of their equivalents.

[0109] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more of them) can be used in combination with each other. For example, a person of ordinary skill in the art can use other embodiments when reading the above description. In addition, in the above-mentioned specific embodiments, various features can be grouped together to simplify the present disclosure. This should not be interpreted as an intention that a disclosed feature that is not required to be protected is necessary for any claim. On the contrary, the subject matter of the present disclosure may be less than all the features of a specific disclosed embodiment. Thus, the following claims are incorporated into the specific embodiments as examples or embodiments, wherein each claim is independently used as a separate embodiment, and it is considered that these embodiments can be combined with each other in various combinations or arrangements. The scope of the present invention should be determined with reference to the attached claims and the full scope of equivalent forms granted by these claims.

[0110] The above embodiments are only exemplary embodiments of the present disclosure and are not intended to limit the present invention. The protection scope of the present invention is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present invention within the essence and protection scope of the present disclosure, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present invention.

Claims

1. A control method for an edge energy router system, characterized in that: The edge energy router system includes a plurality of nodes, each node includes an edge energy router integrating both edge computing devices and energy routers and its corresponding computing load, each node has a neighboring node connected to it via an energy transmission line and has a neighboring node connected to it via an information communication path, and the control method includes, For each node: Establishing a first task change model for the computing load of the node, wherein the first task change model is configured to determine a change rate of the change amount of the computing load at the current moment based on the change amount of the task of the computing load at the current moment, a first input disturbance amount, and a first random change rate, while resisting the change amount of the task of the computing load at the current moment; Establishing a first power change model for each energy transmission line connected to the node, wherein the first power change model is configured to determine a change rate of the energy transmission power change of the energy transmission line at the current moment based on control information about the energy transmission power change of the energy transmission line at the current moment and against the energy transmission power change of the energy transmission line at a moment before a predetermined time period; Establishing a second task change model for each information communication path connected to the node, wherein the second task change model is configured to determine a change rate of the task scheduling change amount of the information communication path at the current moment based on control information about the task scheduling change amount of the information communication path at the current moment and against the task scheduling change amount of the information communication path at a moment before a predetermined time period; For the edge energy router of the node, establish a power balance model based on the first task change model, the first power change model and the second task change model; Establishing a frequency deviation change model of the AC bus for the edge energy router of the node, wherein the frequency deviation change model of the AC bus is configured to determine a change rate of the frequency deviation at a current moment based on the frequency deviation at a current moment and the power change at a current moment; Based on the first task change model, the first power change model, the second task change model, the power balance model and the frequency deviation change model of each node, performing a joint solution to determine the control parameters of each node; and The edge energy router of each node performs control according to the control parameters corresponding to the node.

2. The control method according to claim 1, characterized in that: The first task change model includes a first resistance term, a first disturbance term and a first random term, wherein The first resistance item is configured to resist the change amount of the task of the computing load at the current moment, The first disturbance term is determined based on the first input disturbance amount; The first random term simulates the effect of a first random change rate on the change amount of the task of the computing load at a current moment.

3. The control method according to claim 1, wherein the first power change model comprises a second resistance term and an energy control term, wherein the second resistance term is configured to resist the energy transmission power change amount of the energy transmission line at a time before a predetermined time period; The energy control item is determined based on a control signal of the energy transmission line at a current moment regarding a change in energy transmission power and a control coefficient of the energy transmission line.

4. According to the control method according to claim 1, the second task change model includes a third resistance item and a task control item, the third resistance item is configured to resist the task scheduling change amount of the information communication path at a moment before a predetermined time period, and the task control item is determined based on the control signal of the information communication path at the current moment regarding the task scheduling change amount and the control coefficient of the information communication path.

5. The control method according to claim 1, characterized in that: It also includes, when each node includes traditional power generation equipment and energy storage equipment: Establishing a second power change model for the traditional power generation equipment at the node, wherein the second power change model is configured to determine the change rate of the power change amount of the traditional power generation equipment at the current moment based on the control information about the power change amount of the traditional power generation equipment at the current moment and resisting the power change amount of the traditional power generation equipment at the current moment; Establishing a third power change model for the energy storage device of the node, wherein the third power change model is configured to resist the frequency deviation at the current moment and the energy storage power change amount of the energy storage device at the current moment, so as to determine the change rate of the energy storage power change amount of the energy storage device at the current moment; Establishing a power balance model based on the first task change model, the first power change model, and the second task change model further includes: establishing a power balance model based on the first task change model, the first power change model, the second power change model, the third power change model, and the second task change model; Based on the first task change model, the first power change model, the second task change model, the power balance model and the frequency deviation change model of each node, performing a joint solution to determine the control parameters of each node also includes: based on the first task change model, the first power change model, the second power change model, the third power change model, the second task change model, the power balance model and the frequency deviation change model, performing a joint solution to determine the control parameters of each node.

6. The control method according to claim 5, characterized in that: The second power change model includes a fourth resistance item and a power generation control item. The fourth resistance item is configured to resist the change in the power generation of the traditional power generation equipment at the current moment. The power generation control item is determined based on the control signal of the traditional power generation equipment at the current moment about the change in power generation and the control coefficient of the traditional power generation equipment.

7. The control method according to claim 5, characterized in that: It also includes, when each node also includes renewable energy generation equipment: Establishing a fourth power change model for the renewable energy power generation device of the node, wherein the fourth power change model is configured to determine the rate of change of the power generation change of the renewable energy power generation device at the current moment based on the change of the power generation of the renewable energy power generation device at the current moment, the second input disturbance amount, and the second random change rate, while resisting the change of the power generation of the renewable energy power generation device at the current moment; Establishing a power balance model based on the first task change model, the first power change model, and the second task change model further includes: establishing a power balance model based on the first task change model, the first power change model, the second power change model, the third power change model, the fourth power change model, and the second task change model; Based on the first task change model, the first power change model, the second task change model, the power balance model and the frequency deviation change model, performing a joint solution to determine the control parameters of each node also includes: based on the first task change model, the first power change model, the second power change model, the third power change model, the fourth power change model, the second task change model, the power balance model and the frequency deviation change model, performing a joint solution to determine the control parameters of each node.

8. The control method according to claim 7, characterized in that: The fourth power change model includes a fifth resistance term, a second disturbance term and a second random term, wherein the fifth resistance term is configured to resist the change in the power generation of the renewable energy power generation equipment at the current moment, the second disturbance term is determined based on the second input disturbance, and the second random term is configured to simulate the effect of the second random change rate on the change in the power generation of the renewable energy power generation equipment at the current moment.

9. The control method according to claim 5, characterized in that: Also included, when each node also contains normal load: Establishing a fifth power change model for the common load of the node, wherein the fifth power model is configured to determine the rate of change of the load power of the common load at the current moment based on the load power change of the common load at the current moment, the third input disturbance amount, and the third random change rate, while resisting the load power change of the common load at the current moment; Establishing a power balance model based on the first task change model, the first power change model, and the second task change model further includes: establishing a power balance model based on the first task change model, the first power change model, the second power change model, the third power change model, the fifth power change model, and the second task change model; Based on the first task change model, the first power change model, the second task change model, the power balance model and the frequency deviation change model, performing a joint solution to determine the control parameters of each node also includes: based on the first task change model, the first power change model, the second power change model, the third power change model, the fifth power change model, the second task change model, the power balance model and the frequency deviation change model, performing a joint solution to determine the control parameters of each node.

10. The control method according to claim 9, characterized in that: The fifth power change model includes a sixth resistance term, a third disturbance term and a third random term, wherein the sixth resistance term is configured to resist the change in load power of the ordinary load at the current moment, the third disturbance term is determined based on a third input disturbance, and the third random term is configured to simulate the effect of a third random change rate on the change in load power of the ordinary load at the current moment.

11. The control method according to any one of claims 1 to 10, characterized in that: The joint solution is performed by the following steps: Arranging the various models on which the joint solution is based into a total system differential equation; Based on the total system differential equation, the linear inequality matrix is ​​obtained by Lyapunov functional method. Where Ω = AX + XA' + BY + Y'B' + S, S, X, Y are matrix variables in the linear inequality matrix, Υ represents the disturbance factor, I represents the unit matrix, A represents the constant coefficient matrix corresponding to the system state matrix, A d represents the constant coefficient matrix corresponding to energy transmission with time delay and task scheduling, B represents the constant matrix corresponding to control input, C represents the constant coefficient matrix corresponding to disturbance, D represents the constant coefficient matrix corresponding to the system, R represents the constant coefficient matrix corresponding to the random process, ′ represents the transpose of the matrix, and h represents the derivative of the delay function τ(t); Solve the symmetric matrix X>0, S>0 and Y satisfies the linear inequality matrix; Determine * =Kx,K=YX -1 , where u * represents the control parameters corresponding to each node, x represents the system state matrix in the overall system differential equation, and K represents the amplification factor matrix.

12. The control method according to claim 11, characterized in that: Solving for Y further includes: Taking the value range of the task variation and / or power variation in each model as a constraint condition, solve Y so that it satisfies the linear inequality matrix and minimizes the infinite norm.

13. The control method according to claim 12, characterized in that: Solving for Y is performed using a particle swarm optimization algorithm or a reinforcement learning algorithm.

14. An edge energy router system, characterized in that: The edge energy router system includes a plurality of nodes, each node including an edge energy router integrating both edge computing devices and energy routers and its corresponding computing load, each node having adjacent nodes connected thereto via an energy transmission line, and having adjacent nodes connected thereto via an information communication path, The edge energy router system further includes a controller, the controller being configured to: execute the control method for an edge energy router system according to any one of claims 1 to 13 to determine control parameters of each node and transmit the control parameters to the edge energy router of each node; The edge energy router of each node is configured to receive the control parameters of the corresponding node from the controller and perform control according to the control parameters of the corresponding node.

15. A non-transitory computer medium, characterized in that Computer executable instructions are stored thereon, and when the computer executable instructions are executed by a controller in the edge energy router system, the control method for the edge energy router system according to any one of claims 1-13 is implemented.

Images