Adaptive collaborative optimization control method for photovoltaic power plants

Through multi-layer network physical architecture and adaptive collaborative optimization algorithm, the problem of troubleshooting of photovoltaic power stations is solved, the operation and maintenance efficiency and professionalism are improved, and the rapid and accurate troubleshooting and optimization control of photovoltaic power stations are realized.

CN114172264BActive Publication Date: 2024-11-01HUANENG RENEWABLES CORPORATION LIMITED +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111338813.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-11-12
Publication Date
2024-11-01
Estimated Expiration
2041-11-12

AI Technical Summary

Technical Problem

The distributed structure of photovoltaic power stations leads to difficulties in troubleshooting, affecting power generation efficiency, and the operation and maintenance environment is harsh, increasing the operation and maintenance difficulty.

Method used

Adopting a multi-layer network physical architecture, the state data of photovoltaic power stations is collected and processed through the physical layer, the network space layer and the collaborative optimization layer, and the optimization scheme is generated using the adaptive collaborative optimization algorithm, which is converted into physical equipment commands to guide operation and maintenance, achieving fast and accurate Troubleshooting and optimization control.

Benefits of technology

It improves the convenience and efficiency of photovoltaic power station operation and maintenance, simplifies the troubleshooting process, and improves the professionalism and timeliness of distributed photovoltaic power station operation and maintenance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114172264B_ABST
    Figure CN114172264B_ABST
Patent Text Reader

Abstract

The present invention provides an adaptive collaborative optimization control method for a photovoltaic power station, which belongs to the field of photovoltaic power generation technology. Among them, the control method of the present invention specifically includes: constructing a multi-layer network physical architecture of the photovoltaic power station, collecting the status data of the physical equipment of the photovoltaic power station in the corresponding network physical architecture layer, optimizing the abnormal data in the status data using an adaptive collaborative optimization algorithm, and converting it into commands for physical equipment to guide the operation and maintenance of the photovoltaic power station. The present invention designs the photovoltaic power station as a multi-layer network physical architecture, checks the abnormal points of the photovoltaic power station step by step, makes the operation and maintenance of the power station more refined, and the collaborative optimization control scheme proposed by the present invention helps to improve the convenience and efficiency of the operation and maintenance of distributed photovoltaic power stations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power generation, and in particular relates to an adaptive collaborative optimization control method for a photovoltaic power station. Background Art

[0002] The increasing popularity of photovoltaic power generation has increased the risk of power grid fluctuations. Improving the safety and flexibility of photovoltaic power station operation and maintenance has become an important task in the field of photovoltaic power generation. Photovoltaic power stations have the characteristics of large land area, poor operating environment, and high difficulty in operation and maintenance. They are usually constructed in the form of distributed photovoltaic arrays. If a photovoltaic cell component is abnormal, it is necessary to check the fault point step by step from the photovoltaic array, which affects the power generation efficiency of the array.

[0003] Based on the above technical problems, the present invention proposes a new adaptive collaborative optimization control method for a photovoltaic power station. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art and provides a method for adaptive collaborative optimization control of a photovoltaic power station.

[0005] The present invention provides a photovoltaic power station adaptive collaborative optimization control method, which specifically includes: constructing a multi-layer network physical architecture of the photovoltaic power station;

[0006] The status data of the physical equipment of the photovoltaic power station is collected in the corresponding network physical architecture layer, and the abnormal data in the status data is optimized using an adaptive collaborative optimization algorithm and converted into commands for the physical equipment to guide the operation and maintenance of the photovoltaic power station.

[0007] Optionally, the multi-layer network physical architecture includes: a physical layer, a network space layer, and a collaborative optimization layer.

[0008] Optionally, collecting the status data of the physical equipment of the photovoltaic power station in the corresponding network physical architecture layer includes:

[0009] At the physical layer, multi-source, multi-dimensional and real-time status data of the photovoltaic power station physical equipment are transmitted to the centralized control platform.

[0010] Optionally, after collecting the status data of the photovoltaic power station physical equipment in the corresponding network physical architecture layer, the method further includes:

[0011] In the cyberspace layer, digital twin technology is used to map the status data of physical devices to the virtual space to build a virtual model.

[0012] Optionally, the adaptive collaborative optimization algorithm optimizes abnormal data in the state data and converts it into commands for physical devices to guide the operation and maintenance of the photovoltaic power station, including:

[0013] In the collaborative optimization layer, an adaptive collaborative optimization algorithm is used to generate an optimization plan for abnormal data in the status data, and the optimization plan is imported into the virtual device to convert it into a command of the physical device to guide the operation and maintenance of the photovoltaic power station.

[0014] Optionally, the adaptive collaborative optimization algorithm is an adaptive collaborative optimization algorithm based on analysis target cascade, which is designed with three levels of collaborative optimization control, including: node collaborative optimization control, local collaborative optimization control and global collaborative optimization control.

[0015] Optionally, for abnormal data of the photovoltaic power station that cannot be resolved by the local collaborative optimization control, the global collaborative optimization control is performed on the self-organization centered on the bottleneck resource using constraint theory, and the overall multi-objective function is as follows:

[0016] min TW G =∑TW L (1)

[0017]

[0018] min TE G =∑TE L (3);

[0019] Among them, TW G TW is the global waiting time; L is the local waiting time; C max Represents the completion time of the global job; C j is the completion time of the jth task; J is the maximum value of the task sequence number; TE G Represents the total global energy consumption; TE L Represents the total local energy consumption.

[0020] Optionally, for abnormal data of the photovoltaic power station that cannot be solved by the node collaborative optimization control, the local collaborative optimization control is used to actively form a temporary self-organization between the interactive links to perform control, wherein for the continuous operations (i, j) and (h, j), the variables at the local level are expressed as the following relationship:

[0021]

[0022]

[0023]

[0024] Where M is the number of devices; y 0j is the initial time of job j; P ij is the processing time of operation (i, j); Wihj represents the waiting time between operations (i, j) and (h, j); E ij is the energy consumption of operation (i, j); E ihj Represents the energy consumption between operations (i, j) and (h, j).

[0025] Optionally, for abnormal data of the photovoltaic power station at the vertex, the decision-making knowledge of the node collaborative optimization control itself is used to adjust the control parameters, wherein the variables at the node level can be expressed as the following relationship:

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036] Among them, the subscripts n, m, and l of the photovoltaic array represent the option sequence numbers of the corresponding analysis target cascade elements; is the waiting time after executing operation (i, j); is the transmission time from the node to the load end; is the transmission time from the node to the non-load end; is the waiting time before executing operation (h, j); MS represents the maximum value of photovoltaic manufacturing service; ms represents the photovoltaic manufacturing service option; when the equipment is available, a=1, otherwise it is 0; when the service is selected, b=1, otherwise it is 0; e is the installed power.

[0037] Optionally, the operation and maintenance deviation of the photovoltaic power station is defined as the following relationship:

[0038]

[0039] Among them, ω W is the waiting time weight coefficient; ω C is the weight coefficient of the job completion time; ωE is the energy consumption weight coefficient; P max The maximum time required to optimize the control process; max is the maximum energy consumption; wt max is the maximum waiting time; the target cascade q element is a variable Coupling; is a finite set of sub-elements of q; and represents a small positive threshold specified by the user, and, at each analysis target cascade element, self-organization is formed by minimizing the deviation between the target variable (t) and the response variable (r) to eliminate abnormal data and reach a new balance.

[0040] The present invention provides an adaptive collaborative optimization control method for a photovoltaic power station, which specifically includes: constructing a multi-layer network physical architecture of the photovoltaic power station, collecting the status data of the physical equipment of the photovoltaic power station in the corresponding network physical architecture layer, optimizing the abnormal data in the status data using an adaptive collaborative optimization algorithm, and converting it into commands for physical equipment to guide the operation and maintenance of the photovoltaic power station. The present invention designs the photovoltaic power station as a multi-layer network physical topology architecture, which helps to quickly and accurately troubleshoot fault points by refining each level of the power station. In addition, the collaborative optimization control scheme proposed by the present invention helps to improve the convenience and efficiency of the operation and maintenance of distributed photovoltaic power stations, and has high applicability to photovoltaic power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flowchart of a photovoltaic power station adaptive collaborative optimization control method according to an embodiment of the present invention;

[0042] Figure 2 A schematic diagram of a multi-layer network physical architecture of a photovoltaic power station according to another embodiment of the present invention;

[0043] Figure 3 It is a schematic diagram of a photovoltaic power station topology structure based on analysis target cascade according to another embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0045] like Figure 1 As shown, the present invention provides a photovoltaic power station adaptive collaborative optimization control method S100, which specifically includes the following steps S110-S120:

[0046] S110. Construct a multi-layer network physical architecture for a photovoltaic power station.

[0047] Specifically, since photovoltaic power stations have the characteristics of large land area and poor working environment, in order to facilitate the on-site operation and maintenance of photovoltaic power stations by staff, this embodiment designs a three-layer network physical architecture of photovoltaic power generation system, such as Figure 2 As shown in Figure 1, the network physical architecture includes: physical layer, cyberspace layer and collaborative optimization layer. The high integration of the physical layer and the cyberspace layer provides an implementation basis for the collaborative optimization layer mechanism to solve the problem of building a multi-layer network physical structure for photovoltaic power plants.

[0048] Further, such as Figure 2 As shown in the figure, in the physical layer, the physical manufacturing infrastructure resources of the photovoltaic power generation system include photovoltaic arrays, sensors, junction boxes and other equipment. Each photovoltaic array has a specific association with various sensors, and finally the generated electricity is gathered into the junction box. Therefore, in the physical layer, the multi-source, multi-dimensional and real-time status data of the physical equipment of the photovoltaic power station are transmitted to the centralized control platform. In the cyberspace layer, the data of the physical equipment is mapped to the virtual space to build a virtual model based on the digital twin technology of the synchronized real-time state. In other words, this layer includes visualization equipment, virtual models, etc. In the collaborative optimization layer, an optimization plan is generated for the abnormal data in the status data, and the optimization plan is imported into the virtual device to convert it into a command for the physical device to guide the operation and maintenance of the photovoltaic power station. That is, this layer corresponds to an expert system, as well as information extraction, optimization plan and operation and maintenance status data.

[0049] It should be noted that this embodiment uses the collected real-time status data and knowledge base to detect external and internal anomalies, where the knowledge base directly provides the optimal solution for common abnormal data. New abnormal data is first analyzed to extract knowledge and update the knowledge base, and an optimal solution is generated and imported into the virtual device, which is then converted into commands for physical devices to guide the operation and maintenance of the photovoltaic power station.

[0050] It should be further explained that the digital twin device and model simulation software of this embodiment can be connected to the digital platform in a plug-and-play manner.

[0051] This embodiment designs the photovoltaic power station as a multi-layer network physical topology architecture, which helps to quickly and accurately troubleshoot fault points by refining each level of the power station.

[0052] S120, collecting status data of the physical equipment of the photovoltaic power station in the corresponding network physical architecture layer, optimizing abnormal data in the status data using an adaptive collaborative optimization algorithm, and converting the data into commands for the physical equipment to guide the operation and maintenance of the photovoltaic power station.

[0053] Specifically, the adaptive collaborative optimization algorithm is an adaptive collaborative optimization algorithm based on analytical target cascade. The analytical target cascade is a model-based, hierarchical, multi-level method that is often used in the design and optimization of large-scale systems. Therefore, it can be used to solve collaborative optimization problems. The photovoltaic power generation system is decomposed into three layers: global level, local level and node level. Its topological structure is as follows: Figure 3 The overall optimization goal is propagated from the top-level global-level elements to the lower-level local-level and node-level elements.

[0054] Furthermore, in order to handle abnormal data of different degrees, this embodiment considers key performance indicators (KPIs), which include waiting time, maximum scheduling time and energy consumption. It is designed with three levels of collaborative optimization control, including: node collaborative optimization control, local collaborative optimization control and global collaborative optimization control.

[0055] Specifically, for abnormal data of photovoltaic power plants that cannot be solved by local collaborative optimization control, the constraint theory is used to perform the global collaborative optimization control on the self-organization centered on the bottleneck resource, that is, the entire photovoltaic power plant resources are used to solve the abnormality. The overall multi-objective function can be constructed as follows:

[0056] min TW G =∑TW L (1)

[0057]

[0058] min TE G =∑TE L (3);

[0059] Among them, TW G TW is the global waiting time; L is the local waiting time; C max Represents the completion time of the global job; C j is the completion time of the jth task; J is the maximum value of the task sequence number; TE G Represents the total global energy consumption; TE L Represents the total local energy consumption.

[0060] Furthermore, for abnormal data of photovoltaic power plants that cannot be solved by node collaborative optimization control, local collaborative optimization control is used to actively form temporary self-organization between interactive links to control, without affecting the overall operation of the photovoltaic power plant. For continuous operations (i, j) and (h, j), the variables at the local level are expressed as the following relationship:

[0061]

[0062]

[0063]

[0064] Where M is the number of devices; y 0j is the initial time of job j; P ij is the processing time of operation (i, j); W ihj represents the waiting time between operations (i, j) and (h, j); E ij is the energy consumption of operation (i, j); E ihj Represents the energy consumption between operations (i, j) and (h, j).

[0065] Furthermore, for the small abnormal data of the PV power station at the vertex, the decision-making knowledge of the node collaborative optimization control itself is used to adjust the control parameters. At the node level, the variables can be expressed as the following relationship:

[0066]

[0067]

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076] Among them, the subscripts n, m, and l of the photovoltaic array represent the option sequence numbers of the corresponding analysis target cascade elements; is the waiting time after executing operation (i, j); is the transmission time from the node to the load end; is the transmission time from the node to the non-load end; is the waiting time before executing operation (h, j); MS represents the maximum value of photovoltaic manufacturing service; ms represents the photovoltaic manufacturing service option; when the equipment is available, a=1, otherwise it is 0; when the service is selected, b=1, otherwise it is 0; e is the installed power.

[0077] Furthermore, the operation and maintenance deviation of the photovoltaic power station is defined as the following relationship:

[0078]

[0079] Among them, ω W is the waiting time weight coefficient; ω C is the weight coefficient of the job completion time; ω E is the energy consumption weight coefficient; P max The maximum time required to optimize the control process; max is the maximum energy consumption; wt max is the maximum waiting time; the target cascade q element is a variable Coupling; is a finite set of sub-elements of q; and represents a small positive threshold specified by the user, and, at each analysis target cascade element, self-organization is formed by minimizing the deviation between the target variable (t) and the response variable (r) to eliminate abnormal data and reach a new balance.

[0080] The collaborative optimization control strategy of this embodiment has the ability to coordinate the overall situation and precise optimization performance, and is highly applicable to photovoltaic power stations. In addition, this embodiment is based on a multi-layer network physical architecture and an adaptive collaborative optimization Y-system solution, which solves the problem of troubleshooting abnormal points in the operation and maintenance of photovoltaic power stations, as well as the collaborative optimization control problem of photovoltaic power station scheduling.

[0081] The following will further illustrate the photovoltaic power station adaptive collaborative optimization control method with reference to specific embodiments:

[0082] This embodiment uses a 20MW photovoltaic power station as an example for analysis, and includes the following steps:

[0083] 1. Build a multi-layer network physical architecture for photovoltaic power stations

[0084] In order to facilitate the operation and maintenance of photovoltaic power stations on site, the following Figure 2 The three-layer network physical architecture of the photovoltaic power generation system shown in Figure 1 includes the physical layer, the cyberspace layer and the collaborative optimization layer. The high integration of the physical layer and the cyberspace layer provides the implementation basis for the collaborative optimization layer mechanism.

[0085] The physical layer of the photovoltaic power generation system includes photovoltaic arrays, sensors, junction boxes and other equipment. Each photovoltaic array has a specific association with various sensors, and finally the generated electricity is gathered into the junction box. Therefore, the multi-source, multi-dimensional and real-time status data of the physical equipment are sensed and transmitted to the centralized control platform. In the cyberspace layer, the digital twin technology based on synchronized real-time status maps the data of the physical equipment to the virtual space to build a virtual model. The digital twin device and the model simulation software can be connected to the digital platform in a plug-and-play manner. In the collaborative optimization layer, the collected real-time status data and knowledge base are used to detect external and internal anomalies. For common anomalies, the knowledge base directly provides the optimal solution. New anomalies are first analyzed to extract knowledge and update the knowledge base. Therefore, an optimal solution is generated and imported into the virtual device, and then converted into commands for the physical device to guide the operation and maintenance of the photovoltaic power station.

[0086] 2. Adaptive collaborative optimization algorithm based on photovoltaic power station

[0087] 1) Analyze the target cascade

[0088] The photovoltaic power generation system is decomposed into Figure 3 The topology shown contains global, local and node levels. The overall optimization objective of the PV plant is propagated from the top global level elements to the lower level local and node level elements.

[0089] 2) Adaptive collaborative optimization algorithm

[0090] In order to handle different degrees of abnormalities in photovoltaic power generation systems, three levels of collaborative optimization control are designed, including node collaborative optimization control, local collaborative optimization control, and global collaborative optimization control, considering KPIs (including waiting time, maximum scheduling time, and energy consumption). For photovoltaic power station abnormalities that cannot be solved by local collaborative optimization control, constraint theory is used to perform global collaborative optimization control on the self-organization centered on the bottleneck resource, and the resources of the entire photovoltaic power station are used to solve the abnormalities. The overall multi-objective function can be constructed as:

[0091] min TW G =∑TW L (18)

[0092]

[0093] min TE G =∑TE L (20).

[0094] For the abnormality of the photovoltaic power generation system that cannot be solved by the node, the local collaborative optimization control is carried out by actively forming a temporary self-organization between the interactive links, which does not affect the overall operation of the photovoltaic power station. For continuous operations (i, j) and (h, j), the variables at the local level can be expressed as:

[0095]

[0096]

[0097]

[0098] For small anomalies of the photovoltaic power generation system at the vertex, the node collaborative optimization control's own decision-making knowledge is used to adjust the control parameters and express the node-level variables as:

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107]

[0108]

[0109] The subscripts n, m, and l of the photovoltaic array represent the option sequence numbers of the corresponding analysis target cascade elements.

[0110] At each analysis target cascade element, the operation and maintenance deviation of the PV power station is calculated according to the following relationship (34):

[0111]

[0112] By minimizing the deviation between the target variable (t) and the response variable (r), self-organization is formed to eliminate the abnormality of the photovoltaic power generation system, thereby achieving the purpose of optimizing the scheduling of photovoltaic power stations.

[0113] The present invention provides a photovoltaic power station adaptive collaborative optimization control method, which has the following beneficial effects compared with the prior art:

[0114] First, the present invention designs the photovoltaic power station as a multi-layer network physical architecture, and checks the abnormal points of the photovoltaic power station step by step, making the operation and maintenance of the power station more refined.

[0115] Second, the collaborative optimization control scheme proposed in the present invention helps to improve the convenience and efficiency of distributed photovoltaic power station operation and maintenance.

[0116] Third, the present invention adopts an analytical target cascade algorithm to solve the problem of collaborative optimization control of photovoltaic power plants, thereby improving the timeliness and professionalism of photovoltaic power plant scheduling.

[0117] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A photovoltaic power station adaptive collaborative optimization control method, characterized in that: Specifically include: Construct a multi-layer network physical architecture for photovoltaic power plants; The state data of the physical equipment of the photovoltaic power station is collected in the corresponding network physical architecture layer, and the abnormal data in the state data is optimized by using an adaptive collaborative optimization algorithm, and converted into commands for physical equipment to guide the operation and maintenance of the photovoltaic power station; wherein, The adaptive collaborative optimization algorithm is an adaptive collaborative optimization algorithm based on analytical target cascading, which is designed with three levels of collaborative optimization control, including: node collaborative optimization control, local collaborative optimization control and global collaborative optimization control; For abnormal data of photovoltaic power plants that cannot be solved by the local collaborative optimization control, the global collaborative optimization control is performed on the self-organization centered on the bottleneck resource using constraint theory, and the overall multi-objective function is as follows: minTW G =∑TW L (1) mind G =∑TE L (3); Among them, TW G TW is the global waiting time; L is the local waiting time; C max Represents the completion time of the global job; C j is the completion time of the jth task; J is the maximum value of the task sequence number; TE G Represents the total global energy consumption; TE L Represents the total local energy consumption; For abnormal data of the photovoltaic power station that cannot be solved by the node collaborative optimization control, the local collaborative optimization control is used to actively form a temporary self-organization between the interactive links to perform control, wherein for the continuous operations (i, j) and (h, j), the variables at the local level are expressed as the following relationship: Where M is the number of devices; y 0j is the initial time of job j; P ij is the processing time of operation (i, j); W ihj represents the waiting time between operations (i,j) and (h,j); E ij is the energy consumption of operation (i, j); E ihj Represents the energy consumption between operations (i,j) and (h,j).

2. The method according to claim 1, characterized in that The multi-layer network physical architecture includes: a physical layer, a network space layer and a collaborative optimization layer.

3. The method according to claim 2, characterized in that The collecting of the status data of the physical equipment of the photovoltaic power station in the corresponding network physical architecture layer includes: At the physical layer, multi-source, multi-dimensional and real-time status data of the photovoltaic power station physical equipment are transmitted to the centralized control platform.

4. The method according to claim 2, characterized in that: After collecting the status data of the photovoltaic power station physical equipment in the corresponding network physical architecture layer, the method further includes: In the cyberspace layer, digital twin technology is used to map the status data of physical devices to the virtual space to build a virtual model.

5. The method according to claim 2, characterized in that: The adaptive collaborative optimization algorithm optimizes abnormal data in the state data and converts it into commands for physical devices to guide the operation and maintenance of the photovoltaic power station, including: In the collaborative optimization layer, an adaptive collaborative optimization algorithm is used to generate an optimization plan for abnormal data in the status data, and the optimization plan is imported into the virtual device to convert it into a command of the physical device to guide the operation and maintenance of the photovoltaic power station.

6. The method according to claim 1, characterized in that For abnormal data of the PV power plant at the vertex, the decision-making knowledge of the node collaborative optimization control itself is used to adjust the control parameters, wherein the variables at the node level can be expressed as the following relation: Among them, the subscripts n, m, and l of the photovoltaic array represent the option sequence numbers of the corresponding analysis target cascade elements; is the waiting time after executing operation (i, j); is the transmission time from the node to the load end; is the transmission time from the node to the non-load end; is the waiting time before executing operation (h, j); MS represents the maximum value of photovoltaic manufacturing service; ms represents the photovoltaic manufacturing service option; when the equipment is available, a=1, otherwise it is 0; when the service is selected, b=1, otherwise it is 0; e is the installed power.

7. The method according to any one of claims 1 to 6, characterized in that: The operation and maintenance deviation of the photovoltaic power station is defined as the following relationship: Among them, ω W is the waiting time weight coefficient; ω C is the weight coefficient of the job completion time; ω E is the energy consumption weight coefficient; P max The maximum time required to optimize the control process; max is the maximum energy consumption; wt max is the maximum waiting time; the target cascade q element is a variable Coupling; is a finite set of sub-elements of q; and represents a small positive threshold specified by the user, and, at each analysis target cascade element, self-organization is formed by minimizing the deviation between the target variable (t) and the response variable (r) to eliminate abnormal data and reach a new balance.