Distributed photovoltaic inverter acquisition scheduling method and system

Abstract

The invention discloses a distributed photovoltaic inverter acquisition scheduling method and system, and relates to the technical field of new energy, and the method comprises the steps: constructing distributed photovoltaic power generation real-time monitoring, and carrying out the information acquisition; diagnosing, operating and maintaining equipment faults of the photovoltaic power station based on the knowledge graph; predicting the distributed photovoltaic power to evaluate the operation performance and energy efficiency of the distributed photovoltaic power station; and a distributed photovoltaic power station digital integrated management and control platform is constructed. According to the distributed photovoltaic inverter acquisition and scheduling method provided by the invention, a real-time monitoring system for distributed photovoltaic power generation is constructed, so that the operation safety and stability of a power station are ensured. On the basis of the knowledge graph, equipment faults of the photovoltaic power station are diagnosed, operated and maintained, the diagnosis accuracy and efficiency are improved, the distributed photovoltaic power is predicted, the operation performance and energy efficiency of the distributed photovoltaic power station are evaluated, and the management decision-making capacity of the power station is improved. The method has better effects in the aspects of fault processing efficiency, accuracy of energy efficiency evaluation and operation cost.

Description

Technical Field

[0001] The present invention relates to the field of new energy technology, and in particular to a distributed photovoltaic inverter collection and scheduling method and system. Background Art

[0002] Although the installed capacity and scale of photovoltaic power generation are increasing year by year, the operation and maintenance level and means of most power stations are far lower than the construction speed of power stations, and more and more photovoltaic power stations are facing operation and maintenance problems. In particular, compared with centralized photovoltaic power stations, distributed photovoltaic power stations have poor maintainability. This is mainly because the distributed power stations are too far away and information cannot be accurately transmitted. Therefore, it is impossible to form a highly professional operation and maintenance team on standby at any time like large-scale centralized power stations. At the same time, the current distributed photovoltaic operation and maintenance work is usually undertaken by third-party companies or individuals. The professionalism of the personnel is very unstable, and the operation and maintenance model relies heavily on experience. It often misjudges the failure of the power station, causing personnel to repeatedly travel between the faulty power station and the operation and maintenance site, reducing the operation and maintenance efficiency.

[0003] At present, although photovoltaic power stations have introduced a variety of information management systems to improve their business level, such as photovoltaic monitoring systems, the operation and maintenance of power stations still need to be handled manually, and it is impossible to fully record the operation and maintenance process of distributed photovoltaic power stations. In addition, due to the large number of types and numbers of power generation equipment and the lack of effective means and measures, operation and maintenance personnel can only passively handle faults and cannot achieve timely operation and maintenance. For a large number of repeated equipment operation and maintenance problems, operation and maintenance personnel will waste a lot of time and energy, resulting in low efficiency of operation and maintenance work. In addition, in order to ensure the safe and reliable operation of photovoltaic power generation equipment, operation and maintenance personnel are often required to be on duty 24 hours a day to deal with emergencies. Coupled with the professional level of operation and maintenance personnel and the geographical location of power stations, problems such as high operation and maintenance costs, low efficiency and complicated processes are caused. In general, the current production, operation and maintenance and management problems of photovoltaic power plants are as follows: large site area, many equipment, and complex photovoltaic power generation components; the existing control system cannot monitor production data in real time and effectively count and analyze it; the inspection workload is large, the maintenance and operation efficiency is low, and equipment problems are difficult to find in time. The operation and maintenance of each site is not standardized, and a standardized operation and maintenance management system has not been formed; there is a lack of a complete operation and maintenance management system, and fault troubleshooting and repairs cannot respond quickly; the workload of power station equipment maintenance, inspection and inspection, and operation data statistics is heavy; the sites are numerous and wide-ranging, and personnel are dispersed, and resources cannot be effectively utilized; multiple operating modes coexist, and their effectiveness is not maximized, and the requirements for efficient management cannot be met; the systems are independent of each other, basic data cannot be shared, and information islands are serious; the existing information system data lacks normalization and standardization means and cannot support the needs of data applications.

[0004] The current operation and maintenance level and technology of photovoltaic power stations can no longer meet the urgent needs of photovoltaic power stations, which directly affects the operational stability and power generation of photovoltaic power stations, and the operating profit of power stations cannot be reliably guaranteed. Therefore, it is urgent to research and develop intelligent monitoring and intelligent operation and maintenance systems for distributed photovoltaic power stations to achieve centralized, intensive and intelligent operation and maintenance management of many distributed photovoltaic power stations, which is of great significance to promoting the intelligent development of photovoltaic power stations. Summary of the invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is that the existing photovoltaic power station operation and maintenance method has the problems of low maintenance and repair efficiency, failure to meet the requirements of efficient management, ineffective utilization of resources, and difficulty in timely detection of equipment problems.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: a distributed photovoltaic inverter data collection and scheduling method, including building real-time monitoring of distributed photovoltaic power generation and collecting information; diagnosing and operating photovoltaic power station equipment failures based on knowledge graphs; predicting distributed photovoltaic power to evaluate the operating performance and energy efficiency of distributed photovoltaic power stations; and building a distributed photovoltaic power station digital integrated management and control platform.

[0008] As a preferred solution of the distributed photovoltaic inverter acquisition and scheduling method described in the present invention, wherein: said construction of distributed photovoltaic power generation real-time monitoring and information collection includes comprehensively considering the cost and efficiency of the distributed photovoltaic power generation system, and constructing distributed photovoltaic power generation real-time monitoring;

[0009] Monitor the operating status of the inverter in real time, use sound and light alarms to indicate equipment failures, and check the cause and time of the failure;

[0010] The collected information includes:

[0011] Irradiance, solar irradiance on the system array surface, temperature, PV module temperature, battery temperature, DC current parameters, current, battery voltage and current, inverter input voltage and current, DC load input voltage and current, AC current parameters.

[0012] As a preferred solution of the distributed photovoltaic inverter acquisition and scheduling method described in the present invention, the diagnosis and operation of photovoltaic power station equipment faults based on the knowledge graph includes analyzing the equipment faults of the photovoltaic power station through the knowledge graph technology, performing fault diagnosis at the equipment level and the power station level respectively, and proposing a diagnosis method based on the characteristics of the transformer and the inverter;

[0013] Using fault tree and ontology technology, combined with SWRL rules and Jess reasoning engine, a knowledge graph system is constructed to realize intelligent fault diagnosis.

[0014] As a preferred solution of the distributed photovoltaic inverter acquisition and scheduling method described in the present invention, the diagnosis and operation and maintenance of photovoltaic power station equipment faults based on the knowledge graph includes diagnosing the cognition of operation and maintenance engineers through a Bayesian model, constructing a cognitive skill graph, and optimizing the dispatch strategy;

[0015] Use decision tree algorithm to perform intelligent dispatch based on cognitive skill map data to realize intelligent fault handling;

[0016] Improve the fault expert database and provide fault location and intelligent alarm functions;

[0017] Establish an evaluation method for equipment power generation losses, analyze equipment performance issues and provide improvement suggestions.

[0018] As a preferred solution of the distributed photovoltaic inverter acquisition and scheduling method described in the present invention, wherein: the prediction of distributed photovoltaic power to evaluate the operating performance and energy efficiency of distributed photovoltaic power stations includes proposing a probability model of photovoltaic output power based on orthogonal series density estimation;

[0019] Use particle swarm optimization model, reinforcement learning model and online sequential learning machine to deal with system stability, calculation speed and dynamic data processing issues respectively;

[0020] The correlation coefficient matrix method is used to screen key meteorological factors, and an optical power interval prediction model based on k-means hierarchical clustering is established.

[0021] As a preferred scheme of the distributed photovoltaic inverter acquisition and scheduling method described in the present invention, wherein: the predicted distributed photovoltaic power to evaluate the operating performance and energy efficiency of the distributed photovoltaic power station includes on-site testing of the photovoltaic system links, obtaining efficiency parameters of photovoltaic components, inverters, and system losses, and conducting power generation performance evaluation.

[0022] As a preferred solution of the distributed photovoltaic inverter acquisition and scheduling method described in the present invention, wherein: the construction of a distributed photovoltaic power station digital integrated management and control platform includes displaying the operation status of the power station through three-dimensional visualization technology combined with virtual reality;

[0023] Optimize plant operations with detailed historical analysis of inverters, generators, energy consumption and solar resources;

[0024] Realize real-time monitoring of electrical structure and environmental information and early warning processing of alarm information;

[0025] Based on the distributed database storage architecture, power plant equipment data is stored in real time, and through big data analysis, equipment performance is optimized and operational support is provided.

[0026] Another object of the present invention is to provide a distributed photovoltaic inverter data collection and scheduling system, which can diagnose and operate photovoltaic power station equipment faults based on knowledge graphs, thereby solving the problem of low maintenance efficiency in current photovoltaic power station operation and maintenance methods.

[0027] As a preferred solution of the distributed photovoltaic inverter data collection and scheduling system described in the present invention, it includes: a distributed photovoltaic power generation real-time monitoring and information collection module, a photovoltaic power station equipment fault intelligent diagnosis and operation and maintenance module, a distributed photovoltaic power prediction module; a distributed photovoltaic power station operation performance and energy efficiency comprehensive evaluation module; the distributed photovoltaic power generation real-time monitoring and information collection module is used for real-time monitoring, collection and management of operation data and equipment status of the photovoltaic power generation system; the photovoltaic power station equipment fault intelligent diagnosis and operation and maintenance module is used for intelligent diagnosis and knowledge graph optimization of photovoltaic power station fault diagnosis and maintenance process; the distributed photovoltaic power prediction module is used to predict the future output power of the photovoltaic power station and optimize power scheduling; the distributed photovoltaic power station operation performance and energy efficiency comprehensive evaluation module is used to evaluate the performance and power generation efficiency of key equipment of the photovoltaic power station and optimize equipment maintenance.

[0028] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a distributed photovoltaic inverter acquisition and scheduling method.

[0029] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a distributed photovoltaic inverter acquisition and scheduling method.

[0030] Beneficial effects of the present invention: The distributed photovoltaic inverter acquisition and scheduling method provided by the present invention ensures the safety and stability of power station operation by constructing a real-time monitoring system for distributed photovoltaic power generation, quickly discovers and solves problems, improves the overall operating efficiency of the power station, and reduces power losses caused by equipment failures. Based on the knowledge graph, the faults of photovoltaic power station equipment are diagnosed and maintained, the time for manual troubleshooting is reduced, the accuracy and efficiency of diagnosis are improved, the downtime caused by equipment failure is reduced, and the overall maintenance management of the power station is optimized. The distributed photovoltaic power is predicted to evaluate the operating performance and energy efficiency of the distributed photovoltaic power station, and the management decision-making ability of the power station is improved, so that the power station can adjust the power generation and grid load management according to weather and other uncertain factors, ensuring that the power station can still operate stably and efficiently in a fluctuating environment, and reducing the economic losses caused by power generation uncertainty. The photovoltaic system links are tested on site to evaluate the power generation performance and efficiency. Through continuous performance monitoring and evaluation, it is ensured that the photovoltaic power station equipment always operates at the best efficiency, prolongs the equipment life and improves the overall power generation efficiency of the power station. A digital integrated management and control platform for distributed photovoltaic power stations is constructed, and the management efficiency of the power station is improved through real-time and visual monitoring means. At the same time, the equipment performance is optimized through data analysis, the operating cost is reduced and the power generation income is increased. The present invention achieves better results in terms of fault handling efficiency, accuracy of energy efficiency evaluation and operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0032] Figure 1 The present invention provides an overall flow chart of a distributed photovoltaic inverter data collection and scheduling method according to the first embodiment of the present invention.

[0033] Figure 2 A technical roadmap for a distributed photovoltaic inverter data collection and scheduling method provided for the first embodiment of the present invention.

[0034] Figure 3 This is an overall flow chart of a distributed photovoltaic inverter data collection and scheduling system provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0036] Example 1, reference Figure 1-Figure 2 , as an embodiment of the present invention, provides a distributed photovoltaic inverter collection and scheduling method, comprising:

[0037] S1: Build real-time monitoring of distributed photovoltaic power generation and collect information.

[0038] Furthermore, by comprehensively considering the cost and efficiency of the distributed photovoltaic power generation system, a monitoring system with data collection and processing functions, data wireless transmission function, data display and storage function, historical data query function, and user management function is constructed.

[0039] The monitoring system can complete the real-time operation information and alarm information of solar photovoltaic power generation online monitoring, and can display the current total power generation, daily total power generation, cumulative total power generation and daily power generation curve of photovoltaic power generation. It can monitor the operation status of the inverter in real time, use sound and light alarm to prompt the equipment failure, and check the cause and time of the failure.

[0040] The parameters collected by the acquisition system are as follows:

[0041] Irradiance: total solar irradiance on the horizontal plane G, W / m2; solar irradiance on the system array surface G, W / m2. Temperature: outdoor temperature T, ℃; PV module temperature T, ℃; battery temperature T, ℃. DC current parameters: output voltage U, V; current I, A of the PV array; battery voltage U, V; current I, A; inverter input voltage U, V; current I, A; DC load input voltage U, V; current I, A. AC current parameters: inverter output voltage U, V; current.

[0042] It should be noted that, based on the demand analysis of photovoltaic cloud monitoring servers, the photovoltaic distributed power generation monitoring software platform consists of two parts: cloud monitoring module and local monitoring APP. As the remote monitoring center and data center of the photovoltaic energy storage distributed power station, the cloud monitoring module monitors and intelligently controls the photovoltaic power station through the cloud monitoring software, and stores the power generation status data of the photovoltaic power station through the cloud database. The cloud monitoring server can realize the intelligent management of photovoltaic power generation energy by deploying energy management strategies. As the local monitoring center and data center, the local monitoring APP realizes local monitoring of the photovoltaic power station, and provides a human-machine interface for managers to monitor the photovoltaic power station. Finally, the software-hardware data interaction is realized, data processing is performed, and the computer monitoring system is realized to automatically monitor and control.

[0043] The circuit breakers, disconnectors, grounding switches, transformers, capacitors, AC and DC power supplies and their auxiliary equipment, protection signals and various device status signals of photovoltaic power stations are also included in the monitoring scope of the computer monitoring system. The master station collects and processes the status information, action reports, fault recordings and other related information of power protection through the communication channel.

[0044] S2: Diagnose and maintain photovoltaic power station equipment failures based on knowledge graphs.

[0045] Furthermore, by analyzing the common fault types and influencing factors of the main equipment of photovoltaic power stations, the intelligent diagnosis method of photovoltaic power station equipment faults is studied from the two levels of equipment and power station: at the equipment level, corresponding intelligent fault diagnosis methods are proposed for the operating characteristics of transformers and inverters respectively; at the power station level, a complete intelligent fault diagnosis process is proposed in combination with the fault analysis of transformers, inverters and photovoltaic arrays. A wind power intelligent operation and maintenance system framework with knowledge graph as the core is proposed. The research on intelligent operation and maintenance system includes three main research contents: fault diagnosis decision, cognitive diagnosis, and targeted dispatch. Theoretical analysis is carried out on fault tree, ontology technology, Bayesian reasoning, and decision tree algorithm to lay the foundation for the realization of intelligent operation and maintenance of distributed photovoltaic power stations.

[0046] Based on the fault tree, extracted fault feature parameters and probability importance, the fault ontology model is built using the Protégé tool. The event relationship in the fault tree is represented as the object attribute of the entity, and the probability importance of the event is represented as a data attribute. Then, based on the relationship between the facts built in the model, the SWRL rules and the Jess reasoning engine are integrated to reason about the ontology and discover the implicit facts between the faults. And it can automatically match the rules according to the input fault facts to realize reasoning and fault diagnosis decision-making. Carry out case analysis of abnormal fault diagnosis of transformers, inverters and photovoltaic arrays, verify the operation and maintenance auxiliary decision-making performance of the domain knowledge graph built based on the ontology, and improve the sharing and reuse rate of operation and maintenance knowledge.

[0047] It should be noted that Protégé is used to build a photovoltaic power station fault model. Protégé is a powerful ontology modeling tool that can help us systematically describe the various components in a photovoltaic power station, their relationships, possible fault types, and their associations with operation and maintenance data. This process is divided into the following key steps:

[0048] Step 1: Define the key components of a PV plant

[0049] The equipment and components in the photovoltaic power station are the core of the system. In the ontology model, you need to first clarify these components and their relationships. You can define these devices as classes, such as:

[0050] Photovoltaic Panels

[0051] Inverters

[0052] Battery Storage Systems

[0053] Cables and Connectors

[0054] Monitoring Systems

[0055] The relationships between these classes can be represented by Object Properties. For example, a cable connects the panel and the inverter, the inverter controls the power output, etc.

[0056] The relationship between the devices in the system is as follows:

[0057] Step 2: Define the fault type

[0058] Next, define the failure types associated with each component. Failure types can describe different failure modes, such as:

[0059] Panel Damage

[0060] Inverter Failure

[0061] Battery Aging

[0062] Cable Connection Damage

[0063] Overheating

[0064] These fault types can be defined through classes in Protégé, and detailed attribute descriptions (such as fault cause, impact range, fault detection method, etc.) can be added to each fault.

[0065] Step 3: Add data attributes

[0066] Associate relevant sensor data or monitoring information for each component and fault type. In a photovoltaic power plant, data comes from multiple sensors, such as temperature, voltage, current, irradiance, etc. This data can be associated with components and faults through **DataProperties**. For example:

[0067] Temperature of Panels

[0068] Inverter Output Voltage

[0069] Battery Charge Level

[0070] These properties reflect the status of the component and are updated and evaluated through real-time monitoring data.

[0071] The system supports users to customize the relationships between entities. By constructing these relationships, the knowledge graph of the photovoltaic system can comprehensively and structuredly display all aspects of the photovoltaic system, including equipment, performance, operation and maintenance, environment, etc., providing strong support for the design, operation and maintenance, and management of the photovoltaic system.

[0072] Device association relationship:

[0073] Module-inverter relationship: The connection relationship between PV modules and inverters, indicating how the electrical energy generated by the modules is transmitted to the inverter for conversion.

[0074] Relationship between inverter and grid: The connection between inverter and grid, which explains how inverter converts DC power into AC power and feeds it into the grid.

[0075] Relationship between components: In a large photovoltaic system, there may be series or parallel relationships between components, which affect the voltage and current output of the system.

[0076] Attribute relationship:

[0077] Equipment attributes: such as the power, efficiency, and size of PV modules; the model, rated power, and conversion efficiency of inverters.

[0078] Environmental attributes: such as the geographical location of the PV power station, climate conditions (such as irradiance, temperature), soil type, etc. These attributes affect the performance and operation and maintenance strategy of the PV system.

[0079] Operation and maintenance relationship:

[0080] Maintenance task and equipment relationship: The association between regular maintenance tasks (such as cleaning, inspection, replacement) and specific equipment to ensure that the equipment is in good condition.

[0081] Fault-Solution Relationship: The relationship between equipment fault types and corresponding solutions helps operation and maintenance personnel quickly locate and solve problems.

[0082] Performance relationship:

[0083] System performance and weather relationship: the relationship between the power generation performance of the photovoltaic system and weather conditions (such as irradiance and cloud cover).

[0084] Relationship between system performance and equipment status: the relationship between the overall system performance and the operating status of each device (such as component aging, inverter failure).

[0085] Relationship between space and time:

[0086] Spatial layout relationship: The spatial layout relationship of photovoltaic modules, inverters, energy storage equipment, etc. in photovoltaic power stations affects system efficiency and operation and maintenance convenience.

[0087] Time-varying relationships: changes in system performance over time (such as seasonal changes, daily changes), and the timeline of the equipment aging process.

[0088] Dependency and influence relationship:

[0089] Dependencies between devices: The proper operation of some devices may depend on the status of other devices (e.g. an inverter depends on the power generated by the components).

[0090] Fault impact relationship: The impact that a device failure may have on other devices in the system or on the overall performance.

[0091] Relationship between knowledge and experience:

[0092] Operation and maintenance knowledge and experience: The knowledge and experience accumulated by operation and maintenance personnel, including best practices, common problems and solutions, etc. This knowledge can be integrated into the knowledge graph to provide guidance for system operation and maintenance.

[0093] Step 4: Create inference rules

[0094] In order to achieve automatic fault diagnosis, it is necessary to build inference rules in the system. Protégé can be combined with SWRL (Semantic Web Rule Language) to define rules to trigger fault diagnosis. Common rules can include:

[0095] Overtemperature fault rule: If the panel temperature is above a certain threshold (e.g. 85°C), an “overtemperature fault” may be triggered.

[0096] Voltage Fluctuation Fault Rule: If the inverter output voltage fluctuates beyond the specified range, an “inverter fault” may occur.

[0097] Through these rules, the system can automatically make real-time judgments based on sensor data and prompt operation and maintenance personnel to take corresponding maintenance measures.

[0098] Reasoning based on graph structure: Most of the information in the knowledge graph is fact triples that represent a certain relationship between two entities. From the perspective of the graph, these triples can be viewed as a labeled directed graph, where entities are nodes and relationships are edges. Reasoning methods based on graph structure use graph structure algorithms to reason about knowledge graphs, such as the Path Ranking Algorithm (PRA). PRA predicts a certain specific relationship that may exist between entities by discovering a set of relationship paths connecting two entities. It has the advantages of strong interpretability and the ability to automatically discover reasoning rules.

[0099] Reasoning based on rule learning: This method automatically learns reasoning rules from the knowledge graph through machine learning algorithms. For example, the system can automatically learn the rules for predicting each relationship based on the association rule mining algorithm of the incomplete knowledge base. The reasoning method based on rule learning is accurate and explainable, but rule acquisition is an important task, the search space is large, and the generated rules have low coverage.

[0100] Step 5: Instantiation and verification

[0101] After the ontology model and rules are built, the model needs to be verified with actual data. Actual equipment and faults in the photovoltaic power station can be instantiated, and fault diagnosis tests can be performed by inputting historical data or real-time monitoring data. In this way, it is ensured that the fault model can accurately diagnose problems and provide effective maintenance recommendations.

[0102] Furthermore, the cognitive diagnosis of photovoltaic power station operation and maintenance engineers is carried out based on the Bayesian psychometric model to further realize intelligent dispatch. On the basis of the fault knowledge map, the cognitive diagnosis model of photovoltaic power station operation and maintenance is constructed based on the concepts of cognitive diagnosis model and cognitive attribute. The Bayesian psychometric model and MCMC algorithm are used to carry out cognitive diagnosis of operation and maintenance engineers to form a cognitive skill map. Finally, the decision tree algorithm is used to learn the dispatch decision tree from the data of the operation and maintenance engineer's cognitive skill map to realize dispatch prediction.

[0103] It should be noted that the application of Bayesian statistics in photovoltaic power plants

[0104] Bayesian statistics is a method for dealing with uncertainty that combines prior knowledge with new observation data to predict and update the system. In the management of photovoltaic power plants, Bayesian methods can be applied to the following aspects:

[0105] (1) Fault prediction and diagnosis

[0106] In a photovoltaic power station, equipment may fail or degrade over time. The Bayesian method can combine prior fault data and real-time sensor data to dynamically update the failure probability of the equipment.

[0107] For example:

[0108] Model historical fault data of inverters, photovoltaic panels or battery energy storage systems, and continuously update the health status of the equipment based on Bayesian reasoning.

[0109] Bayesian networks are used to establish causal relationships and fault models for equipment, helping to identify possible root causes of faults.

[0110] (2) Power output prediction

[0111] The power output of a photovoltaic power station is affected by many factors such as weather conditions, temperature, and light intensity. These factors often have large uncertainties. Bayesian methods can optimize power prediction in the following ways:

[0112] The Bayesian regression model is used in combination with historical weather data and photovoltaic power generation data to dynamically adjust the parameters of the prediction model, thereby improving the prediction accuracy of future power generation.

[0113] When new data (such as solar radiation and temperature changes) arrive, the Bayesian model can quickly adjust the prediction results and improve power prediction under abnormal weather conditions.

[0114] Data collection and processing:

[0115] The current, voltage, power and other parameters of the inverter are collected in real time. At the same time, the collected data is preprocessed, including filtering, denoising, normalization and other operations to improve the quality and reliability of the data.

[0116] Bayesian model training and optimization:

[0117] The Bayesian model is trained and optimized using historical failure data and normal operation data.

[0118] By adjusting the parameters and structure of the model, the classification accuracy and generalization ability of the model can be improved.

[0119] Inference algorithm implementation:

[0120] It is necessary to implement the Bayesian reasoning algorithm, including the steps of calculating prior probability, conditional probability and posterior probability. For a complete power generation data prediction process, the neural network needs to include an input layer, a hidden layer and an output layer. Each neuron can be used as the hidden state input of the next time step, or as the input of the next hidden layer or output layer. If it points to the output layer of the neural network, the prediction result can be calculated through the fully connected layer. The specific network structure is shown in the figure.

[0121] User interface:

[0122] Develop a user interactive interface to display the power generation function prediction results and actual power data. By comparison, the interface should be concise, easy to operate and understand, so that users can quickly obtain fault diagnosis results and take corresponding maintenance measures.

[0123] (3) Equipment performance evaluation and life prediction

[0124] Key equipment in photovoltaic power plants (such as inverters and photovoltaic panels) will experience performance degradation during long-term operation. Bayesian statistics can be used to predict the life of equipment and provide more reliable maintenance recommendations. For example:

[0125] The failure time of equipment is estimated through Bayesian Survival Analysis, and updated in combination with real-time operation data to predict the remaining life of the equipment.

[0126] The Bayesian hierarchical model is used to model the differences in the lifespan of different equipment (such as different types of inverters or panels) and optimize the allocation of operation and maintenance resources.

[0127] (4) Uncertainty decision support

[0128] Many decisions in the operation and maintenance of photovoltaic power plants involve uncertainty, such as risks brought about by weather fluctuations and equipment failures. Bayesian methods can provide decision makers with probability-based decision support. For example:

[0129] When choosing whether to replace aging equipment prematurely, Bayesian inference can be used to evaluate the probability of failure and replacement cost to make a more rational decision.

[0130] When faced with issues such as the expansion of a photovoltaic power plant or the installation of a new energy storage system, Bayesian optimization can be used to evaluate the risks and benefits of different options.

[0131] 2. Potential relevance of psychometric models to PV power plants

[0132] Although Bayesian psychometric models are mainly used in the field of psychology to model and analyze individual behaviors and attitudes, their mathematical foundations can be applied across fields, especially in dealing with complex data and uncertainty. In the operation and maintenance of photovoltaic power plants, these psychological methods can be borrowed from other types of measurement and optimization tasks.

[0133] (1) Personnel behavior prediction and management

[0134] The daily operation and maintenance of photovoltaic power plants involves a lot of human resource management, including the work efficiency, task allocation, decision making, etc. of the operation and maintenance personnel. The Bayesian psychometric model can be used to optimize human resource management.

[0135] For example:

[0136] By modeling the task completion and decision-making behavior of the operation and maintenance team, it helps managers optimize workflow and task allocation.

[0137] Through the Bayesian cognitive model, the performance of operation and maintenance personnel under high-stress situations is predicted, thereby reducing the impact of human errors on photovoltaic power plants in task planning.

[0138] (2) Subjective judgment in risk assessment

[0139] In the management decision-making of photovoltaic power plants, it is sometimes necessary to combine subjective judgment with objective data. For example, when judging the maintenance cycle of a certain equipment, the subjective opinions of engineers can be combined with the equipment sensor data. The Bayesian psychometric model can help transform these subjective judgments into quantifiable probability distributions, thereby combining the monitoring data of the equipment to more scientifically assess risks.

[0140] (3) User feedback and satisfaction analysis

[0141] If the electricity from the PV power plant is supplied directly to users or in cooperation with the community, the Bayesian psychometric model can also be used to analyze user satisfaction with energy supply. For example:

[0142] By regularly collecting user feedback and using the Bayesian model to analyze changes in user satisfaction, we can identify problems (such as power outage frequency, reduced energy efficiency, etc.) and optimize power plant operations.

[0143] Combine changes in user satisfaction with system failures, maintenance plans, etc. to identify potential correlations and help improve service quality.

[0144] Furthermore, we will design and develop an intelligent operation and maintenance analysis system for photovoltaic power stations with the following characteristics: by optimizing the fault alarm model of the monitoring and analysis system, we will improve the fault expert database, workflow management, and spare parts management functions; provide hierarchical data transmission based on the alarm model, realize real-time response to alarms and full-cycle tracking of alarms, establish reasonable workflow management, and provide fault location and intelligent alarm functions; establish a complete power station equipment power generation performance evaluation method, accurately evaluate the power generation loss of each key equipment, and realize the comparison of the equipment's own characteristics with various time and space analysis; based on a scalable distributed database storage architecture, massively store the real-time operation data and environmental data of power station equipment, and find ways to improve equipment performance through analysis and mining of these data.

[0145] S3: Predict distributed photovoltaic power to evaluate the operating performance and energy efficiency of distributed photovoltaic power stations.

[0146] Furthermore, in view of the phenomenon that the uncertainty characteristics of distributed photovoltaic power generation systems have not been fully considered in the research, a photovoltaic output power probability model and three light power interval prediction models taking into account randomness are established: one is a particle swarm optimization model aimed at ensuring stability, the second is a reinforcement learning model aimed at speeding up the calculation speed, and the third is an online sequential model aimed at dynamically processing batch data.

[0147] In view of the randomness of the output power of distributed photovoltaic power generation systems, a photovoltaic output power probability model based on orthogonal series density estimation theory is established. This algorithm does not rely on any assumptions about parameter distribution and does not need to establish a physical model of photovoltaic power generation, thus reducing the deviation caused by subjectivity. A hierarchical clustering algorithm based on k-means is proposed to classify photovoltaic power generation operation scenarios according to various meteorological factors, thereby reducing the model error caused by direct classification according to weather types in traditional methods.

[0148] It should be noted that by utilizing the advantages of the extreme learning machine, such as fast calculation speed and strong generalization ability, the prediction interval of distributed photovoltaic output power is directly given, and the output weight of the extreme learning machine model is optimized by the particle swarm algorithm. The entire modeling process avoids the process of inferring and assuming distribution based on point value errors in the traditional interval prediction method. Taking the extreme learning machine as the basic learner, the Adaboost algorithm in reinforcement learning can be used to update the sample weights and learner weights, and an integrated extreme learning machine photovoltaic output power interval prediction model is established.

[0149] The correlation coefficient matrix method is introduced to screen the many meteorological factors that affect photovoltaic output power, retain the factors with high correlation as input variables, and improve the prediction accuracy while simplifying the model. The online sequential learning machine (OS-ELM) can process data in batches and update the output weights in a rolling manner. The photovoltaic online interval prediction model is established to solve the problem that all samples cannot be obtained at one time due to the intermittent and random nature of photovoltaic power generation, and realize the dynamic prediction of distributed photovoltaic output power.

[0150] Furthermore, based on the analysis of the operating performance of the photovoltaic system, on-site tests are conducted on the key links of the photovoltaic system with problems. Through on-site testing of photovoltaic components and photovoltaic arrays, model parameter information such as component efficiency, inconsistency, component attenuation, and photovoltaic array attenuation is obtained; photovoltaic inverter efficiency tests are conducted to obtain photovoltaic inverter model parameters; photovoltaic power station efficiency loss tests are conducted to obtain efficiency loss model parameters, and photovoltaic power station power generation performance evaluation tools are used to evaluate the mid- and long-term power generation performance of the power station.

[0151] S4: Build a digital integrated management and control platform for distributed photovoltaic power stations.

[0152] Furthermore, based on the Internet of Things, cloud computing, big data and other technologies, a digital integrated photovoltaic operation center management and control platform is built from photovoltaic power station equipment to data collectors, to cloud servers, and finally to the client. It combines wireless and wired data transmission, deeply mines and analyzes data, and objectively displays the operating status of equipment in real time, providing strong support for monitoring, operation and maintenance management. The main functions of the management and control platform include:

[0153] Full site visualization

[0154] Panoramic 3D visualization: Combined with advanced virtual reality technology, the 3D engine generates real-time dynamic 3D realistic images, integrating 3D intelligent models with production, operation and business operations, making the power station display more realistic, fault judgment more perspective, and personnel operation more convenient;

[0155] Comprehensive production situation: Display the overall production and operation information of the power station, including power generation plan, power generation, power sales, unit operating capacity, load rate, average monthly utilization hours of power generation equipment and other indicators, and provide trend charts and comparative analysis;

[0156] Plan execution: Real-time dynamic acquisition of power plant power generation data, centralized daily real-time load, yesterday's power generation, this month's cumulative power generation, monthly assessment plan, this year's cumulative power generation and other information, and display the real-time load comparison curve of the entire power plant;

[0157] Real-time operation status: Real-time monitoring of the actual operation status of each unit in the power plant, dynamic grasp of the operation efficiency and other technical and economic indicators of each unit in the plant. If there is any abnormality, automatic real-time warning;

[0158] Equipment start and stop records: Display the power station's operating hours, maintenance hours, and power generation within the cycle. Use Gantt charts to display the unit's standby and operating conditions, implement early warnings for cycle operating hours and maintenance hours, and highlight color markings for abnormal indicators.

[0159] Centralized analysis

[0160] Equipment history analysis: Through the historical data of the inverter, the inverter is comprehensively analyzed from four aspects: loss equivalent time, equipment utilization, inverter equivalent time, and equipment alarm, and the inverter with poor power generation level is located; further through the inverter unit analysis, the hidden defects are determined, preventive maintenance is carried out, and defect orders and contact orders are established to complete the closed-loop management of defect elimination work;

[0161] Power station performance analysis: Analyze the overall efficiency, plan completion rate, abandoned photovoltaic power, loss equivalent time, solar energy resources, and inverter performance level by level, determine the specific factors affecting the power station performance, and deal with them in a timely manner, so as to ensure the overall stable and efficient operation of the power station;

[0162] Energy consumption analysis: Through comprehensive analysis of plant power consumption, grid-purchased power, comprehensive plant power consumption, and energy consumption of power collection lines, we optimize the living power consumption of power station personnel and equipment energy consumption, thereby reducing the operating costs of the power station and improving the efficiency of the power station;

[0163] Solar energy resource analysis: Through comprehensive analysis of measured radiation, feasible radiation, and resource distribution maps, we can understand the true situation of regional energy and provide a reliable basis for investment, design, and equipment selection;

[0164] Inverter energy distribution analysis: supports comparative analysis of power curves and DC currents of all inverters in the station.

[0165] It should be noted that centralized monitoring and integration of data from the photovoltaic power station production management information system can manage the operating indicators of various main and auxiliary equipment in the power station, achieve real-time monitoring of the electrical structure, power station equipment, main operating parameters, etc., and dynamically grasp the real-time status of the power station operation.

[0166] Electrical structure monitoring: By monitoring the electrical structure in the power station, the operation and status of each electrical equipment in the power station, the real-time position of each switch, and the grid-connected status of the power station, the flow of active / reactive power, etc. can be monitored in real time;

[0167] Operation information monitoring: supports real-time monitoring of the photovoltaic power station's power distribution electrical structure, photovoltaic component distribution, inverter and other operation information, as well as the display of key operation data;

[0168] Environmental information monitoring: Automatically collect environmental information of the power station and display it in real time to achieve remote monitoring of the environment. The main information includes: environmental temperature, environmental humidity, dew point temperature, wind speed, wind direction, air pressure, horizontal radiation, inclined radiation, irradiance and other indicators;

[0169] Automatic early warning: Centrally monitor the alarm information of photovoltaic power stations to easily and conveniently understand abnormal operating conditions.

[0170] Embodiment 2, an embodiment of the present invention, provides a distributed photovoltaic inverter collection and scheduling method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0171] First, the experiment was conducted in a small photovoltaic power station. Six photovoltaic devices (devices 1 to 6) were selected for performance comparison. The main parameters monitored and collected included irradiance, photovoltaic module temperature, inverter input voltage and current, and system output power. In addition, the number of faults during equipment operation was also monitored in real time.

[0172] The experiment first built a real-time monitoring and information collection system for photovoltaic power generation. The system can collect irradiance and temperature data through sensors installed on each photovoltaic module, and monitor the real-time changes of voltage and current through the inverter. These data are transmitted to the cloud monitoring module via a wireless network. Managers can view various data in real time through the local monitoring APP and receive system fault alarms. During the experiment, the operating status of each device is recorded around the clock through the monitoring system to ensure the accuracy of the data.

[0173] During the operation of the equipment, the knowledge graph technology is used to automatically diagnose the faults of each device. The system identifies potential problems based on the changes in the equipment's operating parameters and historical failure modes, and issues fault alarms in a timely manner. For equipment that fails, the experiment records the time and cause of the failure in detail, and evaluates the system's intelligent diagnosis function.

[0174] The experiment lasted for one week, during which the system collected and recorded the main parameters of each device and performed real-time analysis. In order to prove the effectiveness of the system, the experimental data will be compared with the performance of the device without the system. The analysis results are shown in Table 1.

[0175]

[0176] Through the analysis of experimental data, we can see the innovative performance of the photovoltaic real-time monitoring and information collection system in monitoring and fault diagnosis. First, from the comparison of system output power, it can be clearly seen that the equipment using the system real-time monitoring (such as equipment 1, equipment 3, and equipment 5) has significantly higher power generation efficiency than other equipment when the photovoltaic irradiance and current conditions are good. For example, equipment 5 has a power generation efficiency of 930W / m 2 Under the conditions of irradiance and 16.2A current, a power output of 26.5kW was achieved, while the power output of device 4 under similar conditions was only 24.5kW, proving that the power generation efficiency of the equipment was improved after the system optimized the energy management strategy.

[0177] Secondly, the system's real-time monitoring and intelligent fault diagnosis functions effectively reduced equipment downtime. During the experiment, equipment 1, equipment 3, and equipment 5 each experienced a fault, but due to the system's timely alarm and intelligent fault diagnosis, managers were able to respond and handle the problem quickly, ensuring that the power generation interruption time was minimized. In contrast, traditional systems may fail to identify these potential faults in a timely manner, resulting in long equipment downtime, which in turn affects the overall power generation efficiency.

[0178] In addition, the system not only collects and transmits basic data, but also conducts in-depth historical data analysis and performance evaluation through cloud big data analysis functions. These analyses help to promptly discover potential problems in equipment operation and provide a basis for decision-making for subsequent equipment maintenance. In this experiment, the system automatically generated a device performance report through comparative analysis of historical data, and pointed out that the performance of the inverter of device 3 degraded rapidly under high temperature conditions. It is recommended to appropriately increase equipment heat dissipation measures in high temperature environments to extend the service life of the equipment.

[0179] In summary, the system of the present invention is obviously innovative. Through real-time monitoring, intelligent fault diagnosis and energy efficiency optimization functions, it significantly improves the operating efficiency of photovoltaic power stations, shortens the equipment fault processing time, and provides scientific data support for managers. Compared with traditional monitoring systems, this system realizes automated and intelligent management of photovoltaic power stations, reflecting higher technical advantages.

[0180] Example 3, reference Figure 3 , which is an embodiment of the present invention, provides a distributed photovoltaic inverter data collection and scheduling system, including a distributed photovoltaic power generation real-time monitoring and information collection module, a photovoltaic power station equipment fault intelligent diagnosis and operation and maintenance module, a distributed photovoltaic power prediction module; and a distributed photovoltaic power station operation performance and energy efficiency comprehensive evaluation module.

[0181] Among them, the distributed photovoltaic power generation real-time monitoring and information collection module is used to monitor, collect and manage the operation data and equipment status of the photovoltaic power generation system in real time. The photovoltaic power station equipment fault intelligent diagnosis and operation and maintenance module is used for intelligent diagnosis and knowledge graph optimization of the fault diagnosis and maintenance process of the photovoltaic power station. The distributed photovoltaic power prediction module is used to predict the future output power of the photovoltaic power station and optimize power dispatching. The distributed photovoltaic power station operation performance and energy efficiency comprehensive evaluation module is used to evaluate the performance and power generation efficiency of key equipment in the photovoltaic power station and optimize equipment maintenance.

[0182] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0183] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0184] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0185] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0186] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A distributed photovoltaic inverter collection and scheduling method, characterized in that: include: Build real-time monitoring of distributed photovoltaic power generation and collect information; Diagnose and maintain photovoltaic power station equipment failures based on knowledge graphs; Predict distributed photovoltaic power to evaluate the operating performance and energy efficiency of distributed photovoltaic power stations; Build a digital integrated management and control platform for distributed photovoltaic power stations.

2. The distributed photovoltaic inverter collection and scheduling method according to claim 1, characterized in that: The construction of distributed photovoltaic power generation real-time monitoring and information collection includes comprehensively considering the cost and efficiency of the distributed photovoltaic power generation system and constructing distributed photovoltaic power generation real-time monitoring; Monitor the operating status of the inverter in real time, use sound and light alarms to indicate equipment failures, and check the cause and time of the failure; The collected information includes: Irradiance, solar irradiance on the system array surface, temperature, PV module temperature, battery temperature, DC current parameters, current, battery voltage and current, inverter input voltage and current, DC load input voltage and current, AC current parameters.

3. The distributed photovoltaic inverter collection and scheduling method according to claim 2, characterized in that: The diagnosis and maintenance of photovoltaic power station equipment faults based on knowledge graphs includes analyzing the equipment faults of photovoltaic power stations through knowledge graph technology, performing fault diagnosis at the equipment level and the power station level, and proposing diagnosis methods based on the characteristics of transformers and inverters; Using fault tree and ontology technology, combined with SWRL rules and Jess reasoning engine, a knowledge graph system is constructed to realize intelligent fault diagnosis.

4. The distributed photovoltaic inverter collection and scheduling method according to claim 3, characterized in that: The diagnosis and operation and maintenance of photovoltaic power station equipment faults based on the knowledge graph includes diagnosing the cognition of operation and maintenance engineers through the Bayesian model, building a cognitive skill graph, and optimizing the dispatch strategy; Use decision tree algorithm to perform intelligent dispatch based on cognitive skill map data to realize intelligent fault handling; Improve the fault expert database and provide fault location and intelligent alarm functions; Establish an evaluation method for equipment power generation losses, analyze equipment performance issues and provide improvement suggestions.

5. The distributed photovoltaic inverter collection and scheduling method according to claim 4, characterized in that: The predicting of distributed photovoltaic power to evaluate the operating performance and energy efficiency of distributed photovoltaic power stations includes proposing a probability model of photovoltaic output power based on orthogonal series density estimation; Use particle swarm optimization model, reinforcement learning model and online sequential learning machine to deal with system stability, calculation speed and dynamic data processing issues respectively; The correlation coefficient matrix method is used to screen key meteorological factors, and an optical power interval prediction model based on k-means hierarchical clustering is established.

6. The distributed photovoltaic inverter collection and scheduling method according to claim 5, characterized in that: The predicted distributed photovoltaic power to evaluate the operating performance and energy efficiency of the distributed photovoltaic power station includes on-site testing of the photovoltaic system links, obtaining efficiency parameters of photovoltaic components, inverters, and system losses, and evaluating power generation performance.

7. The distributed photovoltaic inverter collection and scheduling method according to claim 6, characterized in that: The construction of the distributed photovoltaic power station digital integrated control platform includes displaying the operation status of the power station through three-dimensional visualization technology combined with virtual reality; Optimize plant operations with detailed historical analysis of inverters, generators, energy consumption and solar resources; Realize real-time monitoring of electrical structure and environmental information and early warning processing of alarm information; Based on the distributed database storage architecture, power plant equipment data is stored in real time, and through big data analysis, equipment performance is optimized and operational support is provided.

8. A system using the distributed photovoltaic inverter data collection and scheduling method as claimed in any one of claims 1 to 7, characterized in that: It includes distributed photovoltaic power generation real-time monitoring and information collection module, photovoltaic power station equipment fault intelligent diagnosis and operation and maintenance module, distributed photovoltaic power prediction module; distributed photovoltaic power station operation performance and energy efficiency comprehensive evaluation module; The distributed photovoltaic power generation real-time monitoring and information collection module is used to monitor, collect and manage the operation data and equipment status of the photovoltaic power generation system in real time; The photovoltaic power station equipment fault intelligent diagnosis and operation and maintenance module is used for intelligent diagnosis and knowledge graph optimization of photovoltaic power station fault diagnosis and maintenance process; The distributed photovoltaic power prediction module is used to predict the future output power of the photovoltaic power station and optimize power dispatch; The distributed photovoltaic power station operation performance and energy efficiency comprehensive evaluation module is used to evaluate the performance and power generation efficiency of key equipment in the photovoltaic power station and optimize equipment maintenance.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the distributed photovoltaic inverter data collection and scheduling method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distributed photovoltaic inverter data collection and scheduling method described in any one of claims 1 to 7 are implemented.

Images