Distributed power generation system fault detection and fault tolerance method and system

Abstract

The invention discloses a distributed power generation system fault detection and fault tolerance method and system, and the method comprises the steps: firstly obtaining the position information of each device in the system and the weather information of the position where the device is located, carrying out the clustering analysis, obtaining a device cluster, determining each device as a target power generation device, obtaining the actual value of the power generation power, and carrying out the clustering analysis. Obtaining the real value of the target device in each device of each device cluster, determining an estimated value based on the real value of the power generation power of each device, comparing the estimated value with the real value to determine whether a fault exists, and if the fault exists, selecting a mobile energy storage device to supply power; according to the invention, through clustering analysis based on position and weather and in combination with the difference between the true value and the estimated value of the generated power, whether the equipment has a fault can be accurately identified, the fault equipment can be timely and effectively positioned, fault-tolerant reserve is provided for the power generation system through the mobile energy storage equipment, and the fault-tolerant performance of the power generation system is improved. And the accuracy and the fault-tolerant rate of the power generation system are comprehensively improved.

Description

Technical Field

[0001] The present invention relates to fault detection and fault tolerance means, belonging to the field of electric power technology, and in particular to a distributed power generation system fault detection and fault tolerance method and system. Background Art

[0002] The current power supply system mainly relies on a centralized power supply system, which is characterized by the use of large-capacity generators, extensive grid coverage and high-voltage transmission. This system supports the power demand in most parts of the world, and therefore places extremely high demands on the quality of energy and the stability of power supply. However, the architecture of the centralized power grid also brings potential vulnerabilities: any local failure in the power grid, such as equipment damage or overload, may cause a chain reaction, affecting the stable operation of the entire power grid; in extreme cases, such a chain failure may cause large-scale power outages, or even paralyze the entire power grid, causing serious social and economic losses.

[0003] Distributed power generation technology has become an indispensable and important supplement to centralized power supply technology due to its advantages such as high reliability, high quality, high efficiency and flexibility, and has been widely used as an important development direction in the energy field. According to the different primary energy sources used, distributed power generation can be divided into distributed power generation technology based on fossil energy, distributed power generation technology based on renewable energy and hybrid distributed power generation technology. Although distributed power generation technology effectively solves the defects of a single power supply system, it is mainly connected to the grid through photovoltaic inverters. Such direct grid-connected operation is prone to many problems, such as harmonics, three-phase imbalance, reactive power and voltage, islanding problems, etc. Therefore, how to timely and effectively realize fault detection of distributed power generation systems and provide fault tolerance to improve the accuracy and fault tolerance of distributed power generation systems has become an urgent problem to be solved. Summary of the invention

[0004] The purpose of the present invention is to overcome the above-mentioned defects and problems existing in the prior art and to provide a timely and effective method and system for fault detection and fault tolerance of a distributed power generation system.

[0005] To achieve the above objectives, the technical solution of the present invention is: a distributed power generation system fault detection and fault tolerance method, the distributed power generation system is used to supply power to a target power grid, the distributed power generation system includes a plurality of distributed power generation devices and a plurality of mobile energy storage devices; the method includes:

[0006] S1. Obtaining the location information of each distributed power generation device and the weather information of the location of the device;

[0007] S2. Based on the location information and weather information, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters;

[0008] S3, determining each distributed power generation equipment as a target power generation equipment, and obtaining the actual value of the power generation power of the target power generation equipment;

[0009] S4, obtaining the actual value of the power generation power of each distributed power generation device in each device cluster where the target power generation device is located;

[0010] S5. Determine an estimated value of the power generation of the target power generation equipment based on the actual power generation value of each distributed power generation equipment;

[0011] S6. Compare the actual value of the power generation of the target power generation equipment with the estimated value; the comparison results include the following:

[0012] The first type: if the actual power generation value of the target power generation equipment is greater than or equal to the preset ratio of the estimated value, then the target power generation equipment does not have a fault;

[0013] The second method: if the actual value of the power generation of the target power generation equipment is less than a preset ratio of the estimated value, the target power generation equipment is determined to be a faulty power generation equipment, and step S7 is performed;

[0014] S7. Confirm at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices, and then connect the multiple target mobile energy storage devices to a power grid to supply power to the target power grid.

[0015] The step S2 specifically includes:

[0016] S21, obtaining location information of each distributed power generation device and weather information of the location of the device, then normalizing the location information and weather information, and selecting key features for cluster analysis from the weather information based on the type of distributed power generation device;

[0017] S22. Based on the location information and key features, the spatial distance and weather difference between each distributed power generation device are calculated and weighted to obtain a comprehensive distance metric for each distributed power generation device; the expression of the comprehensive distance metric is as follows:

[0018] D combined =w spatial ×D spatial +w weather ×D weather ;

[0019] w spatial +w weather =1;

[0020] Where: D combined is the comprehensive distance metric, w spatial is the weight of the spatial distance, D spatialis the spatial distance, w weather is the weight of weather difference, D weather For weather differences;

[0021] S23. Based on the clustering algorithm and the comprehensive distance metric, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters.

[0022] The step S5 specifically includes:

[0023] S51, obtaining a target distance between each distributed power generation device and a target power generation device; the target distance refers to a spatial distance between each distributed power generation device and the target power generation device;

[0024] S52, confirming the target weight of each distributed power generation device based on the target distance; the expression of the target weight is as follows:

[0025] w=a*e b*d ;

[0026] Among them: w is the target weight, a and b are preset fixed parameters, and d is the target distance;

[0027] S53: Determine an estimated value of the power generation of the target power generation device based on the target weight and the actual value of the power generation of each distributed power generation device.

[0028] The step S53 specifically includes:

[0029] S531, selecting a number of devices from each distributed power generation device as candidate power generation devices;

[0030] S532, obtaining an estimated value of the power generation power of the target power generation equipment based on the target weight of the candidate power generation equipment and the actual value of the power generation power of the candidate power generation equipment;

[0031] The expression of the estimated power generation value is as follows:

[0032]

[0033] Where: P est is the estimated power generation capacity of the target power generation equipment, P i is the actual value of the power generation of the i-th candidate power generation equipment, w i is the target weight of the i-th candidate power generation equipment.

[0034] The preset ratio of the estimated value is 60%-80%.

[0035] The step S7 specifically includes:

[0036] S71, obtaining the charge and discharge times, charge and discharge time intervals, and remaining power of each mobile energy storage device;

[0037] S72, weighting the number of charge and discharge times, the charge and discharge time interval, and the remaining power, and calculating the priority score of each mobile energy storage device; the expression of the priority score is as follows:

[0038] P s =N t *W n +T i *W t +R c +W r ;

[0039] Where: P s is the priority score, N t is the number of charge and discharge times, W n is the weight of the number of charge and discharge times, T i is the charge and discharge time interval, W t is the weight of the charge and discharge time interval, R c is the remaining power, W r is the weight of the remaining power;

[0040] S73. Sort the mobile energy storage devices based on the priority scores, identify at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices in order, and connect the multiple target mobile energy storage devices to the power grid according to the priorities to supply power to the target power grid.

[0041] A distributed power generation system fault detection and fault tolerance system, the distributed power generation system is used to supply power to a target power grid, the distributed power generation system includes a plurality of distributed power generation devices and a plurality of mobile energy storage devices; the system includes:

[0042] The equipment information acquisition module is used to obtain the location information of each distributed power generation equipment and the weather information of the location where the equipment is located;

[0043] A cluster analysis module is used to perform cluster analysis on each distributed power generation device based on location information and weather information to obtain multiple device clusters;

[0044] A first power acquisition module, used to determine each distributed power generation device as a target power generation device, and obtain the real value of the power generation power of the target power generation device;

[0045] The second power acquisition module is used to obtain the actual value of the power generation power of each distributed power generation device in each device cluster where the target power generation device is located;

[0046] An estimated power acquisition module is used to determine the estimated power generation value of the target power generation device based on the actual power generation value of each distributed power generation device;

[0047] The fault confirmation module is used to compare the actual value of the power generation of the target power generation equipment with the estimated value; the comparison results include the following:

[0048] The first type: if the actual power generation value of the target power generation equipment is greater than or equal to the preset ratio of the estimated value, then the target power generation equipment does not have a fault;

[0049] The second type: if the actual value of the power generation of the target power generation equipment is less than the preset ratio of the estimated value, the target power generation equipment is determined to be a faulty power generation equipment, and the relevant steps of the fault-tolerant execution module are performed;

[0050] The fault-tolerant execution module is used to identify at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices, and then connect the multiple target mobile energy storage devices to the power grid to supply power to the target power grid.

[0051] The cluster analysis module obtains the device clustering cluster according to the following steps:

[0052] S21, obtaining location information of each distributed power generation device and weather information of the location of the device, then normalizing the location information and weather information, and selecting key features for cluster analysis from the weather information based on the type of distributed power generation device;

[0053] S22. Based on the location information and key features, the spatial distance and weather difference between each distributed power generation device are calculated and weighted to obtain a comprehensive distance metric for each distributed power generation device; the expression of the comprehensive distance metric is as follows:

[0054] D combined =w spatial ×D spatial +w weather ×D weather ;

[0055] w spatial +w weather =1;

[0056] Where: D combined is the comprehensive distance metric, w spatial is the weight of the spatial distance, D spatial is the spatial distance, w weather is the weight of weather difference, D weather For weather differences;

[0057] S23. Based on the clustering algorithm and the comprehensive distance metric, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters.

[0058] The estimated power acquisition module obtains the estimated value of the generated power according to the following steps:

[0059] S51, obtaining a target distance between each distributed power generation device and a target power generation device; the target distance refers to a spatial distance between each distributed power generation device and the target power generation device;

[0060] S52, confirming the target weight of each distributed power generation device based on the target distance; the expression of the target weight is as follows:

[0061] w=a*e b*d ;

[0062] Among them: w is the target weight, a and b are preset fixed parameters, and d is the target distance;

[0063] S53, determining an estimated value of the power generation of the target power generation device based on the target weight and the actual value of the power generation of each distributed power generation device;

[0064] S531, selecting a number of devices from each distributed power generation device as candidate power generation devices;

[0065] S532, obtaining an estimated value of the power generation power of the target power generation equipment based on the target weight of the candidate power generation equipment and the actual value of the power generation power of the candidate power generation equipment;

[0066] The expression of the estimated power generation value is as follows:

[0067]

[0068] Where: P est is the estimated power generation capacity of the target power generation equipment, P i is the actual value of the power generation of the i-th candidate power generation equipment, w i is the target weight of the i-th candidate power generation equipment.

[0069] The fault-tolerant execution module performs fault tolerance according to the following steps:

[0070] S71, obtaining the charge and discharge times, charge and discharge time intervals, and remaining power of each mobile energy storage device;

[0071] S72, weighting the number of charge and discharge times, the charge and discharge time interval, and the remaining power, and calculating the priority score of each mobile energy storage device; the expression of the priority score is as follows:

[0072] P s =Nt *W n +T i *W t +R c +W r ;

[0073] Where: P s is the priority score, N t is the number of charge and discharge times, W n is the weight of the number of charge and discharge times, T i is the charge and discharge time interval, W t is the weight of the charge and discharge time interval, R c is the remaining power, W r is the weight of the remaining power;

[0074] S73. Sort the mobile energy storage devices based on the priority scores, identify at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices in order, and connect the multiple target mobile energy storage devices to the power grid according to the priorities to supply power to the target power grid.

[0075] Compared with the prior art, the present invention has the following beneficial effects:

[0076] The present invention discloses a fault detection and fault-tolerance method and system for a distributed power generation system. The method first obtains the location information of each device in the system and the weather information of the device location and performs cluster analysis to obtain a device clustering cluster. Then, each device is respectively determined as a target power generation device to obtain its true value of power generation. Then, the true value of each device in each device clustering cluster is obtained. Then, based on the true value of power generation of each device, an estimated value is determined and compared with the true value to determine whether a fault exists. If a fault exists, a mobile energy storage device is selected for power supply. In application, the present design can accurately identify whether a device fails through cluster analysis based on location and weather, combined with the difference between the true value and the estimated value of power generation, and can timely and effectively locate the faulty device. Furthermore, a fault-tolerant reserve is provided for the power generation system through the mobile energy storage device, thereby comprehensively improving the accuracy and fault-tolerance rate of the power generation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a flow chart of the method steps of the present invention.

[0078] Figure 2 It is a schematic diagram of the system structure of the present invention.

[0079] Figure 3 It is a schematic diagram of the device structure of the present invention.

[0080] In the figure: device information acquisition module 1, cluster analysis module 2, first power acquisition module 3, second power acquisition module 4, estimated power acquisition module 5, fault confirmation module 6, fault-tolerant execution module 7, processor 8, memory 9, computer program code 91. DETAILED DESCRIPTION

[0081] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0082] Embodiment 1:

[0083] See also Figure 1 , a distributed power generation system fault detection and fault tolerance method, the distributed power generation system includes a plurality of distributed power generation devices and a plurality of mobile energy storage devices; the method includes:

[0084] S1. Obtaining the location information of each distributed power generation device and the weather information of the location of the device;

[0085] In this embodiment, most modern distributed power generation equipment (such as solar photovoltaic panels, wind turbines, etc.) have built-in GPS positioning systems, and the location information can be obtained in real time through the device's built-in GPS function to obtain the device's latitude and longitude information (i.e., location information).

[0086] In addition, use professional meteorological data service platforms (such as the National Meteorological Information Center, commercial meteorological data providers, etc.) to obtain real-time and historical weather information at the location of the equipment. For example, some meteorological data service platforms provide data customization services, which can provide weather data for specific areas and specific time periods according to user needs. Through the API interface, meteorological data can be integrated into the control center of the distributed power generation system to realize automatic data processing and analysis. These meteorological data service platforms usually provide a variety of meteorological data including temperature, humidity, wind speed, wind direction, precipitation, solar radiation intensity, etc.

[0087] For distributed power generation systems such as photovoltaic power stations and wind farms, special weather stations can also be set up to monitor and record local weather conditions. The weather stations can be connected to the power station's SCADA system or remote monitoring system to transmit meteorological data to the control center in real time.

[0088] S2. Based on the location information and weather information, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters;

[0089] Furthermore, the step S2 specifically includes:

[0090] S21, obtaining location information of each distributed power generation device and weather information of the location of the device, then normalizing the location information and weather information, and selecting key features for cluster analysis from the weather information based on the type of distributed power generation device;

[0091] In this embodiment, since the units and dimensions of the location information (latitude and longitude) and weather information (such as temperature, humidity, etc.) are different, these data need to be standardized or normalized to the same scale in advance for comparison and calculation.

[0092] The key features refer to the features selected from weather information that are most helpful for cluster analysis. The types of distributed power generation equipment refer to different types of power generation equipment such as solar power generation equipment and wind power generation equipment. The key features will vary depending on the type of power generation equipment. For example, for solar power generation equipment, solar radiation intensity may be a key feature; while for wind power generation equipment, wind speed and wind direction may be more important.

[0093] S22. Based on the location information and key features, the spatial distance and weather difference between each distributed power generation device are calculated and weighted to obtain a comprehensive distance metric for each distributed power generation device; the expression of the comprehensive distance metric is as follows:

[0094] D combined =w spatial ×D spatial +w weather ×D weather ;

[0095] w spatial +w weather =1;

[0096] Where: D combined is the comprehensive distance metric, w spatial is the weight of the spatial distance, D sptial is the spatial distance, w weather is the weight of weather difference, D weather For weather differences;

[0097] In this embodiment, the spatial distance between the two devices is calculated using longitude and latitude information, such as the Haversine formula, which can be used to calculate the great circle distance between two points. In addition to the spatial distance, the similarity of weather conditions also needs to be considered, and the difference in weather information is quantified as a distance metric, which is then combined with the spatial distance to form a comprehensive distance metric. For example, a weighted sum method can be used to assign different weights to spatial distance and weather differences according to business needs.

[0098] For weather differences, since they may include multiple meteorological factors (such as temperature, humidity, wind speed, etc.), we first need to define a comprehensive weather difference metric, which can be achieved by weighted summing the differences of various meteorological factors, or by directly using a distance metric (such as Euclidean distance) to calculate the differences between vectors composed of multiple meteorological factors.

[0099] Assume that there are two distributed generation devices A and B, which need to be clustered according to their geographical locations and weather conditions.

[0100] First, use the Haversing formula to calculate the spatial distance D between A and B spatial = 100 km, and then calculate the weather difference between A and B considering only the temperature and wind speed factors; assuming the temperature difference is 5 degrees, the wind speed difference is 2 meters per second, and assigning equal weights to these two factors, the weather difference can be simply expressed as some combination of these two differences, such as the weather difference D weather The Euclidean distance calculation formula is as follows:

[0101]

[0102] Assuming that the business considers the geographical location to be more important, the spatial distance weight w is assigned spatial为 0.7, then the weight of weather difference w weather is 0.3;

[0103] Then the weighted calculation is performed as follows:

[0104] D combined =0.7×100+0.3×5.4≈70+1.6=71.6;

[0105] Where D combined It represents the comprehensive distance between devices A and B. This distance takes into account the differences in geographical location and weather conditions, and assigns different weights to the two according to business needs. In the clustering process, this comprehensive distance can be used to measure the similarity between devices.

[0106] S23. Based on the clustering algorithm and the comprehensive distance metric, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters.

[0107] In this embodiment, the clustering algorithm may be a K-means algorithm, and the number of clusters K is preliminarily set (assuming K=5, there are 5 different types of power generation environments), and the devices are clustered according to the comprehensive distance.

[0108] Specifically, each device is assigned to the nearest cluster center, and the cluster center of each cluster is recalculated (usually the average of the comprehensive distances of all devices in the cluster); then the above steps are repeated until the cluster center no longer changes significantly or the preset number of iterations is reached. The silhouette coefficient, Calinski-Harabasz index or other cluster evaluation indicators can be used to evaluate the clustering effect.

[0109] Assume that there are 10 distributed power generation devices, and the comprehensive distance between each device has been calculated, and these distances are stored in a distance matrix; according to business needs or data visualization analysis, the data is divided into 5 clusters, and 5 devices are randomly selected as the initial cluster centers; in the first iteration, each device is assigned to the nearest cluster center according to the comprehensive distance, and the cluster center of each cluster is recalculated (that is, the average value of the comprehensive distance of all devices in the cluster is calculated, but here the centroid of all points in the cluster is usually used as the new cluster center, which may not be a simple average value in multidimensional space); then repeat the above steps until the cluster center no longer changes or reaches the preset number of iterations, and then use the silhouette coefficient or other indicators to evaluate the clustering effect. If the clustering effect is not ideal, you can try to adjust the number of clusters K or reinitialize the cluster center until you get the final multiple device clusters.

[0110] In practical applications, the selection of the number of clusters K usually needs to be determined in combination with business requirements and clustering evaluation indicators. In addition, the K-means algorithm is sensitive to the selection of the initial cluster centers, and different initial selections may lead to different clustering results. Therefore, it is sometimes necessary to run the K-means algorithm multiple times, each time using a different initial cluster center, and then select the best result.

[0111] S3, determining each distributed power generation equipment as a target power generation equipment, and obtaining the actual value of the power generation power of the target power generation equipment;

[0112] In this embodiment, assuming that there are N distributed power generation devices, the N devices are respectively identified as target power generation devices and subsequent calculations and judgments are performed, so as to screen out devices that may have faults.

[0113] S4, obtaining the actual value of the power generation power of each distributed power generation device in each device cluster where the target power generation device is located;

[0114] Modern power generation equipment is usually equipped with a monitoring system that can monitor and record the operating status and power generation of the equipment in real time. By accessing these monitoring systems, the true value of the power generation can be easily obtained.

[0115] S5. Determine an estimated value of the power generation of the target power generation equipment based on the actual power generation value of each distributed power generation equipment;

[0116] Furthermore, the step S5 specifically includes:

[0117] S51, obtaining a target distance between each distributed power generation device and a target power generation device; the target distance refers to a spatial distance between each distributed power generation device and the target power generation device;

[0118] S52, confirming the target weight of each distributed power generation device based on the target distance; the expression of the target weight is as follows:

[0119] w=a*e b*d ;

[0120] Among them: w is the target weight, a and b are preset fixed parameters, and d is the target distance;

[0121] S53: Determine an estimated value of the power generation of the target power generation device based on the target weight and the actual value of the power generation of each distributed power generation device.

[0122] The step S53 specifically includes:

[0123] S531, selecting a number of devices from each distributed power generation device as candidate power generation devices;

[0124] S532, obtaining an estimated value of the power generation power of the target power generation equipment based on the target weight of the candidate power generation equipment and the actual value of the power generation power of the candidate power generation equipment;

[0125] The expression of the estimated power generation value is as follows:

[0126]

[0127] Where: P est is the estimated power generation capacity of the target power generation equipment, P i is the actual value of the power generation of the i-th candidate power generation equipment, w i is the target weight of the i-th candidate power generation equipment.

[0128] In this embodiment, the actual value of the power generation power of the candidate power generation equipment can be directly obtained through the monitoring system equipped with the power generation equipment, and the calculation method of the target weight of the candidate power generation equipment is consistent with step S52.

[0129] Assume that there are three candidate power generation devices, and their actual power generation values ​​are as follows:

[0130] P 1 =100kW, P 2 =150kW, P3 =200kW;

[0131] The corresponding target weights are: W 1 =0.2, w 2 =0.3, w 3 =0.5;

[0132] Then the estimated power generation power of the target power generation equipment is P est The calculation is as follows:

[0133]

[0134] The target weight is the weight set for each distributed generation device relative to the target generation device, which is used to ensure that in subsequent calculations, scheduling or optimization, distributed generation devices that are farther away from the target generation device can receive more attention or resources; when calculating, first set the parameters: for example, a = 1, b = 0.1, and ensure that when d = 0, w = 1. It can be adjusted according to actual needs to ensure that the weight increases appropriately when the distance increases.

[0135] Then, for each distributed generation device, substitute the target distance d into the function w = e 0.1*d In the calculation, the corresponding target weight w is calculated; if the weight distribution is unreasonable or does not meet the actual needs, the function parameters a and b can be adjusted and the weight can be recalculated.

[0136] S6. Compare the actual value of the power generation of the target power generation equipment with the estimated value; the comparison results include the following:

[0137] The first type: if the actual power generation value of the target power generation equipment is greater than or equal to the preset ratio of the estimated value, then the target power generation equipment does not have a fault;

[0138] The second method: if the actual value of the power generation of the target power generation equipment is less than a preset ratio of the estimated value, the target power generation equipment is determined to be a faulty power generation equipment, and step S7 is performed;

[0139] The preset ratio is flexibly set and adjusted in advance according to the specific characteristics and operating environment of the power generation equipment, and should be set within a range that can accurately reflect the equipment failure state and have a certain fault tolerance and robustness. The preferred preset ratio is set to a specific value between 60% and 80%.

[0140] S7. Confirm at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices, and then connect the multiple target mobile energy storage devices to a power grid to supply power to the target power grid.

[0141] Furthermore, the step S7 specifically includes:

[0142] S71, obtaining the charge and discharge times, charge and discharge time intervals, and remaining power of each mobile energy storage device;

[0143] S72, weighting the number of charge and discharge times, the charge and discharge time interval, and the remaining power, and calculating the priority score of each mobile energy storage device; the expression of the priority score is as follows:

[0144] P s =N t *W n +T i *W t +R c +W r ;

[0145] Where: P s is the priority score, N t is the number of charge and discharge times, W n is the weight of the number of charge and discharge times, T i is the charge and discharge time interval, W t is the weight of the charge and discharge time interval, R c is the remaining power, W r is the weight of the remaining power;

[0146] The number of charge and discharge times reflects the frequency of use and potential aging of the device. Frequently charged and discharged devices may require higher priority maintenance or more cautious usage strategies. The weight can be set higher (such as 30%) based on the design life and usage of the device.

[0147] The charge and discharge interval reflects the usage status or battery status of the device. Devices that have not been used for a long time may need to check their battery status to prevent performance degradation caused by over-discharge or self-discharge of the battery. Its weight takes into account the need for battery health maintenance and can be set to a moderate weight (such as 30%).

[0148] The remaining power reflects the remaining power of the device, and devices with low remaining power may be more important in an emergency because they need to be charged or replaced faster. In order to ensure availability in an emergency, a higher weight (such as 40%) can be set.

[0149] When calculating, the raw data of each factor were first converted to a unified scale for comparison and weighting.

[0150] For example, the number of charge and discharge times can be normalized by dividing by the maximum number of charge and discharge times in the design life of the device; the time interval can be normalized by converting to days and dividing by the longest acceptable non-use time interval; the remaining power can be directly expressed as a percentage.

[0151] S73. Sort the mobile energy storage devices based on the priority scores, identify at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices in order, and connect the multiple target mobile energy storage devices to the power grid according to the priorities to supply power to the target power grid.

[0152] In this embodiment, the order sorting refers to sorting the priority scores from high to low. Assuming there are N mobile energy storage devices in total, but when meeting the fault tolerance target, only some of them may be needed to achieve it, then some mobile energy storage devices are selected in sequence to be confirmed as target mobile energy storage devices. There may be multiple target mobile energy storage devices, that is, to confirm multiple target mobile energy storage devices, multiple target mobile energy storage devices are connected to the power grid to achieve power supply fault tolerance.

[0153] Embodiment 2:

[0154] See also Figure 2 , a distributed power generation system fault detection and fault tolerance system, the distributed power generation system is used to supply power to a target power grid, the distributed power generation system includes a plurality of distributed power generation devices and a plurality of mobile energy storage devices; the system includes:

[0155] The device information acquisition module 1 is used to obtain the location information of each distributed power generation device and the weather information of the location where the device is located;

[0156] Cluster analysis module 2, used to perform cluster analysis on each distributed power generation device based on location information and weather information to obtain multiple device clusters;

[0157] Furthermore, the cluster analysis module 2 obtains the device clusters according to the following steps:

[0158] S21, obtaining location information of each distributed power generation device and weather information of the location of the device, then normalizing the location information and weather information, and selecting key features for cluster analysis from the weather information based on the type of distributed power generation device;

[0159] S22. Based on the location information and key features, the spatial distance and weather difference between each distributed power generation device are calculated and weighted to obtain a comprehensive distance metric for each distributed power generation device; the expression of the comprehensive distance metric is as follows:

[0160] D combined =w spatial ×D spatial +w wseather ×D weather ;

[0161] w spatial +w weather =1;

[0162] Where: D combined is the comprehensive distance metric, w spatial is the weight of the spatial distance, D spatial is the spatial distance, w weather is the weight of weather difference, D weather For weather differences;

[0163] S23. Based on the clustering algorithm and the comprehensive distance metric, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters.

[0164] A first power acquisition module 3, used to determine each distributed power generation device as a target power generation device, and obtain the real value of the power generation power of the target power generation device;

[0165] The second power acquisition module 4 is used to obtain the actual power generation value of each distributed power generation device in each device cluster where the target power generation device is located;

[0166] The estimated power acquisition module 5 is used to determine the estimated value of the power generation of the target power generation equipment based on the actual value of the power generation of each distributed power generation equipment;

[0167] Furthermore, the estimated power acquisition module 5 obtains the estimated value of the generated power according to the following steps:

[0168] S51, obtaining a target distance between each distributed power generation device and a target power generation device; the target distance refers to a spatial distance between each distributed power generation device and the target power generation device;

[0169] S52, confirming the target weight of each distributed power generation device based on the target distance; the expression of the target weight is as follows:

[0170] w=a*e b*d ;

[0171] Among them: w is the target weight, a and b are preset fixed parameters, and d is the target distance;

[0172] S53: Determine an estimated value of the power generation of the target power generation device based on the target weight and the actual value of the power generation of each distributed power generation device.

[0173] The step S53 specifically includes:

[0174] S531, selecting a number of devices from each distributed power generation device as candidate power generation devices;

[0175] S532, obtaining an estimated value of the power generation power of the target power generation equipment based on the target weight of the candidate power generation equipment and the actual value of the power generation power of the candidate power generation equipment;

[0176] The expression of the estimated power generation value is as follows:

[0177]

[0178] Where: P est is the estimated power generation capacity of the target power generation equipment, P i is the actual value of the power generation of the i-th candidate power generation equipment, w i is the target weight of the i-th candidate power generation equipment.

[0179] The fault confirmation module 6 is used to compare the actual value of the power generation of the target power generation equipment with the estimated value; the comparison results include the following:

[0180] The first type: if the actual power generation value of the target power generation equipment is greater than or equal to the preset ratio of the estimated value, then the target power generation equipment does not have a fault;

[0181] The second type: if the actual value of the power generation of the target power generation equipment is less than the preset ratio of the estimated value, the target power generation equipment is determined to be a faulty power generation equipment, and the relevant steps of the fault-tolerant execution module 7 are performed;

[0182] The fault-tolerant execution module 7 identifies at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices, and then connects the multiple target mobile energy storage devices to the power grid to supply power to the target power grid.

[0183] Further, the fault-tolerant execution module 7 performs fault tolerance according to the following steps:

[0184] S71, obtaining the charge and discharge times, charge and discharge time intervals, and remaining power of each mobile energy storage device;

[0185] S72, weighting the number of charge and discharge times, the charge and discharge time interval, and the remaining power, and calculating the priority score of each mobile energy storage device; the expression of the priority score is as follows:

[0186] P s =N t *W n +T i *W t +R c +W r ;

[0187] Where: P s is the priority score, N t is the number of charge and discharge times, W n is the weight of the number of charge and discharge times, T i is the charge and discharge time interval, W tis the weight of the charge and discharge time interval, R c is the remaining power, W r is the weight of the remaining power;

[0188] S73. Sort the mobile energy storage devices based on the priority scores, identify at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices in order, and connect the multiple target mobile energy storage devices to the power grid according to the priorities to supply power to the target power grid.

[0189] Embodiment 3:

[0190] See also Figure 3 , In this embodiment, a distributed power generation system fault detection and fault tolerance device is also included, and the device includes a processor 8 and a memory 9;

[0191] The memory 9 is used to store computer program code 91 and transmit the computer program code 91 to the processor 8;

[0192] The processor 8 is used to execute the distributed power generation system fault detection and fault tolerance method described in Example 1 according to the instructions in the computer program code 91.

[0193] Generally speaking, the computer instructions for implementing the method of the present invention may be carried in any combination of one or more computer-readable storage media. Non-transitory computer-readable storage media may include any computer-readable media, except for the signal itself that is temporarily propagating.

[0194] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EKROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0195] Computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, SMalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages, in particular, Python suitable for neural network computing and platform frameworks based on TensorFlow, PyTorch, etc. can be used. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer or to an external computer (for example, using an Internet service provider to connect via the Internet) through any type of network, including a local area network (LAN) or a wide area network (WAN).

[0196] The above-mentioned device and non-temporary computer-readable storage medium can refer to the specific description of a distributed power generation system fault detection and fault tolerance method and its beneficial effects, which will not be repeated here.

[0197] Although the embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A distributed power generation system fault detection and fault tolerance method, characterized in that: The distributed power generation system is used to supply power to a target power grid, and the distributed power generation system includes a plurality of distributed power generation devices and a plurality of mobile energy storage devices; the method includes: S1. Obtaining the location information of each distributed power generation device and the weather information of the location of the device; S2. Based on the location information and weather information, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters; S3, determining each distributed power generation equipment as a target power generation equipment, and obtaining the actual value of the power generation power of the target power generation equipment; S4, obtaining the actual value of the power generation power of each distributed power generation device in each device cluster where the target power generation device is located; S5. Determine an estimated value of the power generation of the target power generation equipment based on the actual power generation value of each distributed power generation equipment; S6. Compare the actual value of the power generation of the target power generation equipment with the estimated value; the comparison results include the following: The first type: if the actual power generation value of the target power generation equipment is greater than or equal to the preset ratio of the estimated value, then the target power generation equipment does not have a fault; The second method: if the actual value of the power generation of the target power generation equipment is less than a preset ratio of the estimated value, the target power generation equipment is determined to be a faulty power generation equipment, and step S7 is performed; S7. Confirm at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices, and then connect the multiple target mobile energy storage devices to a power grid to supply power to the target power grid.

2. A distributed power generation system fault detection and fault tolerance method according to claim 1, characterized in that: The step S2 specifically includes: S21, obtaining location information of each distributed power generation device and weather information of the location of the device, then normalizing the location information and weather information, and selecting key features for cluster analysis from the weather information based on the type of distributed power generation device; S22. Based on the location information and key features, the spatial distance and weather difference between each distributed power generation device are calculated and weighted to obtain a comprehensive distance metric for each distributed power generation device; the expression of the comprehensive distance metric is as follows: D combined =w spatial ×D spatial +w weather ×D weather ; In spatial +in weather =1; Where: D combined is the comprehensive distance metric, w spatial is the weight of the spatial distance, D spatial is the spatial distance, w weather is the weight of weather difference, D weather For weather differences; S23. Based on the clustering algorithm and the comprehensive distance metric, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters.

3. A distributed power generation system fault detection and fault tolerance method according to claim 1, characterized in that: The step S5 specifically includes: S51, obtaining a target distance between each distributed power generation device and a target power generation device; the target distance refers to a spatial distance between each distributed power generation device and the target power generation device; S52, confirming the target weight of each distributed power generation device based on the target distance; the expression of the target weight is as follows: w=a* b*d ; Among them: w is the target weight, a and b are preset fixed parameters, and d is the target distance; S53: Determine an estimated value of the power generation of the target power generation device based on the target weight and the actual value of the power generation of each distributed power generation device.

4. A distributed power generation system fault detection and fault tolerance method according to claim 3, characterized in that: The step S53 specifically includes: S531, selecting a number of devices from each distributed power generation device as candidate power generation devices; S532, obtaining an estimated value of the power generation power of the target power generation equipment based on the target weight of the candidate power generation equipment and the actual value of the power generation power of the candidate power generation equipment; The expression of the estimated power generation value is as follows: Where: P est is the estimated power generation capacity of the target power generation equipment, P i is the actual value of the power generation of the i-th candidate power generation equipment, w i is the target weight of the i-th candidate power generation equipment.

5. A distributed power generation system fault detection and fault tolerance method according to claim 1, characterized in that: The preset ratio of the estimated value is 60%-80%.

6. A distributed power generation system fault detection and fault tolerance method according to claim 1, characterized in that: The step S7 specifically includes: S71, obtaining the charge and discharge times, charge and discharge time intervals, and remaining power of each mobile energy storage device; S72, weighting the number of charge and discharge times, the charge and discharge time interval, and the remaining power, and calculating the priority score of each mobile energy storage device; the expression of the priority score is as follows: P s =N t *W n +T i *W t +R C +W r ; Where: P s is the priority score, N t is the number of charge and discharge times, W n is the weight of the number of charge and discharge times, T i is the charge and discharge time interval, W t is the weight of the charging and discharging time interval, R c is the remaining power, W r is the weight of the remaining power; S73. Sort the mobile energy storage devices based on the priority scores, identify at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices in order, and connect the multiple target mobile energy storage devices to the power grid according to the priorities to supply power to the target power grid.

7. A distributed power generation system fault detection and fault tolerance system, characterized in that: The distributed power generation system is used to supply power to the target power grid, and the distributed power generation system includes a plurality of distributed power generation devices and a plurality of mobile energy storage devices; the system includes: The device information acquisition module (1) is used to acquire the location information of each distributed power generation device and the weather information of the location where the device is located; A cluster analysis module (2), used to perform cluster analysis on each distributed power generation device based on location information and weather information to obtain multiple device clusters; A first power acquisition module (3) is used to determine each distributed power generation device as a target power generation device and obtain the real value of the power generation of the target power generation device; A second power acquisition module (4) is used to acquire the actual power generation value of each distributed power generation device in each device cluster where the target power generation device is located; An estimated power acquisition module (5) is used to determine an estimated value of the power generation of the target power generation device based on the actual value of the power generation of each distributed power generation device; The fault confirmation module (6) is used to compare the actual value of the power generation of the target power generation equipment with the estimated value; the comparison results include the following: The first type: if the actual power generation value of the target power generation equipment is greater than or equal to the preset ratio of the estimated value, then the target power generation equipment does not have a fault; The second method: if the actual value of the power generation of the target power generation equipment is less than a preset ratio of the estimated value, the target power generation equipment is determined to be a faulty power generation equipment, and the relevant steps of the fault-tolerant execution module (7) are performed; The fault-tolerant execution module (7) is used to identify at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices, and then connect the multiple target mobile energy storage devices to the power grid to supply power to the target power grid.

8. A distributed power generation system fault detection and fault tolerance system according to claim 7, characterized in that: The cluster analysis module (2) obtains the device clustering cluster according to the following steps: S21, obtaining location information of each distributed power generation device and weather information of the location of the device, then normalizing the location information and weather information, and selecting key features for cluster analysis from the weather information based on the type of distributed power generation device; S22. Based on the location information and key features, the spatial distance and weather difference between each distributed power generation device are calculated and weighted to obtain a comprehensive distance metric for each distributed power generation device; the expression of the comprehensive distance metric is as follows: D combined =w spatial ×D spatial +w weather ×D weather ; In spatial +in weather =1; Where: D combined is the comprehensive distance metric, w spatial is the weight of the spatial distance, D spatial is the spatial distance, w weather is the weight of weather difference, D weather For weather differences; S23. Based on the clustering algorithm and the comprehensive distance metric, cluster analysis is performed on each distributed power generation device to obtain multiple device clusters.

9. A distributed power generation system fault detection and fault tolerance system according to claim 7, characterized in that: The estimated power acquisition module (5) obtains the estimated value of the generated power according to the following steps: S51, obtaining a target distance between each distributed power generation device and a target power generation device; the target distance refers to a spatial distance between each distributed power generation device and the target power generation device; S52, confirming the target weight of each distributed power generation device based on the target distance; the expression of the target weight is as follows: w=a*e b*d ; Among them: w is the target weight, a and b are preset fixed parameters, and d is the target distance; S53, determining an estimated value of the power generation of the target power generation device based on the target weight and the actual value of the power generation of each distributed power generation device; S531, selecting a number of devices from each distributed power generation device as candidate power generation devices; S532, obtaining an estimated value of the power generation power of the target power generation equipment based on the target weight of the candidate power generation equipment and the actual value of the power generation power of the candidate power generation equipment; The expression of the estimated power generation value is as follows: Where: P est is the estimated power generation capacity of the target power generation equipment, P i is the actual value of the power generation of the i-th candidate power generation equipment, w i is the target weight of the i-th candidate power generation equipment.

10. A distributed power generation system fault detection and fault tolerance system according to claim 7, characterized in that: The fault-tolerant execution module (7) executes fault tolerance according to the following steps: S71, obtaining the charge and discharge times, charge and discharge time intervals, and remaining power of each mobile energy storage device; S72, weighting the number of charge and discharge times, the charge and discharge time interval, and the remaining power, and calculating the priority score of each mobile energy storage device; the expression of the priority score is as follows: P s =N t *W n +T i *W t +R c +W r ; Where: P s is the priority score, N t is the number of charge and discharge times, W n is the weight of the number of charge and discharge times, T i is the charge and discharge time interval, W t is the weight of the charging and discharging time interval, R c is the remaining power, W r is the weight of the remaining power; S73. Sort the mobile energy storage devices based on the priority scores, identify at least some of the multiple mobile energy storage devices as multiple target mobile energy storage devices in order, and connect the multiple target mobile energy storage devices to the power grid according to the priorities to supply power to the target power grid.

Images