A multi-device polling control method and system for PLC power communication automatic control cabinet

Abstract

The present invention provides a multi-device polling control method and system for a PLC power communication automatic control cabinet. It relates to the technical field of PLC power communication. The method includes: obtaining historical state collection data of multiple slave devices, performing state fluctuation collaborative clustering on multiple slave devices to obtain multiple state collaborative device sets, generating global fault probability parameters and fault coordination parameters of the state collaborative device sets according to historical fault data of the slave devices; performing host polling allocation analysis on multiple state collaborative device sets to determine the number of polling control hosts of the state collaborative device sets; segmenting the state collaborative device sets to obtain multiple target collaborative device sets, determining the target polling priority parameters and global monitoring sorting reference list of each target collaborative device set, and generating polling control strategies for multiple slave devices, and performing polling control on multiple slave devices based on the polling control strategies. The present invention achieves the goal of improving the comprehensiveness of multi-device polling data collection.

Description

Technical Field

[0001] The present invention relates to the technical field of PLC power communication, and in particular to a multi-device polling control method and system for a PLC power communication automatic control cabinet. Background Art

[0002] With the rapid development of smart grid and automation technology, the application of automatic control cabinets based on PLC power communication technology in power systems is becoming more and more extensive. In this process, the automatic control cabinet, as the host device and core control unit, realizes remote monitoring and control of many power equipment such as transformers, current protection equipment, electric meters, sensors, etc. through PLC power communication technology.

[0003] However, some PLC power communication systems usually use polling mode for data collection, that is, the host device collects data from the slave device through periodic polling. Although this method is simple, it may still cause resource occupation when facing the needs of multi-device and high-frequency data collection. Even if the polling strategy is simply set according to the data collection priority of different power equipment, some devices have synergistic effects, that is, there are synergistic changes in state parameters. For example, if the voltage data of a certain device increases, the current of the device that is synergistically associated with it may also increase synergistically. If these devices all need to collect data first due to priority division, and one of the devices is abnormal and may change due to the synergistic change of the device state, so that the data corresponding to the device that is synergistically associated with it can be detected, then the data of these devices will be collected at the same time, which will lead to resource occupation. When other devices fail and need to collect data, the data collection will be delayed due to the low priority. The control management of polling data collection needs to be optimized. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a multi-device polling control method and system for a PLC power communication automatic control cabinet. The equipment is classified by analyzing the coordinated change characteristics of the states of different power equipment, and the local priority relationship between the equipment in each equipment set and the global priority relationship that needs to be monitored between different equipment sets are analyzed in combination with the historical fault data of the power equipment. The data polling and collection strategy of a large number of power equipment under multi-host management is optimized to avoid multiple devices frequently occupying polling resources and improve the working efficiency of the power communication system.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A first aspect of the present invention provides a multi-device polling control method for a PLC power communication automatic control cabinet, comprising:

[0007] Acquire historical state collection data corresponding to a plurality of slave devices respectively, extract a state fluctuation sequence of the slave devices from the historical state collection data, perform state fluctuation collaborative clustering on the plurality of slave devices based on the state fluctuation sequence, and obtain a plurality of state collaborative device sets;

[0008] Extract historical fault data corresponding to multiple slave devices in each state collaborative device set, extract local fault probability parameters and fault state sequences of each slave device from multiple groups of historical fault data, and generate global fault probability parameters and fault coordination parameters of the state collaborative device set according to the local fault probability parameters and fault state sequences;

[0009] Determine the local polling priority parameter of each state collaborative device set based on the global fault probability parameter of the state collaborative device set, perform host polling allocation analysis on multiple state collaborative device sets according to the global fault probability parameter and the fault coordination parameter, and determine the number of polling control hosts for each state collaborative device set;

[0010] Acquire monitoring importance parameters of multiple slave devices, perform monitoring sorting on multiple slave devices in each state collaborative device set according to the monitoring importance parameters and local failure probability parameters, and obtain a local monitoring sorting reference list in each state collaborative device set;

[0011] According to the number of polling control hosts of the state collaborative device set and the local monitoring sorting reference list, multiple state collaborative device sets are divided to obtain multiple target collaborative device sets, and the target polling priority parameter and the global monitoring sorting reference list of each target collaborative device set are determined;

[0012] A polling control strategy for the slave device is determined according to the target polling priority parameter and the global monitoring sorting reference list, and polling control is performed on multiple slave devices based on the polling control strategy.

[0013] Preferably, generating a global fault probability parameter and a fault coordination parameter of a state coordination device set according to a local fault probability parameter and a fault state sequence includes:

[0014] Determine multiple fault events of the slave device according to the historical fault data of the slave device, determine the local fault probability parameter of the slave device according to the number of occurrences of the fault event, and take the mean value corresponding to the local fault probability parameters of multiple slave devices in the state collaborative device set as the global fault probability parameter of the state collaborative device set;

[0015] The timestamp of each fault event of the slave device is extracted, and the fault state sequence of the slave device is generated according to the timestamp of the fault event. The fault state timing matrix of the state coordination device set is constructed according to multiple fault state sequences. The fault state timing matrix is ​​analyzed using the following formula to calculate the fault coordination parameters of the state coordination device set:

[0016]

[0017] In the formula, Indicates the fault coordination parameters of the state coordination device set, Indicates the number of slave devices in the state coordination device set, Indicates the fault overlap parameter, Indicates the status of the collaborative device set The slave device The timestamp of the fault event, Indicates A reference time window, Indicates the status of the collaborative device set The number of fault events for slave devices, represents the number of reference time windows, express For the fault overlap parameter, if the timestamp of the fault event belongs to the corresponding reference time window, it is recorded as 1, otherwise it is recorded as 0.

[0018] Preferably, performing host polling allocation analysis on multiple state collaborative device sets according to the global fault probability parameter and the fault coordination parameter to determine the number of polling control hosts for each state collaborative device set includes:

[0019] The ratio between the global failure probability parameter and the failure coordination parameter of the state coordination device set is used as the distributed control parameter of the state coordination device set, and the distributed control parameters corresponding to the multiple state coordination device sets are calculated;

[0020] Determine the distributed control range according to the distributed control parameters corresponding to multiple state collaborative device sets, obtain the number of host devices, divide the distributed control range according to the number of host devices to obtain multiple local control ranges, determine the number of reference hosts for each local control range, determine the local control range corresponding to each state collaborative device set according to the distributed control parameters corresponding to multiple state collaborative device sets, and use the number of reference hosts associated with the local control range corresponding to the state collaborative device set as the number of polling control hosts for the state collaborative device set.

[0021] Preferably, multiple state collaborative device sets are segmented according to the number of polling control hosts of the state collaborative device sets and the local monitoring sorting reference list to obtain multiple target collaborative device sets, and the target polling priority parameter and the global monitoring sorting reference list of each target collaborative device set are determined, including:

[0022] According to the number of polling control hosts of the state collaborative device set, multiple target collaborative device sets of each state collaborative device set are constructed, and after the multiple target collaborative device sets of each state collaborative device set are randomly sorted, multiple slave devices in the state collaborative device set are sequentially allocated to the multiple sorted target collaborative device sets according to the monitoring sort reference list, and multiple cycles are performed until the state collaborative device set is split, so as to obtain multiple target collaborative device sets corresponding to each state collaborative device set;

[0023] Generate a target polling priority parameter for each target collaborative device set according to an order determined by randomly sorting multiple target collaborative device sets of each state collaborative device set, and generate a global monitoring sorting reference list for the target collaborative device set according to the priority parameters of multiple slave devices in the target collaborative device set;

[0024] The priority parameter of the slave device is calculated based on the monitoring importance parameter and the local fault probability parameter of the slave device, and the priority parameter of the slave device is generated after weighted optimization of the monitoring importance parameter of the slave device by the local fault probability parameter.

[0025] Preferably, determining a polling control strategy for a slave device according to a target polling priority parameter and a global monitoring order reference list includes:

[0026] generating an allocation priority list for multiple target collaborative device sets according to the target polling priority parameter, and after determining the allocation order of the multiple host devices, allocating the first target collaborative device set to each host device according to the allocation order of the host device and the allocation priority list for the multiple target collaborative device sets;

[0027] Load limits and collaborative interference limits are allocated to the remaining multiple target collaborative device sets in the allocation priority list, wherein, for the target collaborative device set with the largest target polling priority parameter in the allocation priority list, the target load quantity of each current host device is determined, and after screening out multiple host devices without collaborative interference, the target collaborative device set with the largest target polling priority parameter is allocated to the host device with the lowest target load quantity among the multiple host devices without collaborative interference, and the target load quantity of each host device is updated after each allocation is completed until the allocation of the remaining multiple target collaborative device sets in the allocation priority list is completed, and a polling control strategy for the slave device is generated in combination with the global monitoring sorting reference list of each target collaborative device set;

[0028] Among them, for multiple host devices without coordinated interference, if among the multiple target collaborative device sets currently assigned to the host device, there is no collaborative device set belonging to the same state as the target collaborative device set currently to be assigned, it is considered that there is no coordinated interference between the host device and the target collaborative device set currently to be assigned.

[0029] Preferably, the target polling priority parameter of each target collaborative device set further includes:

[0030] Determine a local polling priority parameter of a state collaborative device set, determine at least one adjacent state collaborative device set of each state collaborative device set, and limit target polling priority parameters of multiple target collaborative device sets under the state collaborative device set according to the local polling priority parameter of at least one adjacent state collaborative device set of the state collaborative device set.

[0031] A second aspect of the present invention provides a PLC power communication automatic control cabinet multi-device polling control system, which is used to implement the above-mentioned PLC power communication automatic control cabinet multi-device polling control method, including:

[0032] The device state fluctuation analysis module is used to obtain the historical state collection data corresponding to multiple slave devices, extract the state fluctuation sequence of the slave devices from the historical state collection data, and perform state fluctuation collaborative clustering on multiple slave devices based on the state fluctuation sequence to obtain multiple state collaborative device sets;

[0033] A local fault analysis module is used to extract historical fault data corresponding to multiple slave devices in each state collaborative device set, extract local fault probability parameters and fault state sequences of each slave device from multiple groups of historical fault data, and generate global fault probability parameters and fault coordination parameters of the state collaborative device set based on the local fault probability parameters and fault state sequences;

[0034] A fault discrete analysis module is used to determine the local polling priority parameter of each state collaborative device set based on the global fault probability parameter of the state collaborative device set, perform host polling allocation analysis on multiple state collaborative device sets according to the global fault probability parameter and the fault coordination parameter, and determine the number of polling control hosts for each state collaborative device set;

[0035] A local monitoring sequence analysis module is used to obtain monitoring importance parameters of multiple slave devices, and to perform monitoring sorting on multiple slave devices in each state collaborative device set according to the monitoring importance parameters and local failure probability parameters, so as to obtain a local monitoring sorting reference list in each state collaborative device set;

[0036] A collaborative segmentation module is used to segment multiple state collaborative device sets according to the number of polling control hosts of the state collaborative device sets and the local monitoring sorting reference list to obtain multiple target collaborative device sets, and determine the target polling priority parameter and the global monitoring sorting reference list of each target collaborative device set;

[0037] The polling control optimization module is used to determine the polling control strategy for the slave device according to the target polling priority parameter and the global monitoring sorting reference list, and perform polling control on multiple slave devices based on the polling control strategy.

[0038] The present invention has the following beneficial effects:

[0039] The present invention collects data of historical states of different slave devices through core analysis, mines the cooperative association characteristics between slave devices and performs state fluctuation cooperative clustering on multiple slave devices, so as to divide them into different state cooperative device sets according to the cooperative working characteristics between the devices, so as to facilitate targeted optimization according to the characteristics of different types of devices, and further analyzes the fault conditions of different slave devices, analyzes the overall fault state of multiple devices in the state cooperative device set and the fault conditions between different state cooperative device sets, determines the priority information that needs to be polled about each state cooperative device set, and divides the state cooperative device set in combination with the fault cooperative characteristics of multiple devices in the state cooperative device set, and finally formulates a polling control strategy for slave devices in a multi-host device scenario in combination with the monitoring importance parameters and fault occurrence conditions of different slave devices, and performs polling control on multiple slave devices based on the polling control strategy, so that during the polling process, the power system can obtain the state information of different types of slave devices in a more comprehensive manner at different time periods, ensure that high-priority devices can be monitored in a timely manner, and avoid waste of resources or reduced efficiency due to the cooperative characteristics between slave devices, improve the monitoring efficiency of the system, maintain reasonable use of resources, and reduce system bottlenecks and monitoring blank periods. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A flowchart of a multi-device polling control method for a PLC power communication automatic control cabinet is provided as one of the embodiments of the present invention.

[0041] Figure 2 A schematic structural diagram of a PLC electric power communication automatic control cabinet multi-device polling control system provided in one of the embodiments of the present invention. DETAILED DESCRIPTION

[0042] In order to make those skilled in the art better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0043] See also Figure 1 The embodiment of the present invention provides a PLC power communication automatic control cabinet multi-device polling control method, comprising the following steps:

[0044] Step S1, obtaining historical state collection data corresponding to multiple slave devices respectively, extracting the state fluctuation sequence of the slave devices from the historical state collection data, and performing state fluctuation collaborative clustering on the multiple slave devices based on the state fluctuation sequence to obtain multiple state collaborative device sets.

[0045] In this embodiment, the historical state acquisition data of multiple slave devices within a period of time is first obtained. These data include detection data corresponding to the state parameters collected or collected by each slave device during operation, including different state parameters of the power equipment such as voltage, current, load and other information. In the PLC power communication system, the host device, such as the automatic control cabinet, is responsible for overall control, coordination, data acquisition, fault detection and other functions. The host device sends a polling request to the slave device to obtain data. The slave device is a specific power device or a control unit of the device, such as a certain electric meter, a certain sensor or a transformer, etc., which is responsible for responding to the request of the host device and providing the operation data and state information of the power equipment. There may be different associations between slave devices. For example, some protection devices, sensors and controllers may work together for the same protection target, such as current protection, overload protection, etc., so that one of the devices fails or the data is abnormal, which is likely to cause the state change of other devices. For example, when the load of a transformer is too high, it may cause abnormal fluctuations in current. This coordinated fluctuation is collected by different slave devices and sent to the host device.

[0046] In order to analyze the behavior changes of multiple slave devices under the same or similar working environment, the state fluctuation sequence of each device, such as voltage sequence, current sequence, temperature sequence, etc., can be extracted from the historical state data of these slave devices, so as to identify the potential associations between different slave devices. For the multiple extracted state fluctuation sequences, clustering algorithms such as K-means and DBSCAN are used to collaboratively cluster multiple slave devices, with the goal of classifying devices with similar state fluctuation characteristics into one category, thereby generating multiple state collaborative device sets. Taking the DBSCAN algorithm as an example, the distance between different devices is analyzed through the state fluctuation sequence of the device, so that multiple slave devices with high state coordination, or with similar fluctuation trends in the same time period, are clustered together.

[0047] Step S2, extract the historical fault data corresponding to multiple slave devices in each state collaborative device set, extract the local fault probability parameters and fault state sequence of each slave device from the multiple groups of historical fault data, and generate the global fault probability parameters and fault coordination parameters of the state collaborative device set according to the local fault probability parameters and fault state sequence.

[0048] In this embodiment, the fault conditions of each slave device in each state collaborative device set are deeply analyzed. Specifically, the historical fault data of the slave device can be determined based on the historical state collection data, that is, the information that records the time type and other information of the fault in detail, and then the probability of each slave device failing in a period of history is determined to obtain the local fault probability parameter of the slave device, and according to the multiple fault events in the historical fault data and the time information corresponding to the fault events, a fault state sequence about the time characteristics of the fault event is constructed. By deeply analyzing the local fault probability parameters and the fault state sequence, the global fault probability parameter of each state collaborative device set is calculated to reflect the overall situation of the probability of failure of multiple slave devices in the state collaborative device set. In order to better understand the correlation between the failures of multiple devices, the fault coordination parameter of each state collaborative device set is further calculated to measure the coordination degree of failure of devices in the same type of device set. If the probability of multiple devices failing at the same time is high, it means that these devices have strong coordination when the failure occurs, so as to manage multiple slave devices more carefully according to the coordination of the failure.

[0049] Step S3: determine the local polling priority parameter of each state collaborative device set based on the global fault probability parameter of the state collaborative device set, perform host polling allocation analysis on multiple state collaborative device sets according to the global fault probability parameter and the fault coordination parameter, and determine the number of polling control hosts for each state collaborative device set.

[0050] In this embodiment, the polling priority parameter of each device set is determined by a global failure probability parameter that characterizes the overall characteristics of failures of multiple slave devices in a state collaborative device set. Specifically, a device set with a higher failure probability has a greater risk of failure, and therefore needs to be polled and monitored first. These device sets should be given a higher polling priority to ensure that device failures can be captured in a timely manner. Among them, the local polling priority parameter can simply characterize the priority of polling data collection for multiple state collaborative device sets, for example, according to the order of the global failure probability parameter from high to low, it is recorded as 1, 2, 3 and other local polling priority parameters. Then, according to the collaborative characteristics of failures of multiple slave devices in the device set, it is determined whether the device set should be monitored separately by a host or jointly monitored by multiple hosts, and a specific quantitative analysis is performed based on the global failure probability parameter and the failure coordination parameter to obtain a specific device allocation situation. For example, a set of devices with a higher risk of failure and lower fault coordination may require multiple hosts to monitor simultaneously to ensure real-time data collection and fault detection, and to avoid the situation where, during monitoring by a single host, it is difficult to promptly discover possible faults through the status data of a few devices due to polling efficiency limitations. For a set of devices with higher fault coordination, possible faults can be discovered through the status data of a few slave devices, without occupying too many polling resources.

[0051] Step S4, obtaining monitoring importance parameters of multiple slave devices, and performing monitoring sorting on multiple slave devices in each state collaborative device set according to the monitoring importance parameters and local failure probability parameters to obtain a local monitoring sorting reference list in each state collaborative device set.

[0052] In this embodiment, the monitoring importance parameter assigned to each slave device is collected. This parameter is pre-evaluated based on the function of the device, the frequency of use, the impact on the system security and other factors. The higher the importance of the device, the greater its impact on the entire power system, so it needs to be monitored first. For example, the technical personnel of the power equipment operation and maintenance can divide different devices into multiple monitoring importance levels according to the importance of the equipment, such as level one, level two, level three, etc., and make a preliminary division based on expert experience in this way. However, this method of biased experience setting has limitations. The present invention further combines the local fault probability parameter of each slave device to optimize the acquired monitoring importance parameter based on the occurrence of the fault. Specifically, according to the local fault probability parameter, different slave devices that are pre-divided and belong to the same monitoring importance level are further monitored and sorted. For example, a certain electric meter and a certain sensor both correspond to the same monitoring importance, but the frequency of the sensor device detecting fault data in the historical data is significantly higher than that of the electric meter device. In the actual data collection, the sensor device needs to obtain data with priority collection compared to the electric meter device. In this way, multiple slave devices in each state collaborative device set are monitored and sorted to generate a corresponding local monitoring sort reference list. Devices with higher failure probability and higher monitoring importance are placed at the front of the list, and data collection and fault detection are given priority. Devices with lower failure risk or lower monitoring importance can be placed at the back to reduce their collection priority.

[0053] Step S5: divide multiple state collaborative device sets into multiple target collaborative device sets according to the number of polling control hosts of the state collaborative device sets and the local monitoring sorting reference list, and determine the target polling priority parameter and the global monitoring sorting reference list of each target collaborative device set.

[0054] In this embodiment, according to the number of polling control hosts for each state collaborative device set determined in the aforementioned steps, and a local monitoring ranking reference list characterizing the monitoring importance ranking of different slave devices in different slave device groups, multiple state collaborative device sets are segmented to obtain multiple target collaborative device sets. Because some state collaborative device sets need to be allocated to multiple host devices for polling data collection, this process is combined with the local monitoring ranking reference list for multiple slave devices in the device set. When the host devices are roughly regarded as having the same priority, multiple slave devices are allocated to multiple host devices as evenly as possible to obtain multiple target collaborative device sets. At the same time, according to the global failure probability parameters corresponding to the multiple state collaborative device sets before segmentation, the target polling priority parameters characterizing the relative importance of monitoring of the target collaborative device set after segmentation are determined, as well as the global monitoring ranking reference list characterizing the relative importance of detection of multiple slave devices contained in the target collaborative device set, so that different slave devices can be reasonably polled and managed based on this information to ensure effective use of resources and minimize resource waste.

[0055] Step S6: determine a polling control strategy for the slave device according to the target polling priority parameter and the global monitoring order reference list, and perform polling control on the multiple slave devices based on the polling control strategy.

[0056] In this embodiment, based on the target polling priority parameters of each target collaborative device set and the global monitoring sorting reference list, I determine the polling control strategy for multiple slave devices. The strategy determines the polling monitoring order of different slave devices under the joint management of multiple hosts and the coordination rules when multiple hosts work together. The power system performs periodic polling monitoring on multiple slave devices through the polling control strategy. During the polling data collection process, high-priority devices are given priority for monitoring based on the device priority and the polling strategy. In particular, considering that in each type of slave device with a collaborative change relationship, multiple host devices will at least poll and collect some of the more important devices in a relatively short period of time, avoiding In a short period of time, multiple host devices perform unnecessary centralized data collection on multiple slave devices in a collaborative change relationship, because for multiple slave devices in a collaborative change relationship, the data collected from a few devices can represent the status of this type of device to a certain extent, thereby providing early warning of the status of other collaboratively associated devices in the current period. Therefore, it is necessary to poll as many slave devices in different collaborative relationships as possible in a relatively short period of time, so as to grasp the short-term performance of a type of slave devices corresponding to different sets of devices in each period as much as possible. Considering the priority of the slave devices that need to be polled for data collection, combined with the limitation of polling resources and the necessary time spent on polling, multiple slave devices are globally monitored and managed as much as possible. In the above manner, the polling control optimization scheme for multiple power equipment provided by the present invention can greatly improve the resource utilization efficiency of the system while ensuring the integrity of equipment monitoring, and ensure that when a device fails, fault information can be captured in a timely and accurate manner, providing a strong guarantee for the safe and stable operation of the power system.

[0057] In one embodiment, for the above step S2, generating a global fault probability parameter and a fault coordination parameter of a state coordination device set according to a local fault probability parameter and a fault state sequence includes the following contents:

[0058] According to the historical fault data of the slave device, multiple fault events of the slave device are determined, and the local fault probability parameter of the slave device is determined according to the number of occurrences of the fault events. The mean value corresponding to the local fault probability parameters of multiple slave devices in the state collaborative device set is taken as the global fault probability parameter of the state collaborative device set.

[0059] In this embodiment, multiple fault events contained in the historical fault data and the timestamp of each fault event are determined based on the historical fault data of the slave device, and the number of fault events corresponding to the slave device in the corresponding time period in the historical fault data is counted, so as to calculate a local fault probability parameter characterizing the probability of fault occurrence, such as the ratio of the number of fault event occurrences to the total number of sampling time periods, and further calculate the average of the local fault probability parameters of multiple slave devices in the state collaborative device set as the global fault probability parameter of the state collaborative device set to reflect the overall failure probability of the device set.

[0060] Extract the timestamp of each fault event of the slave device, and generate the fault state sequence of the slave device according to the timestamp of the fault event, wherein the fault state sequence is specifically used to characterize the situation where the fault occurs in multiple sampling periods. The sampling period can be reasonably set according to different power demand scenarios. This implementation does not specifically limit it. For example, taking one day as an example, every half hour is regarded as a sampling period. If a fault event occurs in the sampling period, the element value corresponding to the period can be recorded as 1, otherwise it is recorded as 0. In this way, the fault state sequence of the slave device is generated, and multiple fault state sequences corresponding to the same state collaborative device set are spliced ​​to construct the fault state timing matrix of the state collaborative device set. The fault state timing matrix is ​​analyzed using the following formula to calculate the fault coordination parameters of the state collaborative device set:

[0061]

[0062] In the formula, Indicates the fault coordination parameter of the state coordination device set, which is used to reflect the coordination degree of device failures in the state coordination device set. Indicates the number of slave devices in the state coordination device set, Indicates the fault overlap parameter, Indicates the status of the collaborative device set The slave device The timestamp of the fault event, Indicates A reference time window, Indicates the status of the collaborative device set The number of fault events for slave devices, represents the number of reference time windows, express The total window length corresponding to the reference time window is 1, and the reference time window contains at least one sampling period. For the fault overlap parameter, if the timestamp of the fault event belongs to the corresponding reference time window, it is recorded as 1, otherwise it is recorded as 0.

[0063] The coordination degree of device failures in each reference time window is calculated in the above manner to measure the coordination of device failures in the state coordination device set. If the probability of devices failing in the same time window is high, it means that these devices have strong coordination when failures occur. For example, if the failure of one device causes other devices to collect fault data and show a fault state, the fault coordination parameter is large. On the contrary, if the failure time distribution of the device is relatively discrete, the fault coordination is small.

[0064] In one embodiment, for the above step S3, host polling allocation analysis is performed on multiple state collaborative device sets according to the global fault probability parameter and the fault coordination parameter to determine the number of polling control hosts for each state collaborative device set, specifically including:

[0065] The ratio between the global failure probability parameter and the failure coordination parameter of the state coordination device set is used as the distributed control parameter of the state coordination device set, and the distributed control parameters corresponding to the multiple state coordination device sets are calculated.

[0066] In this embodiment, when performing host polling allocation analysis for multiple state collaborative device sets, the failure occurrence and coordination of each state collaborative device set are specifically considered, so as to determine the number of polling control hosts for each state collaborative device set based on this information. In this process, the global failure probability parameter is combined with the failure coordination parameter to calculate the distribution control parameter, which is used to describe the risk level and failure fluctuation characteristics of the device cluster when a failure occurs. The significance is that if the global failure probability is high and the failure coordination is low, the distribution control parameter of the device cluster is large, which means that the cluster has a higher risk when a failure occurs, and because of the low failure coordination, more hosts are required for polling monitoring, so as to avoid the failure data of multiple devices that may occur in the short term during the monitoring of a single host; on the contrary, if the global failure probability is low and the failure coordination is high, the distribution control parameter of the cluster is small, which means that fewer hosts can be used for polling.

[0067] Determine the distributed control range according to the distributed control parameters corresponding to multiple state collaborative device sets, obtain the number of host devices, divide the distributed control range according to the number of host devices to obtain multiple local control ranges, and determine the number of reference hosts for each local control range.

[0068] In this embodiment, according to the distributed control parameters of multiple device clusters, a reasonable control range is further determined, and these device clusters are allocated to each host device according to the number of existing hosts. In this process, the distributed control parameters are first counted to generate a distributed control range, and then the distributed control range is divided into equal numbers according to the number of existing host devices to obtain a local control range with the same number of existing host devices, and then the number of reference hosts for each local control range can be further determined according to the number of existing host devices. Specifically, the local control range can be sorted according to the size of the distributed control parameters specifically involved. For example, if the corresponding distributed control parameter belongs to the local control range of the highest level, the number of reference hosts can be the maximum number, and then it decreases in sequence. If the corresponding distributed control parameter belongs to the local control range of the lowest level, only one host device is allocated. In this way, the number of reference hosts for the local control range is preliminarily allocated. It is worth noting that in this case, considering that there is no restriction and priority distinction between host devices, those skilled in the art can also reasonably set the number of reference hosts for different local control ranges in combination with different host devices. This is only explained as an example.

[0069] According to the distributed control parameters corresponding to the multiple state collaborative device sets, the local control range corresponding to each state collaborative device set is determined respectively, and the number of reference hosts associated with the local control range corresponding to the state collaborative device set is used as the number of polling control hosts of the state collaborative device set.

[0070] In this embodiment, after determining the number of reference hosts for each local control range, the distributed control parameters corresponding to the state collaborative device set can be matched with multiple local control ranges to determine the local control range corresponding to each state collaborative device set, and then the number of reference hosts associated with the local control range is recorded as the number of polling control hosts for the state collaborative device set, thereby achieving efficient allocation and monitoring of device clusters. The allocation principle of the polling control host is specifically to ensure that high-risk and high-failure probability device clusters can obtain more monitoring resources while avoiding waste of resources.

[0071] In one of the embodiments, for the above step S5, multiple state collaborative device sets are segmented according to the number of polling control hosts of the state collaborative device set and the local monitoring sorting reference list to obtain multiple target collaborative device sets, and the target polling priority parameter and the global monitoring sorting reference list of each target collaborative device set are determined, specifically including:

[0072] According to the number of polling control hosts of the state collaborative device set, multiple target collaborative device sets for each state collaborative device set are constructed. After randomly sorting the multiple target collaborative device sets for each state collaborative device set, multiple slave devices in the state collaborative device set are allocated to the multiple sorted target collaborative device sets in sequence according to the monitoring sorting reference list. After multiple cycles, the state collaborative device set is split until the division is completed, and multiple target collaborative device sets corresponding to each state collaborative device set are obtained.

[0073] In this embodiment, the target collaborative device set is used to reallocate multiple slave devices in the state collaborative device set. For multiple empty target collaborative device sets that have not been allocated devices, they are first randomly sorted. For example, if a certain state collaborative device set corresponds to three target collaborative device sets, they are randomly sorted as first, second, and third. For multiple slave devices in the state collaborative device set, they are allocated to the sorted multiple target collaborative device sets in turn according to the monitoring priorities of different slave devices represented by the monitoring sorting reference list, and multiple cycles are performed. For example, the slave device ranked fourth in the list continues to be allocated from the target collaborative device set ranked first until the state collaborative device set is split. In this way, multiple state collaborative device sets are split to obtain target collaborative device sets. Because different target collaborative device sets need to be allocated to different host devices for monitoring, this method of sorting according to monitoring importance can avoid slave devices with higher importance from being allocated to the same host device for management.

[0074] Then, according to the order determined by randomly sorting multiple target collaborative device sets in each state collaborative device set, the target polling priority parameter of each target collaborative device set is generated, and the global monitoring sorting reference list of the target collaborative device set is generated according to the priority parameters of multiple slave devices in the target collaborative device set.

[0075] In this embodiment, the target polling priority parameters of each target collaborative device set are determined according to the sorting of the state collaborative device sets. In this process, the local polling priority parameters of the state collaborative device set and at least one adjacent state collaborative device set of the state collaborative device set can be determined first, and then the target polling priority parameters of multiple target collaborative device sets under the state collaborative device set can be restricted according to the local polling priority parameters of at least one adjacent state collaborative device set of the state collaborative device set. Exemplarily, for a state collaborative device set with a local polling priority parameter of 2, after being segmented to obtain three target collaborative device sets, the three target collaborative device sets can be divided into priorities of 2.1, 2.2 and 2.3 in the order determined by random sorting, so that the three target collaborative device sets will not exceed the local polling priority parameter of at least one adjacent state collaborative device set of the state collaborative device set, that is, the priority is lower than the previous state collaborative device set and higher than the next state collaborative device set. For multiple slave devices in each target collaborative device set, the priority parameter of the slave device is determined according to the previously determined local monitoring sorting reference list. The priority parameter of the slave device can be calculated according to the monitoring importance parameter of the slave device and the local fault probability parameter. The monitoring importance parameter of the slave device is weightedly optimized by the local fault probability parameter to generate the priority parameter of the slave device. Then, the multiple slave devices in the target collaborative device set are further sorted according to the priority parameter to generate a global monitoring sorting reference list of the target collaborative device set, which is used to represent the polling monitoring priority relationship of different slave devices in the device set obtained after segmentation.

[0076] In the above steps, multiple slave devices have been analyzed according to the fault data, monitoring importance, priority parameters, etc. of the slave devices, and multiple target collaborative device sets have been generated. Then, based on the target polling priority parameters and the global monitoring sorting reference list, as well as the existing host device situation, a polling control strategy for the slave device is generated. Specifically, in the above step S6, the polling control strategy for the slave device is determined according to the target polling priority parameters and the global monitoring sorting reference list, including the following contents:

[0077] An allocation priority list for multiple target collaborative device sets is generated according to the target polling priority parameter. After determining the allocation order of multiple host devices, the first target collaborative device set is allocated to each host device according to the allocation order of the host device and the allocation priority list for multiple target collaborative device sets.

[0078] In this embodiment, multiple target collaborative device sets are preliminarily sorted by target polling priority parameters to generate an allocation priority list, and then multiple target collaborative device sets in the allocation priority list are allocated to multiple host devices, and then the allocation order of the allocated host devices is determined, which can be determined randomly or divided according to the actual performance of the host device, for example, high-performance host devices are allocated first to ensure that the target collaborative device set that most needs polling monitoring is allocated to the high-performance host device. Assuming that each host device is unallocated, the first target collaborative device set is allocated to each host device based on the allocation order of the host device and the allocation priority list of multiple target collaborative device sets. For example, if there are five host devices, the first five target collaborative device sets in the allocation priority list are first allocated to multiple host devices in the corresponding order, and for the remaining multiple target collaborative device sets in the allocation priority list, further allocation based on load limit and collaborative interference limit is performed to avoid some host devices from bearing too heavy polling tasks, and to avoid collaborative interference between host devices, for example, multiple target collaborative device sets belonging to the same state collaborative device set are allocated to the same host device.

[0079] In this process, for the target collaborative device set with the largest target polling priority parameter in the allocation priority list, the target load quantity of each host device is determined, which can be specifically the number of slave devices corresponding to the target collaborative device set that has been allocated to the host device. Exemplarily, the PLC power communication is composed of a host and several slaves. Each host is connected to a maximum of 500 slaves. The host uses a polling mode to collect slave information. The one-way communication between the host and the slave takes about 200ms; the polling time is about 400ms, and the polling time increases by 400ms for each slave device. Therefore, the current load characteristics of the host device are roughly characterized based on the number of slave devices. It is worth noting that in some scenarios, some slave devices may involve complex data collection, and the actual data transmission volume is large, which results in more polling time. For this situation, the load characteristics can be further determined based on the actual time consumption data. In this embodiment, only the case where multiple slave devices have similar characteristics is used as an example.

[0080] Then, multiple host devices without coordinated interference are screened out, and the target coordinated device set with the largest target polling priority parameter is assigned to the host device with the lowest target load among the multiple host devices without coordinated interference. Among them, for multiple host devices without coordinated interference, if there is no coordinated device set in the same state as the target coordinated device set to be assigned among the multiple target coordinated device sets currently assigned to the host device, it is considered that there is no coordinated interference between the host device and the target coordinated device set to be assigned.

[0081] The allocation of the target collaborative device set first in the allocation priority list is completed in the above manner, and the target load quantity of each host device is updated after each allocation is completed until the allocation of the remaining multiple target collaborative device sets in the allocation priority list is completed. Combined with the global monitoring sorting reference list of each target collaborative device set, a polling control strategy for the slave device is generated.

[0082] In the above manner, multiple target collaborative device sets belonging to the same state collaborative device set are assigned to different host devices. As for the process of polling control of multiple slave devices based on the polling control strategy, the specific polling rules can be determined according to the target polling priority parameters of the multiple target collaborative device sets under each host device. For example, for multiple target collaborative devices under any host device, the polling order is, according to the target polling priority parameters of the target collaborative device set, poll the first slave device of each target collaborative device set, and after completing a round of data collection for each target collaborative device set, poll the second slave device, the third slave device, etc. of each target collaborative device set. In this way, in each shorter time period, at least one slave device belonging to a collaborative device set in different states is polled to collect data as much as possible, and some slave devices with higher fault frequency or greater monitoring importance are given priority to collect data. For some target collaborative device sets with more discrete fault coordination, data may be collected by multiple host devices. This method ensures that each type of slave device with different state collaborative changes has part of the data collected in multiple time periods, so that during the polling process, the power system can have a more comprehensive grasp of the actual situation of different types of slave devices, ensuring that high-priority devices can be monitored in a timely manner, while avoiding resource waste or efficiency reduction due to the collaborative characteristics between slave devices. This polling control strategy can greatly improve the monitoring efficiency of the system, while maintaining the rational use of resources and reducing system bottlenecks and monitoring gaps.

[0083] See also Figure 2 The embodiment of the present invention also provides a PLC power communication automatic control cabinet multi-device polling control system, which is specifically used to implement the above-mentioned PLC power communication automatic control cabinet multi-device polling control method, and the system includes:

[0084] The device state fluctuation analysis module is used to obtain the historical state collection data corresponding to multiple slave devices, extract the state fluctuation sequence of the slave devices from the historical state collection data, and perform state fluctuation collaborative clustering on multiple slave devices based on the state fluctuation sequence to obtain multiple state collaborative device sets;

[0085] A local fault analysis module is used to extract historical fault data corresponding to multiple slave devices in each state collaborative device set, extract local fault probability parameters and fault state sequences of each slave device from multiple groups of historical fault data, and generate global fault probability parameters and fault coordination parameters of the state collaborative device set based on the local fault probability parameters and fault state sequences;

[0086] A fault discrete analysis module is used to determine the local polling priority parameter of each state collaborative device set based on the global fault probability parameter of the state collaborative device set, perform host polling allocation analysis on multiple state collaborative device sets according to the global fault probability parameter and the fault coordination parameter, and determine the number of polling control hosts for each state collaborative device set;

[0087] A local monitoring sequence analysis module is used to obtain monitoring importance parameters of multiple slave devices, and to perform monitoring sorting on multiple slave devices in each state collaborative device set according to the monitoring importance parameters and local failure probability parameters, so as to obtain a local monitoring sorting reference list in each state collaborative device set;

[0088] A collaborative segmentation module is used to segment multiple state collaborative device sets according to the number of polling control hosts of the state collaborative device sets and the local monitoring sorting reference list to obtain multiple target collaborative device sets, and determine the target polling priority parameter and the global monitoring sorting reference list of each target collaborative device set;

[0089] The polling control optimization module is used to determine the polling control strategy for the slave device according to the target polling priority parameter and the global monitoring sorting reference list, and perform polling control on multiple slave devices based on the polling control strategy.

[0090] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A PLC power communication automatic control cabinet multi-device polling control method, characterized in that: include: Acquire historical state collection data corresponding to a plurality of slave devices respectively, extract a state fluctuation sequence of the slave devices from the historical state collection data, perform state fluctuation collaborative clustering on the plurality of slave devices based on the state fluctuation sequence, and obtain a plurality of state collaborative device sets; Extract historical fault data corresponding to multiple slave devices in each state collaborative device set, extract local fault probability parameters and fault state sequences of each slave device from multiple groups of historical fault data, and generate global fault probability parameters and fault coordination parameters of the state collaborative device set according to the local fault probability parameters and fault state sequences; Determine the local polling priority parameter of each state collaborative device set based on the global fault probability parameter of the state collaborative device set, perform host polling allocation analysis on multiple state collaborative device sets according to the global fault probability parameter and the fault coordination parameter, and determine the number of polling control hosts for each state collaborative device set; Acquire monitoring importance parameters of multiple slave devices, perform monitoring sorting on multiple slave devices in each state collaborative device set according to the monitoring importance parameters and local failure probability parameters, and obtain a local monitoring sorting reference list in each state collaborative device set; According to the number of polling control hosts of the state collaborative device set and the local monitoring sorting reference list, multiple state collaborative device sets are segmented to obtain multiple target collaborative device sets, including constructing multiple target collaborative device sets of each state collaborative device set according to the number of polling control hosts of the state collaborative device set, randomly sorting the multiple target collaborative device sets of each state collaborative device set, and then sequentially allocating multiple slave devices in the state collaborative device set to the multiple sorted target collaborative device sets according to the monitoring sorting reference list, performing multiple cycles until the segmentation of the state collaborative device set is completed, and obtaining multiple target collaborative device sets corresponding to each state collaborative device set; Determining a target polling priority parameter and a global monitoring sorting reference list for each target collaborative device set, including generating a target polling priority parameter for each target collaborative device set according to an order determined by randomly sorting multiple target collaborative device sets of each state collaborative device set, and generating a global monitoring sorting reference list for the target collaborative device set according to priority parameters of multiple slave devices in the target collaborative device set; The priority parameter of the slave device is calculated according to the monitoring importance parameter of the slave device and the local fault probability parameter, and the priority parameter of the slave device is generated after weighted optimization of the monitoring importance parameter of the slave device by the local fault probability parameter; A polling control strategy for the slave device is determined according to the target polling priority parameter and the global monitoring sorting reference list, and polling control is performed on multiple slave devices based on the polling control strategy.

2. A PLC power communication automatic control cabinet multi-device polling control method according to claim 1, characterized in that: The global fault probability parameter and fault coordination parameter of the state coordination device set are generated according to the local fault probability parameter and the fault state sequence, including: Determine multiple fault events of the slave device according to the historical fault data of the slave device, determine the local fault probability parameter of the slave device according to the number of occurrences of the fault event, and take the mean value corresponding to the local fault probability parameters of multiple slave devices in the state collaborative device set as the global fault probability parameter of the state collaborative device set; The timestamp of each fault event of the slave device is extracted, and the fault state sequence of the slave device is generated according to the timestamp of the fault event. The fault state timing matrix of the state coordination device set is constructed according to multiple fault state sequences. The fault state timing matrix is ​​analyzed using the following formula to calculate the fault coordination parameters of the state coordination device set: In the formula, Indicates the fault coordination parameters of the state coordination device set, Indicates the number of slave devices in the state coordination device set, Indicates the fault overlap parameter, Indicates the status of the collaborative device set The slave device The timestamp of the fault event, Indicates A reference time window, Indicates the status of the collaborative device set The number of fault events for slave devices, represents the number of reference time windows, express For the fault overlap parameter, if the timestamp of the fault event belongs to the corresponding reference time window, it is recorded as 1, otherwise it is recorded as 0.

3. A PLC power communication automatic control cabinet multi-device polling control method according to claim 1, characterized in that: Perform host polling allocation analysis on multiple state collaborative device sets according to global fault probability parameters and fault coordination parameters to determine the number of polling control hosts for each state collaborative device set, including: The ratio between the global failure probability parameter and the failure coordination parameter of the state coordination device set is used as the distributed control parameter of the state coordination device set, and the distributed control parameters corresponding to the multiple state coordination device sets are calculated; Determine the distributed control range according to the distributed control parameters corresponding to multiple state collaborative device sets, obtain the number of host devices, divide the distributed control range according to the number of host devices to obtain multiple local control ranges, determine the number of reference hosts for each local control range, determine the local control range corresponding to each state collaborative device set according to the distributed control parameters corresponding to multiple state collaborative device sets, and use the number of reference hosts associated with the local control range corresponding to the state collaborative device set as the number of polling control hosts for the state collaborative device set.

4. A PLC power communication automatic control cabinet multi-device polling control method according to claim 1, characterized in that: Determine the polling control strategy for the slave device based on the target polling priority parameter and the global monitoring sorting reference list, including: generating an allocation priority list for multiple target collaborative device sets according to the target polling priority parameter, and after determining the allocation order of the multiple host devices, allocating the first target collaborative device set to each host device according to the allocation order of the host device and the allocation priority list for the multiple target collaborative device sets; Load limits and collaborative interference limits are allocated to the remaining multiple target collaborative device sets in the allocation priority list, wherein, for the target collaborative device set with the largest target polling priority parameter in the allocation priority list, the target load quantity of each current host device is determined, and after screening out multiple host devices without collaborative interference, the target collaborative device set with the largest target polling priority parameter is allocated to the host device with the lowest target load quantity among the multiple host devices without collaborative interference, and the target load quantity of each host device is updated after each allocation is completed until the allocation of the remaining multiple target collaborative device sets in the allocation priority list is completed, and a polling control strategy for the slave device is generated in combination with the global monitoring sorting reference list of each target collaborative device set; Among them, for multiple host devices without coordinated interference, if among the multiple target collaborative device sets currently assigned to the host device, there is no collaborative device set belonging to the same state as the target collaborative device set currently to be assigned, it is considered that there is no coordinated interference between the host device and the target collaborative device set currently to be assigned.

5. A PLC power communication automatic control cabinet multi-device polling control method according to claim 4, characterized in that: For each target collaborative device set, the target polling priority parameter also includes: Determine a local polling priority parameter of a state collaborative device set, determine at least one adjacent state collaborative device set of each state collaborative device set, and limit target polling priority parameters of multiple target collaborative device sets under the state collaborative device set according to the local polling priority parameter of at least one adjacent state collaborative device set of the state collaborative device set.

6. A PLC power communication automatic control cabinet multi-device polling control system, characterized in that: The system is used to implement a PLC power communication automatic control cabinet multi-device polling control method as described in any one of claims 1 to 5, comprising: The device state fluctuation analysis module is used to obtain the historical state collection data corresponding to multiple slave devices, extract the state fluctuation sequence of the slave devices from the historical state collection data, and perform state fluctuation collaborative clustering on multiple slave devices based on the state fluctuation sequence to obtain multiple state collaborative device sets; A local fault analysis module is used to extract historical fault data corresponding to multiple slave devices in each state collaborative device set, extract local fault probability parameters and fault state sequences of each slave device from multiple groups of historical fault data, and generate global fault probability parameters and fault coordination parameters of the state collaborative device set based on the local fault probability parameters and fault state sequences; A fault discrete analysis module is used to determine the local polling priority parameter of each state collaborative device set based on the global fault probability parameter of the state collaborative device set, perform host polling allocation analysis on multiple state collaborative device sets according to the global fault probability parameter and the fault coordination parameter, and determine the number of polling control hosts for each state collaborative device set; A local monitoring sequence analysis module is used to obtain monitoring importance parameters of multiple slave devices, and to perform monitoring sorting on multiple slave devices in each state collaborative device set according to the monitoring importance parameters and local failure probability parameters, so as to obtain a local monitoring sorting reference list in each state collaborative device set; A collaborative segmentation module is used to segment multiple state collaborative device sets according to the number of polling control hosts of the state collaborative device sets and the local monitoring sorting reference list to obtain multiple target collaborative device sets, and determine the target polling priority parameter and the global monitoring sorting reference list of each target collaborative device set; The polling control optimization module is used to determine the polling control strategy for the slave device according to the target polling priority parameter and the global monitoring sorting reference list, and perform polling control on multiple slave devices based on the polling control strategy.

Images