Cloud edge collaborative management and control system for electromechanical converter

Abstract

The invention relates to the technical field of electromechanical cloud management and control, in particular to an electromechanical converter cloud edge collaborative management and control system which comprises a cloud edge collaborative management and control center, an edge management module, a task division module, a dynamic distribution module, a distribution evaluation module, a collaborative management module and an execution feedback layer. According to the method, analysis is preliminarily carried out from the angle of historical data processing capability of the edge node so as to preferentially match the equipment end, and meanwhile, division processing is carried out according to the task quantity and the task type, so that the tasks submitted by the equipment end are reasonably distributed to the edge node and the cloud server, and the efficiency is improved. Analysis is deeply carried out from a single-point angle and an overall angle through an information feedback mode, that is, rationalized and targeted management is carried out on the edge nodes from the single-point angle, so that the resource dynamic allocation performance of the edge nodes is improved, collaborative management demand analysis is carried out on collaborative performance data from the overall angle, and the resource dynamic allocation performance of the edge nodes is improved. The collaborative management efficiency among the equipment end, the edge end and the cloud server is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electromechanical cloud control, and particularly to an electromechanical converter cloud-edge collaborative control system. Background Art

[0002] An electromechanical converter is a device that performs bidirectional conversion between mechanical energy and electrical energy, and realizes the transformation of energy forms through physical mechanisms such as electromagnetic induction and magnetic field coupling. Its core functions include: energy conversion: converting rotational mechanical energy into electrical energy (generator mode) or electrical energy into mechanical energy (motor mode); multi-port control: supporting multiple inputs and multiple outputs (such as dual electrical ports + dual mechanical ports), and adapting to complex scenarios such as hybrid power systems;

[0003] Cloud-edge collaborative control refers to realizing dynamic scheduling, task coordination and data intercommunication of cloud and edge resources through a unified management framework, aiming to optimize computing power allocation, reduce latency and improve the overall efficiency of the system. However, in the prior art, it is impossible to reasonably divide and manage the tasks submitted by the electromechanical converter, resulting in low data processing efficiency of the submitted tasks, and it is impossible to analyze the dynamic resource allocation performance of the edge nodes, resulting in reduced management efficiency of the edge nodes. At the same time, it is impossible to supervise the overall collaborative performance of the control system, which is not conducive to the collaborative management among the device side, the edge side and the cloud server;

[0004] In view of the above technical deficiencies, a solution is proposed herein. Summary of the Invention

[0005] The purpose of the present invention is to provide an electromechanical converter cloud-edge collaborative control system to solve the above-mentioned technical deficiencies. The present invention initially analyzes from the perspective of the historical data processing ability of the edge nodes to help preferentially match the device side, which is helpful to improve the management effect of the device side. At the same time, it performs reasonable division management analysis based on the tasks submitted by the device side, so as to perform division processing according to the task volume and task type, thereby helping to reasonably allocate the tasks submitted by the device side to the edge nodes and the cloud server, avoiding the problems of resource idle waste caused by traditional cloud-edge collaboration and long task queuing time caused by cloud-edge capacity differences. By means of information feedback, in-depth analysis is carried out from both single-point and overall perspectives, that is, from the single-point perspective, the task dynamic allocation ability of the edge nodes is rationally and targeted managed to improve the dynamic resource allocation performance of the edge nodes, and from the overall perspective, the collaborative management demand analysis of the collaborative performance data is carried out, so as to optimize the management of the entire control system according to the current collaborative management performance, so as to improve the collaborative management efficiency among the device side, the edge side and the cloud server.

[0006] The object of the present invention can be achieved by the following technical solutions: An electromechanical converter cloud-edge collaborative management and control system, including a cloud-edge collaborative management and control center, an edge management module, a task division module, a dynamic allocation module, an allocation evaluation module, a collaborative management module, and an execution feedback layer;

[0007] The cloud-edge collaborative management and control center is used to retrieve the historical data processing capacity value of the to-be-selected node, and send the historical data processing capacity value to the edge management module for edge management matching analysis to obtain the first edge node;

[0008] The task division module is used to perform division management analysis on the task volume corresponding to the task processing request submitted by the device side to obtain a cloud processing signal or a processing signal. When a processing signal is generated, the collected task information is discriminated and processed to obtain a progressive signal or a cloud analysis signal;

[0009] When a progressive signal is generated, the dynamic allocation module is used to collect the status performance data of the first edge node, and perform task dynamic allocation management analysis on the status performance data to obtain an edge processing signal or an alternative signal;

[0010] The allocation evaluation module is used to perform dynamic allocation management ability acquisition analysis on the task queuing duration of each collected edge node to obtain a reasonable signal or a management and regulation feedback signal. The collaborative management module is used to collect the collaborative performance data of the device side, the edge side, and the cloud server, and perform collaborative management requirement analysis on the collaborative performance data to obtain a normal signal or a management and control signal.

[0011] Preferably, the edge management matching analysis process is as follows: The operation period of the electromechanical converter is collected and set as the time threshold. The electromechanical converter is set as the device side, and the connectable edge nodes in the area where the device side is located within the time threshold are obtained, and the connectable edge nodes in the area where the device side is located are set as the to-be-selected nodes;

[0012] The historical data processing capacity values of each to-be-selected node within the time threshold are obtained. The historical data processing capacity value represents the ratio between the data task volume and the processing duration corresponding to the data task volume. At the same time, the total historical processing times of each to-be-selected node within the time threshold are obtained, and the data processing capacity mean value of each to-be-selected node is obtained based on the historical data processing capacity value and the historical processing total times;

[0013] The maximum value of the data processing capacity mean value is obtained, and the to-be-selected node corresponding to the maximum value of the data processing capacity mean value is set as the first edge node, and the device side is connected to the first edge node.

[0014] Preferably, the division management analysis process is as follows:

[0015] Obtain the task processing requests submitted by the device end to the first edge node within the time threshold, obtain the task volume corresponding to the task processing requests submitted by the device end, and perform comparison and analysis on the task volume corresponding to the task processing requests submitted by the device end to obtain a cloud processing signal or a processing signal;

[0016] When a processing signal is generated, obtain the task information of the task volume corresponding to the task processing requests submitted by the device end. The task information includes training task data and non-training task data, and perform discriminant processing on the task information to obtain a cloud analysis signal or a progressive signal.

[0017] Preferably, the task dynamic allocation management analysis process is as follows:

[0018] Obtain the status performance data of the first edge node within the time threshold. The status performance data represents the product value obtained by multiplying the resource occupancy depth by the corresponding value of the data processing capacity per unit time at present, and perform discriminant processing on the status performance data of the first edge node to obtain an edge processing signal or an alternative signal. When an alternative signal is generated, set the selected node with the maximum average data processing capacity after removing the first edge node from the to-be-selected nodes as the alternative edge node;

[0019] The resource occupancy depth represents the number of corresponding values of the CPU occupancy rate, GPU occupancy rate, and storage space occupancy rate of the first edge node that are greater than or equal to the preset threshold.

[0020] Preferably, the dynamic allocation management ability acquisition analysis process is as follows: Obtain the task queuing duration of each edge node within the time threshold, and perform comparison and analysis on the task queuing duration. If the task queuing duration is greater than the preset task queuing duration threshold, set the corresponding edge node as a high-load node. If the task queuing duration is less than or equal to the preset task queuing duration threshold, set the corresponding edge node as an idle-load node.

[0021] Preferably, obtain the number of high-load nodes and the number of idle-load nodes corresponding thereto, set the ratio between the number of high-load nodes and the number of idle-load nodes as the dynamic allocation management ability, and perform comparison and analysis on the dynamic allocation management ability. If the dynamic allocation management ability is less than the preset dynamic allocation management ability threshold, generate a reasonable signal. If the dynamic allocation management ability is greater than or equal to the preset dynamic allocation management ability threshold, generate a management adjustment feedback signal.

[0022] Preferably, the collaborative management requirement analysis process is as follows:

[0023] Obtain the collaborative performance data of the device side, edge side, and cloud server corresponding to the control system within the time threshold. The collaborative performance data includes the control delay rate of the edge node, the bandwidth occupancy rate of the cloud server, and the fault diagnosis accuracy rate. Compare and analyze the collaborative performance data to obtain the comparison result of the collaborative performance data. The comparison result includes qualified and unqualified. Obtain the occupancy ratio corresponding to the qualified comparison result of the collaborative performance data, and perform discriminant analysis on the occupancy ratio corresponding to the qualified comparison result of the collaborative performance data to obtain a normal signal or a control signal.

[0024] Preferably, qualified and unqualified: Compare and analyze each parameter in the collaborative performance data according to the set qualified standard and unqualified standard. If the qualified standard is met, the corresponding parameter is determined to be qualified. If the unqualified standard is met, the corresponding parameter is determined to be unqualified.

[0025] The beneficial effects of the present invention are as follows:

[0026] (1) The present invention initially analyzes from the perspective of the historical data processing ability of the edge node to give priority to matching the device side, which helps to improve the management effect of the device side. At the same time, based on the tasks submitted by the device side, reasonable division management analysis is carried out, so as to divide and process according to the task volume and task type. Furthermore, it helps to reasonably allocate the tasks submitted by the device side to the edge node and the cloud server, avoiding the problems of resource idle waste caused by traditional cloud-edge collaboration and long task queuing time due to cloud-edge capacity differences.

[0027] (2) The present invention deeply analyzes from two perspectives of single point and overall through the way of information feedback. That is, from the single point perspective, the task dynamic allocation ability of the edge node is rationally and specifically managed to improve the resource dynamic allocation performance of the edge node. From the overall perspective, the collaborative management demand analysis of the collaborative performance data is carried out, so as to manage and optimize the entire control system according to the current collaborative management performance, and improve the collaborative management efficiency between the device side, edge side, and cloud server. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention will be further described below with reference to the drawings;

[0029] Figure 1 is the system flow block diagram of the present invention;

[0030] Figure 2 is the local analysis reference diagram of the present invention;

[0031] Figure 3 is the local reference diagram of Embodiment 1 of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0032] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0033] As used herein, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments;

[0034] Embodiment 1:

[0035] Please refer to Figures 1 to 2 As shown, the present invention is an edge-cloud collaborative control system for an electromechanical converter, including an edge-cloud collaborative control center, an edge management module, a task division module, a dynamic allocation module, an allocation evaluation module, a collaborative management module, and an execution feedback layer. The edge-cloud collaborative control center is unidirectionally communicatively connected to both the edge management module and the execution feedback layer. The edge-cloud collaborative control center is bidirectionally communicatively connected to both the allocation evaluation module and the collaborative management module. The edge management module is unidirectionally communicatively connected to the task division module. The task division module is unidirectionally communicatively connected to the dynamic allocation module. The dynamic allocation module is unidirectionally communicatively connected to the execution feedback layer;

[0036] The edge-cloud collaborative control center is used to retrieve the historical data processing capacity value of the to-be-selected node and send the historical data processing capacity value to the edge management module for edge management matching analysis to obtain the first edge node. The specific edge management matching analysis process is as follows:

[0037] Collect the operation period of the electromechanical converter and set it as the time threshold. Set the electromechanical converter as the device end, obtain the connectable edge nodes in the area where the device end is located within the time threshold, and set the connectable edge nodes in the area where the device end is located as the to-be-selected nodes;

[0038] Obtain the historical data processing capacity values of each to-be-selected node within the time threshold. The historical data processing capacity value represents the ratio between the data task volume and the processing duration corresponding to the data task volume. At the same time, obtain the total number of historical processes of each to-be-selected node within the time threshold, and obtain the data processing capacity mean value of each to-be-selected node based on the historical data processing capacity value and the total number of historical processes;

[0039] Obtain the maximum value of the average data processing capacity, and set the to-be-selected node corresponding to the maximum value of the average data processing capacity as the first edge node, and connect the device side to the first edge node;

[0040] The task division module is used to perform division management analysis on the task volume corresponding to the task processing request submitted by the device side, so as to perform division processing according to the task volume and task type, so as to improve the task processing efficiency. The specific division management analysis process is as follows:

[0041] Obtain the task processing requests submitted by the device side to the first edge node within the time threshold, obtain the task volume corresponding to the task processing request submitted by the device side, and perform comparison and analysis on the task volume corresponding to the task processing request submitted by the device side. If the task volume corresponding to the task processing request submitted by the device side is greater than the preset threshold, generate a cloud processing signal, and the feedback layer responds to the cloud processing signal and submits the task processing request submitted by the device side to the cloud server for processing;

[0042] If the task volume corresponding to the task processing request submitted by the device side is less than or equal to the preset threshold, generate a processing signal. When the processing signal is generated, obtain the task information of the task volume corresponding to the task processing request submitted by the device side. The task information includes training task data and non-training task data, and perform discrimination processing on the task information:

[0043] If the task information is training task data, generate a cloud analysis signal, and the feedback layer responds to the cloud analysis signal and submits the task processing request submitted by the device side to the cloud server for processing

[0044] If the task information is non-training task data, generate a progressive signal;

[0045] Training task data refers to task data used for training models (such as fault diagnosis models, fault feature recognition models, etc.);

[0046] Training task data refers to task data not used for training models;

[0047] When the progressive signal is generated, the dynamic allocation module is used to collect the status performance data of the first edge node, and perform task dynamic allocation management analysis on the status performance data, which helps to reasonably allocate the tasks submitted by the device side to the edge node and the cloud server, avoiding the problems of resource idle waste and long task queuing time caused by the traditional cloud-edge collaboration. The specific task dynamic allocation management analysis process is as follows:

[0048] The state performance data of the first edge node within the time threshold is obtained. The state performance data represents the product value obtained by multiplying the resource occupancy depth by the corresponding value of the data processing capacity per unit time. The state performance data of the first edge node is discriminated. If the state performance data of the first edge node is less than the preset threshold, an edge processing signal is generated. When the edge processing signal is generated, the first edge node processes the task processing request submitted by the device side. If the state performance data of the first edge node is greater than or equal to the preset threshold, an alternative signal is generated. When the alternative signal is generated, the selected node with the maximum average data processing capacity after removing the first edge node from the nodes to be selected is set as the alternative edge node, and the task processing request submitted by the device side is submitted to the alternative edge node for processing. This helps to reasonably allocate the tasks submitted by the device side to the edge nodes and the cloud server, avoiding the problems of resource idle waste caused by traditional cloud-edge collaboration and long task queuing time due to the difference in cloud-edge capabilities;

[0049] The resource occupancy depth represents the number of corresponding values of the CPU occupancy rate, GPU occupancy rate, storage space occupancy rate, etc. of the first edge node that are greater than or equal to the preset threshold;

[0050] The data processing capacity value per unit time represents the amount of data tasks per unit time of the current first edge node.

[0051] Embodiment 2:

[0052] The allocation evaluation module is used to obtain and analyze the dynamic allocation management ability of the task queuing duration of each edge node collected, so as to rationally and specifically manage the task dynamic allocation ability of the edge node, and improve the resource dynamic allocation performance of the edge node. The specific process of obtaining and analyzing the dynamic allocation management ability is as follows:

[0053] The task queuing duration of each edge node within the time threshold is obtained, and the task queuing duration is compared and analyzed. If the task queuing duration is greater than the preset task queuing duration threshold, the corresponding edge node is set as a high-load node. If the task queuing duration is less than or equal to the preset task queuing duration threshold, the corresponding edge node is set as an idle-load node;

[0054] Obtain the number of high-load nodes and the number of idle load nodes, set the ratio between the number of high-load nodes and the number of idle load nodes as the dynamic allocation management ability, and perform a comparison and analysis on the dynamic allocation management ability. If the dynamic allocation management ability is less than the preset dynamic allocation management ability threshold, generate a reasonable signal. If the dynamic allocation management ability is greater than or equal to the preset dynamic allocation management ability threshold, generate a management and adjustment feedback signal, and send the reasonable signal or the management and adjustment feedback signal to the execution feedback layer. After receiving the reasonable signal or the management and adjustment feedback signal, the execution feedback layer immediately performs the preset warning operation corresponding to the reasonable signal or the management and adjustment feedback signal, so as to rationally and specifically manage the task dynamic allocation ability of the edge nodes and improve the dynamic allocation performance of the resources of the edge nodes;

[0055] The collaborative management module is used to collect the collaborative performance data of the device side, the edge side, and the cloud server, and perform an analysis on the collaborative management requirements of the collaborative performance data, so as to manage and optimize the entire control system according to the current collaborative management performance, and improve the collaborative management efficiency between the device side, the edge side, and the cloud server. The specific process of the collaborative management requirement analysis is as follows:

[0056] Obtain the collaborative performance data of the control system corresponding to the device side, the edge side, and the cloud server within the time threshold. The collaborative performance data includes the control delay rate of the edge nodes, the bandwidth occupancy rate of the cloud server, the fault diagnosis accuracy rate, etc., and perform a comparison and analysis on the collaborative performance data to obtain the comparison result of the collaborative performance data. The comparison result includes qualified and unqualified. Obtain the occupancy ratio corresponding to the comparison result of the collaborative performance data being qualified, and perform a discriminant analysis on the occupancy ratio corresponding to the comparison result of the collaborative performance data being qualified:

[0057] If the occupancy ratio corresponding to the comparison result of the collaborative performance data being qualified is greater than or equal to the preset threshold, generate a normal signal;

[0058] If the occupancy ratio corresponding to the comparison result of the collaborative performance data being qualified is less than the preset threshold, generate a management and control signal, and send the normal signal or the management and control signal to the execution feedback layer. After receiving the normal signal or the management and control signal, the execution feedback layer immediately performs the preset warning operation corresponding to the normal signal or the management and control signal, so as to perform collaborative management and control on the current device side, edge side, and cloud server, improve the collaborative management efficiency between the device side, edge side, and cloud server, and improve the control stability and efficiency of the entire control system;

[0059] In the embodiment of the present invention, the edge side represents the edge nodes;

[0060] Qualified and unqualified: Each parameter in the collaborative performance data is compared and analyzed according to the set qualified standard and unqualified standard. If the qualified standard is met, the corresponding parameter is determined to be qualified; if the unqualified standard is met, the corresponding parameter is determined to be unqualified.

[0061] For example: the average control delay rate of the edge node. The qualified standard is less than or equal to 5ms, and the unqualified standard is greater than 5ms. If the average control delay rate of the edge node is less than or equal to 5ms and meets the qualified standard, it is determined that the average control delay rate of the edge node is qualified; if the average control delay rate of the edge node is greater than 5ms and meets the unqualified standard, it is determined that the average control delay rate of the edge node is unqualified.

[0062] In summary, the present invention initially analyzes from the perspective of the historical data processing ability of the edge node to facilitate the priority matching of the device side, which helps to improve the management effect of the device side. At the same time, a reasonable division management analysis is carried out based on the tasks submitted by the device side, so as to carry out division processing according to the task volume and task type, and then helps to reasonably allocate the tasks submitted by the device side to the edge node and the cloud server, avoiding the problems of resource idle waste caused by traditional cloud-edge collaboration and long task queuing time caused by the cloud-edge ability difference. Through the way of information feedback, in-depth analysis is carried out from both the single point and the overall perspectives, that is, the task dynamic allocation ability of the edge node is rationally and targeted managed from the single point perspective to improve the resource dynamic allocation performance of the edge node, and the collaborative management requirement analysis of the collaborative performance data is carried out from the overall perspective to optimize the management of the entire control system according to the current collaborative management performance, so as to improve the collaborative management efficiency among the device side, the edge side and the cloud server.

[0063] The setting of the threshold value is for the convenience of comparison. Regarding the size of the threshold value, it depends on the amount of sample data and the base quantity set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantified value.

[0064] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. An edge-cloud collaborative control system for an electromechanical transducer, characterized in that, It includes a cloud-edge collaborative control center, an edge management module, a task division module, a dynamic allocation module, an allocation evaluation module, a collaborative management module, and an execution feedback layer; The cloud-edge collaborative control center is used to retrieve the historical data processing capacity value of the to-be-selected node, and send the historical data processing capacity value to the edge management module for edge management matching analysis to obtain the first edge node; The task division module is used to perform division management analysis on the task volume corresponding to the task processing request submitted by the device end, to obtain a cloud processing signal or a processing signal. When a processing signal is generated, it discriminates and processes the collected task information to obtain a progressive signal or a cloud analysis signal; When a progressive signal is generated, the dynamic allocation module is used to collect the status performance data of the first edge node, and perform task dynamic allocation management analysis on the status performance data to obtain an edge processing signal or an alternative signal; The allocation evaluation module is used to perform dynamic allocation management ability acquisition analysis on the task queuing duration of each collected edge node to obtain a reasonable signal or a management adjustment feedback signal. The collaborative management module is used to collect the collaborative performance data of the device end, the edge end, and the cloud server, and perform collaborative management requirement analysis on the collaborative performance data to obtain a normal signal or a control signal.

2. The cloud-edge collaborative management and control system of an electromechanical transducer according to claim 1, characterized in that The edge management matching analysis process is as follows: The operation period of the electromechanical converter is collected and set as the time threshold. The electromechanical converter is set as the device end, the connectable edge nodes in the area where the device end is located within the time threshold are obtained, and the connectable edge nodes in the area where the device end is located are set as the to-be-selected nodes; The historical data processing capacity values of each to-be-selected node within the time threshold are obtained. The historical data processing capacity value represents the ratio between the data task volume and the processing duration corresponding to the data task volume. At the same time, the total historical processing times of each to-be-selected node within the time threshold are obtained, and the data processing capacity mean value of each to-be-selected node is obtained based on the historical data processing capacity value and the historical processing total times; The maximum value of the data processing capacity mean value is obtained, and the to-be-selected node corresponding to the maximum value of the data processing capacity mean value is set as the first edge node, and the device end is connected to the first edge node.

3. The cloud-edge collaborative management and control system of an electromechanical converter according to claim 1, characterized in that, The division management analysis process is as follows: The task processing request submitted by the device end to the first edge node within the time threshold is obtained, the task volume corresponding to the task processing request submitted by the device end is obtained, and the task volume corresponding to the task processing request submitted by the device end is compared and analyzed to obtain a cloud processing signal or a processing signal; When a processing signal is generated, the task information of the task volume corresponding to the task processing request submitted by the device end is obtained. The task information includes training task data and non-training task data, and the task information is discriminated and processed to obtain a cloud analysis signal or a progressive signal.

4. The cloud-edge collaborative management and control system of an electromechanical converter according to claim 1, characterized in that, The task dynamic allocation management analysis process is as follows: Obtain the status performance data of the first edge node within the time threshold. The status performance data represents the product value obtained by multiplying the resource occupancy depth by the corresponding value of the data processing capacity per unit time at present, and perform discrimination processing on the status performance data of the first edge node to obtain an edge processing signal or an alternative signal. When generating an alternative signal, set the selected node corresponding to the maximum value of the average data processing capacity after removing the first edge node from the nodes to be selected as the alternative edge node; The resource occupancy depth represents the number of corresponding values of the CPU occupancy rate, GPU occupancy rate, and storage space occupancy rate of the first edge node that are greater than or equal to the preset threshold.

5. The cloud-edge collaborative control system of an electromechanical converter according to claim 1, wherein, The process of obtaining and analyzing the dynamic allocation management ability is as follows: Obtain the task queuing duration of each edge node within the time threshold, and perform comparison and analysis on the task queuing duration. If the task queuing duration is greater than the preset task queuing duration threshold, set the corresponding edge node as a high-load node. If the task queuing duration is less than or equal to the preset task queuing duration threshold, set the corresponding edge node as an idle-load node.

6. The cloud-edge collaborative management and control system of an electromechanical transducer according to claim 5, characterized in that, Obtain the number corresponding to the high-load nodes and the number corresponding to the idle-load nodes, set the ratio between the number corresponding to the high-load nodes and the number corresponding to the idle-load nodes as the dynamic allocation management ability, and perform comparison and analysis on the dynamic allocation management ability. If the dynamic allocation management ability is less than the preset dynamic allocation management ability threshold, generate a reasonable signal. If the dynamic allocation management ability is greater than or equal to the preset dynamic allocation management ability threshold, generate a management and adjustment feedback signal.

7. The cloud-edge collaborative control system of an electromechanical converter according to claim 1, characterized in that The process of analyzing the collaborative management requirements is as follows: Obtain the collaborative performance data of the control systems corresponding to the device side, edge side, and cloud server within the time threshold. The collaborative performance data includes the control delay rate of the edge node, the bandwidth occupancy rate of the cloud server, and the fault diagnosis accuracy rate, and perform comparison and analysis on the collaborative performance data to obtain the comparison result of the collaborative performance data. The comparison result includes qualified and unqualified. Obtain the occupancy ratio corresponding to the comparison result of the collaborative performance data being qualified, and perform discrimination analysis on the occupancy ratio corresponding to the comparison result of the collaborative performance data being qualified to obtain a normal signal or a management and control signal.

8. The cloud-edge collaborative control system of an electromechanical transducer according to claim 7, characterized in that, Qualified and unqualified: Compare and analyze each parameter in the collaborative performance data according to the set qualified standard and unqualified standard degree. If the qualified standard is met, determine the corresponding parameter as qualified. If the unqualified standard is met, determine the corresponding parameter as unqualified.

Images