Device temperature patrol method and system fusing short-time load prediction

Abstract

The present application relates to the technical field of equipment inspection and task scheduling, and provides an equipment temperature inspection method and system fusing short-time load prediction, which comprises the following steps: generating a load prediction curve in a future preset time length through a load prediction algorithm; outputting an object sequence in a unified format according to the load prediction curve; setting a TopN time window parameter to obtain a specific TopN time window entity; drawing the object sequence from the TopN time window entity to obtain a plurality of time window sequences, and obtaining a final time window sequence according to the time window sequences; dynamically generating a task execution strategy according to the final time window sequence, and binding the task execution strategy with a specific equipment temperature inspection task; and executing the equipment temperature inspection task according to the task execution strategy to realize tracking of load operation. The present application uses load prediction results to guide equipment temperature inspection, realizes tracking and monitoring of equipment temperature changes, improves monitoring accuracy, and guarantees safe operation of equipment.

Description

Technical Field

[0001] This invention relates to the fields of equipment inspection technology and task scheduling technology, and in particular to a method and system for equipment temperature inspection that integrates short-term load prediction. Background Technology

[0002] During the operation of power grid equipment, the temperature may rise or fall due to various factors, thus affecting the normal operation of the equipment. Therefore, timely monitoring of equipment temperature changes is of great significance for maintaining safe operation and extending equipment life.

[0003] Existing equipment temperature monitoring technologies mainly include two methods: sensor monitoring and infrared thermal imaging monitoring. Sensor monitoring requires installing temperature sensors on multiple components of the equipment to monitor temperature changes, which is very costly to install and maintain in substations. Infrared thermal imaging monitoring can reflect the temperature distribution of the equipment through infrared thermal radiation images. Currently, it mainly uses infrared thermal imaging cameras or infrared thermometers. Infrared thermal imaging can be used for periodic inspections of equipment, but the operating conditions of the equipment cannot be consistently matched with the inspection cycle. For example, there are differences in equipment operating conditions between weekdays and weekends, and between rainy and dry seasons. It is easy to miss inspections when the equipment is overheating, making it impossible to accurately track equipment temperature changes. Infrared thermometers require manual on-site inspections, which are highly dependent on the subjective operation of the inspectors and the completeness of their adherence to inspection procedures. This results in low inspection efficiency and a large workload for personnel. Summary of the Invention

[0004] The purpose of this invention is to solve at least one technical problem in the background art and to provide a method and system for inspecting equipment temperature that integrates short-term load prediction.

[0005] To achieve the above objectives, the present invention provides a method for monitoring equipment temperature by integrating short-term load forecasting, comprising:

[0006] The load forecast curve for a preset future time period is generated using a load forecasting algorithm.

[0007] The load forecast curve is transformed and packaged to output a sequence of objects with a uniform format;

[0008] Set the TopN time window parameters to obtain the specific TopN time window entities;

[0009] The object sequence is passed through the TopN time window entity to obtain several discrete time window sequences within a preset future duration. The time window sequences are then aggregated and filtered to obtain the final time window sequence.

[0010] Based on the obtained final time window sequence, a task execution strategy is dynamically generated, and this task execution strategy is bound to a specific equipment temperature inspection task;

[0011] The equipment temperature monitoring task is executed according to the bound task execution strategy to achieve tracking of load operation.

[0012] According to one aspect of the present invention, the generation of the load forecast curve for a future preset time period by the load forecasting algorithm is as follows:

[0013] Based on the fuzzy neural network load forecasting algorithm, the short-term parameter of the algorithm is set to 24 hours, the trigger time is set to a fixed time, and after the algorithm service runs, it outputs the load forecast curve for the next 24 hours at a fixed time every day.

[0014] According to one aspect of the present invention, the transformation and packaging of the load curve to output a uniformly formatted object sequence is as follows:

[0015] The load forecast curve is read by the result wrapper and the format is converted and wrapped to generate a unified time series object sequence. Each object in the object sequence contains at least two attributes: load and time. At the same time, the business tag to which the object sequence belongs is wrapped into the header of the object sequence.

[0016] According to one aspect of the present invention, the step of setting the TopN time window parameters to obtain the specific TopN time window entity is as follows:

[0017] Set the coordinates of the bottom frame, window height, and window width of the TopN time windows to obtain the specific TopN time window entities.

[0018] According to one aspect of the present invention, the final time window sequence obtained by aggregating and filtering the time window sequence is as follows:

[0019] The time window sequences are aggregated and filtered using a time window filter to obtain the final effective time window sequence.

[0020] According to one aspect of the present invention, the task execution strategy includes: task trigger time, task duration, and task cycle.

[0021] Furthermore, to achieve the above objectives, the present invention also provides an equipment temperature monitoring system integrating short-term load forecasting, comprising:

[0022] The algorithm model expansion dock management module loads the load prediction algorithm model and configures the necessary parameters. After the algorithm starts, it outputs the load prediction curve in real time.

[0023] The result wrapper in the algorithm model expansion dock management module reads the load forecast curve, packages the load forecast curve into an object sequence in a unified time series format, marks it with a business tag, and then sends the object sequence to the trigger of the algorithm model expansion dock management module.

[0024] The trigger distributes the object sequence to the time series preprocessing module registered on it based on different business tags;

[0025] The time series preprocessing module receives the object sequence, first reads the time window parameter configuration, and then extracts the valid time window sequence from the time window sequence through the internally set TopN time window picking unit according to the parameters. The time window sequence is further aggregated and filtered through the time window filter to obtain the final valid time window sequence, and then sent to the automatic scheduling module.

[0026] The patrol task construction module, in which the patrol task management unit first creates patrol tasks to be executed, and each new patrol task to be executed has an automatically generated policy container and is bound to it. The task execution policy depends on the specific policy loaded in the policy container.

[0027] The automatic scheduling module receives the time window sequence and hands it over to the internal time window conversion unit, which converts the time window data into the task execution strategy of the inspection task. At the same time, it reads the inspection task to be executed from the inspection task construction module, and sends the task execution strategy and inspection task to the task automatic orchestration unit. The task automatic orchestration unit dynamically binds the task execution strategy into the corresponding strategy container according to the inspection task information, and finally generates an executable inspection task and sends it to the inspection task execution module.

[0028] The patrol task execution module receives executable patrol tasks, puts the patrol tasks into the patrol task queue, the built-in task execution unit retrieves the patrol tasks from the patrol task queue, and automatically executes them according to the task execution policy bound in the policy container. The result of the patrol task execution is sent to the message push module.

[0029] The message push module provides a registration function to the business system and pushes the results of the inspection task to the business system.

[0030] Furthermore, to achieve the above objectives, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the device temperature inspection method as described above, which integrates short-term load prediction.

[0031] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the device temperature inspection method as described above, which integrates short-term load prediction.

[0032] According to the present invention, this invention addresses the problems of high installation and maintenance costs associated with sensor-based temperature monitoring in power grid equipment operation, and the tendency for missed inspections, large workload of manual operation, and low inspection efficiency when using infrared thermal imaging for periodic inspections. It provides the aforementioned equipment temperature inspection method that integrates short-term load forecasting. This method utilizes the advantages of infrared thermal imaging cameras—easy installation and maintenance, and a wide field of view—combined with a pre-trained short-term load forecasting algorithm model as a front-end module to accurately guide equipment temperature inspections. Furthermore, the provided automatic scheduling system enables adaptive adjustment of equipment temperature inspection tasks, thereby improving inspection efficiency, accuracy, and timeliness.

[0033] According to the present invention, the equipment temperature inspection method integrating short-term load forecasting is adopted. Based on the load forecast curve within a future time interval, a high-load time window is calculated, and the triggering time and execution cycle of the equipment temperature inspection task are dynamically planned according to the time window. This allows for timely tracking of equipment temperature changes, reducing the workload of manual inspection, and significantly improving inspection efficiency, accuracy, and timeliness.

[0034] According to the present invention, the equipment temperature inspection system integrating short-term load forecasting is actually an automatic scheduling system that implements the above-mentioned equipment temperature inspection method integrating short-term load forecasting and its automatic scheduling inspection tasks. This system can track the predicted state of external conditions in the future and adaptively schedule the inspection tasks to be executed according to the predicted state, thereby enhancing the coordination between the execution strategy and the execution tasks, reducing the resource overhead of maintenance work required due to changes in external conditions, and improving the system's flexibility, adaptability and application efficiency. Attached Figure Description

[0035] Figure 1 A flowchart illustrating a method for inspecting equipment temperature by incorporating short-term load forecasting according to an embodiment of the present invention is shown in the schematic diagram.

[0036] Figure 2 The schematic diagram illustrates the principle block diagram of an equipment temperature monitoring system that integrates short-term load prediction according to one embodiment of the present invention. Detailed Implementation

[0037] The invention will now be discussed with reference to exemplary embodiments. It should be understood that the described embodiments are merely intended to enable those skilled in the art to better understand and thus implement the invention, and are not intended to imply any limitation on the scope of the invention.

[0038] As used herein, the term "comprising" and its variations are to be interpreted as open-ended terms meaning "including but not limited to". The term "based on" is to be interpreted as "at least partially based on". The terms "one embodiment" and "an embodiment" are to be interpreted as "at least one embodiment".

[0039] Figure 1 This is a schematic flowchart illustrating a method for monitoring equipment temperature based on short-term load forecasting according to an embodiment of the present invention. Figure 1 As shown, in this embodiment, the equipment temperature monitoring method integrating short-term load forecasting includes:

[0040] Load forecast curves for a preset future time period are generated using different load forecasting algorithms;

[0041] The different load forecast curves generated by different load forecasting algorithms are transformed and packaged to output a sequence of objects with a unified format.

[0042] Set the TopN time window parameters to obtain the specific TopN time window entities;

[0043] The object sequence is passed through the TopN time window entity to obtain several discrete time window sequences within a preset future duration. The time window sequences are then aggregated and filtered to obtain the final time window sequence.

[0044] Based on the obtained final time window sequence, a task execution strategy is dynamically generated, and this task execution strategy is bound to a specific equipment temperature inspection task;

[0045] The equipment temperature monitoring task is executed according to the bound task execution strategy to achieve tracking of load operation.

[0046] Furthermore, according to one embodiment of the present invention, the load forecast curve for a future preset time period is generated by the load forecasting algorithm as follows:

[0047] The load forecasting algorithm based on fuzzy neural network (FNN) has a short-term parameter set to 24 hours and a trigger time set to a fixed time (e.g., 1:00 AM every day). After the algorithm service runs, it outputs the load forecast curve for the next 24 hours at 1:00 AM every day.

[0048] Furthermore, according to one embodiment of the present invention, the load curve is transformed and packaged, and the output object sequence with a unified format is:

[0049] The load forecast curve is read by the result wrapper and the format is converted and wrapped to generate a unified time series object sequence. Each object in the object sequence contains at least two attributes: load and time. At the same time, the business tag to which the object sequence belongs is wrapped into the header of the object sequence.

[0050] In this implementation, the service tag is used to mark the object sequence, enabling the trigger to accurately locate the corresponding time series preprocessing service based on the tag. For example, if the load forecast result is a 24-hour curve object sequence, and this curve is labeled "Main Equipment Temperature Inspection," then the corresponding main equipment time series preprocessing service generates an inspection time window T1 from the object sequence based on the main equipment's operating characteristics. If the curve is labeled "Oil Chromatography Inspection," then the corresponding oil chromatography time series preprocessing service generates an inspection time window T2 from the object sequence based on the oil chromatography's operating characteristics. The T1 and T2 time windows may not overlap.

[0051] Furthermore, according to one embodiment of the present invention, the TopN time window parameters are set to obtain the specific TopN time window entity as follows:

[0052] Set the coordinates of the bottom frame, window height, and window width of the TopN time windows to obtain the specific TopN time window entities.

[0053] Furthermore, according to one embodiment of the present invention, the final time window sequence is obtained by aggregating and filtering the time window sequence as follows:

[0054] The time window sequences are aggregated and filtered using a time window filter to obtain the final effective time window sequence.

[0055] Furthermore, according to one embodiment of the present invention, the task execution strategy includes: task trigger time, task duration, and task cycle.

[0056] According to the above-described scheme of the present invention, the above six steps of the equipment temperature inspection method integrating short-term load forecasting of the present invention are executed iteratively. The load forecasting algorithm can continuously learn and optimize during the iterative process. As time goes by, the prediction effect of the algorithm will become more and more accurate, and the execution of equipment temperature inspection will become more and more efficient.

[0057] According to the above-described solution of the present invention, the present invention addresses the problems of high installation and maintenance costs associated with sensor-based temperature monitoring in power grid equipment operation, and the tendency for missed inspections, large workload of manual operation, and low inspection efficiency when using infrared thermal imaging for periodic inspections. It provides the aforementioned equipment temperature inspection method that integrates short-term load forecasting. This method utilizes the advantages of infrared thermal imaging cameras—easy installation and maintenance, and a wide field of view—combined with a pre-trained short-term load forecasting algorithm model as a front-end module to accurately guide equipment temperature inspections. Furthermore, the provided automatic scheduling system enables adaptive adjustment of equipment temperature inspection tasks, thereby improving inspection efficiency, accuracy, and timeliness.

[0058] According to the above-described scheme of the present invention, the equipment temperature inspection method integrating short-term load forecasting of the present invention calculates the high load time window based on the load forecast curve within a future time interval, and dynamically plans the trigger time and execution cycle of the equipment temperature inspection task according to the time window, thereby tracking equipment temperature changes in a timely manner, reducing the workload of manual inspection, and significantly improving inspection efficiency, accuracy and timeliness.

[0059] Furthermore, to achieve the above objectives, the present invention also provides a device temperature monitoring system that integrates short-term load forecasting, such as... Figure 2 As shown, it includes:

[0060] The algorithm model expansion dock management module loads multiple load forecasting algorithms and configures the necessary parameters. After the algorithm starts, it outputs the load forecasting curve in real time.

[0061] The result wrapper in the algorithm model expansion dock management module reads the load forecast curve, packages the load forecast curve into an object sequence in a unified time series format, marks it with a business tag, and then sends the object sequence to the trigger of the algorithm model expansion dock management module.

[0062] The trigger distributes the object sequence to the time series preprocessing module registered on it based on different business tags;

[0063] The time series preprocessing module receives the object sequence, first reads the time window parameter configuration, and then extracts the valid time window sequence from the time window sequence through the internally set TopN time window picking unit according to the parameters. The time window sequence is further aggregated and filtered through the time window filter to obtain the final valid time window sequence, and then sent to the automatic scheduling module.

[0064] The patrol task construction module, in which the patrol task management unit first creates patrol tasks to be executed, and each new patrol task to be executed has an automatically generated policy container and is bound to it. The task execution policy depends on the specific policy loaded in the policy container.

[0065] The automatic scheduling module receives the time window sequence and hands it over to the internal time window conversion unit, which converts the time window data into the task execution strategy of the inspection task. At the same time, it reads the inspection task to be executed from the inspection task construction module, and sends the task execution strategy and inspection task to the task automatic orchestration unit. The task automatic orchestration unit dynamically binds the task execution strategy into the corresponding strategy container according to the inspection task information, and finally generates an executable inspection task and sends it to the inspection task execution module.

[0066] The patrol task execution module receives executable patrol tasks, puts the patrol tasks into the patrol task queue, the built-in task execution unit retrieves the patrol tasks from the patrol task queue, and automatically executes them according to the task execution policy bound in the policy container. The result of the patrol task execution is sent to the message push module.

[0067] The message push module provides a registration function to the business system and pushes the results of the inspection task to the business system.

[0068] Furthermore, according to one embodiment of the present invention, the algorithm model expansion dock management module performs unified modeling of load forecasting algorithms. The algorithm model expansion dock management module supports the simultaneous loading of multiple load forecasting algorithms, which are logically isolated from each other and do not affect each other. Algorithm configuration parameters can be dynamically bound to the algorithm service, and an algorithm instance can be started. After the algorithm instance runs, it outputs a load forecast curve.

[0069] In this embodiment, since the load forecast curves generated by different forecasting algorithms may have different formats, the result wrapper reads the load forecast curves, performs format conversion and wrapping, and generates a unified time series format object sequence. Each object in the object sequence contains at least two attributes: load and time. At the same time, the business tag to which the sequence belongs is wrapped into the header of the sequence.

[0070] Furthermore, according to one embodiment of the present invention, the time series preprocessing module supports the creation of different preprocessing services based on different business tags, and these services need to be registered to triggers.

[0071] The trigger reads the packaged time series (object series), first reads the business tag in the header, and then sends the object series to the corresponding service of the corresponding time series preprocessing module through the business tag.

[0072] In this embodiment, after receiving the time series, the service of the time series preprocessing module first reads the time window configuration parameters. Based on the configuration parameters, it uses the TopN time window picking unit to extract the top N time window sequences in terms of load from the time series. The value of N can be dynamically set according to the length of the time series and the width of the time window. At this time, there are cases of time overlap and invalid windows in the TopN time window sequences. It is necessary to aggregate and filter them through the time window filter to obtain the final valid time window sequence, and then send it to the automatic scheduling module.

[0073] Furthermore, according to one embodiment of the present invention, the task management unit of the inspection task construction module includes operations for adding, modifying, deleting and querying inspection tasks. After the task is created, a business tag is marked for the task. At the same time, the strategy container service in the module automatically builds a new strategy container for the task and binds the container to the task. The role of the strategy container is to load multiple task execution strategies. After the above steps are completed, the task is sent to the automatic scheduling module.

[0074] Furthermore, according to one embodiment of the present invention, the automatic scheduling module receives a time window sequence and a task instance with a strategy container. The time window conversion unit in the automatic scheduling module reads the time window sequence and converts the time window data into a task execution strategy, which includes the task trigger time, task duration, and task cycle. The strategy is then sent to the task automatic orchestration unit for the orchestration of specific tasks. The corresponding task can be found according to the business tag on the task execution strategy. The strategy is loaded into the strategy container corresponding to the task to generate a task to be executed. The task to be executed is then sent to the task execution module.

[0075] Furthermore, according to one embodiment of the present invention, the patrol task execution module receives tasks to be executed and first places the tasks into the patrol task queue. The patrol task execution unit retrieves the tasks to be executed from the task queue, then obtains the task execution strategy according to the strategy container bound to the task, and finally obtains the task result.

[0076] The task execution module sends the task execution results to the message push module, which has a built-in business system registration unit and result distribution unit.

[0077] In this embodiment, the result distribution unit first reads the registration information from the business system registration unit, identifies the business system to which the task result needs to be sent through the registration information, and accurately pushes the calculation result to each business system through the service bus.

[0078] According to the above-described scheme of the present invention, the equipment temperature inspection system integrating short-term load forecasting of the present invention is actually an automatic scheduling system that implements the above-described equipment temperature inspection method integrating short-term load forecasting and its automatic scheduling inspection tasks. The system can track the predicted state of external conditions in the future and adaptively schedule the inspection tasks to be executed according to the predicted state, enhance the coordination between the execution strategy and the execution tasks, reduce the resource overhead of maintenance work required due to changes in external conditions, and improve the system's flexibility, adaptability and application efficiency.

[0079] Furthermore, to achieve the above objectives, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the device temperature inspection method as described above, which integrates short-term load prediction.

[0080] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the device temperature inspection method for integrating short-term load prediction as described above.

[0081] Those skilled in the art will recognize that the modules and algorithm steps described in conjunction with the embodiments disclosed herein can be implemented using electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0082] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and equipment can be referred to the corresponding process in the foregoing method implementation, and will not be repeated here.

[0083] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0084] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0085] In addition, the functional modules in the embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0086] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the energy-saving signal transmission / reception methods of various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0087] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in this application.

[0088] It should be understood that the sequence number of each step in the invention and its embodiments does not absolutely imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

Claims

1. A method for monitoring equipment temperature by integrating short-term load forecasting, characterized in that, include: The load forecast curve for a preset future time period is generated using a load forecasting algorithm. The load forecast curve is transformed and packaged to output a sequence of objects with a uniform format; Set the TopN time window parameters to obtain the specific TopN time window entities; The object sequence is passed through the TopN time window entity to obtain several discrete time window sequences within a preset future duration. The time window sequences are then aggregated and filtered to obtain the final time window sequence. Based on the obtained final time window sequence, a task execution strategy is dynamically generated, and this task execution strategy is bound to a specific equipment temperature inspection task; The equipment temperature monitoring task is executed according to the bound task execution strategy to achieve tracking of load operation; The process of transforming and packaging the load curve to output a uniformly formatted object sequence is as follows: The load forecast curve is read by the result wrapper and the format is converted and wrapped to generate a unified time series format object sequence. Each object in the object sequence contains at least two attributes: load and time. At the same time, the business tag to which the object sequence belongs is wrapped into the header of the object sequence. The specific TopN time window entity obtained by setting the TopN time window parameters is as follows: Set the bottom frame coordinates, window height, and window width of the TopN time windows to obtain the specific TopN time window entities; The final time window sequence obtained by aggregating and filtering the time window sequence is as follows: The time window sequences are aggregated and filtered using a time window filter to obtain the final effective time window sequence.

2. The equipment temperature inspection method integrating short-term load forecasting according to claim 1, characterized in that, The load forecast curve for a preset future time period generated by the load forecasting algorithm is as follows: Based on the fuzzy neural network load forecasting algorithm, the short-term parameter of the algorithm is set to 24 hours, the trigger time is set to a fixed time, and after the algorithm service runs, it outputs the load forecast curve for the next 24 hours at a fixed time every day.

3. The equipment temperature monitoring method based on short-term load forecasting according to claim 1 or 2, characterized in that, The task execution strategy includes: task trigger time, task duration, and task cycle.

4. An equipment temperature monitoring system integrating short-term load forecasting, characterized in that, include: The algorithm model expansion dock management module loads the load prediction algorithm model and configures the necessary parameters. After the algorithm starts, it outputs the load prediction curve in real time. The parameters configured are the short-term parameters and trigger time of the prediction algorithm. The result wrapper in the algorithm model expansion dock management module reads the load forecast curve, packages the load forecast curve into an object sequence in a unified time series format, marks it with a business tag, and then sends the object sequence to the trigger of the algorithm model expansion dock management module. The trigger distributes the object sequence to the time series preprocessing module registered on it based on different business tags; The time series preprocessing module receives the object sequence, first reads the time window parameter configuration, and then extracts the valid time window sequence from the time window sequence through the internally set TopN time window picking unit according to the parameters. The time window sequence is further aggregated and filtered through the time window filter to obtain the final valid time window sequence, and then sent to the automatic scheduling module. The patrol task construction module, in which the patrol task management unit first creates patrol tasks to be executed, and each new patrol task to be executed has an automatically generated policy container and is bound to it. The task execution policy depends on the specific policy loaded in the policy container. The automatic scheduling module receives the time window sequence and hands it over to the internal time window conversion unit, which converts the time window data into the task execution strategy of the inspection task. At the same time, it reads the inspection task to be executed from the inspection task construction module, and sends the task execution strategy and inspection task to the task automatic orchestration unit. The task automatic orchestration unit dynamically binds the task execution strategy into the corresponding strategy container according to the inspection task information, and finally generates an executable inspection task and sends it to the inspection task execution module. The patrol task execution module receives executable patrol tasks, puts the patrol tasks into the patrol task queue, the built-in task execution unit retrieves the patrol tasks from the patrol task queue, and automatically executes them according to the task execution policy bound in the policy container. The result of the patrol task execution is sent to the message push module. The message push module provides a registration function to the business system and pushes the results of the inspection task to the business system.

5. An electronic device, characterized in that, The device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the device temperature inspection method for fusion short-term load prediction as described in any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the equipment temperature inspection method for fusion short-term load prediction as described in any one of claims 1-3.

Images