Fogging early warning method based on multi-source environmental parameter fusion of fuzzy PID control

Abstract

The application discloses a multi-source environmental parameter fusion condensation warning method based on fuzzy PID control, and builds a simulation environment of a high-voltage switch room and collects multi-source environmental parameters; moreover, the interference of equipment operation in the high-voltage switch room on the condensation formation environment is considered, an integrated data set of the condensation formation related environmental data set and the interference factor data set is formed, and a condensation formation prediction model is constructed in the simulation environment through machine learning; the real-time temperature and humidity data of the high-voltage switch room are used to determine the actual condensation state in the switch room and form a warning notice; according to the actual condensation state, the discharge risk state of the switch room is estimated, and matched fuzzy PID environmental factor regulation is implemented. Through the construction of the prediction model based on the multi-source environmental parameters of the high-voltage switch room and the environmental interference formed by the operation of electrical components, advanced condensation formation identification monitoring is carried out, condensation formation is timely and efficiently inhibited, and the operation safety of the high-voltage switch room is ensured.

Description

Technical Field

[0001] This invention relates to the field of environmental parameters, and particularly to a method for condensation early warning based on fuzzy PID control and multi-source environmental parameter fusion. Background Technology

[0002] High-voltage switchgear rooms house numerous electrical components widely used in substation equipment. For heat dissipation, these rooms are not sealed environments but are connected to the outside atmosphere through ventilation holes. When the ambient temperature and humidity meet certain conditions, condensation will form inside the switchgear room. The occurrence of condensation depends primarily on the relative humidity and dew point temperature. Given that the switchgear room operates within a high-voltage electric field, electrical charges will adhere to the area around the condensation. When the condensation density reaches a certain level and sufficient charge accumulates around it, condensation-induced power generation can occur, posing a risk of internal high-voltage breakdown to the electrical components. This risk is particularly high in humid southern climates, significantly increasing the probability of condensation formation inside the switchgear room and raising the risk of damage.

[0003] Currently, condensation detection in high-voltage switchgear rooms typically employs infrared imaging to directly acquire infrared images of the switchgear's interior. Utilizing the absorption characteristics of infrared light by condensation, the formation of condensation within the switchgear room can be accurately determined. However, the large number and complex distribution of electrical components inside switchgear rooms can easily obstruct the camera view. Furthermore, condensation formation is a gradual process; by the time the infrared camera detects condensation, it indicates a significant amount of condensation has formed, making early warning of condensation formation impossible. Therefore, monitoring and early warning based on environmental parameters, as well as suppressing and controlling condensation formation throughout the entire process, are crucial for ensuring the safe operation of high-voltage switchgear rooms. Summary of the Invention

[0004] Considering that existing methods for monitoring condensation formation in high-voltage switchgear rooms using infrared cameras suffer from significant obstruction and recognition delays, they cannot provide early detection and monitoring of condensation formation across the entire interior of the switchgear room. This prevents timely and appropriate measures from being taken to suppress condensation, thus compromising the operational safety of the switchgear room. In view of these problems, this invention provides a condensation early warning method based on fuzzy PID control and multi-source environmental parameter fusion, comprising: Step S1: Build a simulation environment for the high-voltage switch room and divide the simulation environment into several subspaces; collect real-time multi-source environmental parameters of the high-voltage switch room and identify the useful multi-source environmental parameters therein; Step S2: Based on the useful multi-source environmental parameters, obtain the condensation generation associated environmental dataset for each subspace; acquire the equipment operation data within each subspace to obtain the interference factor dataset; merge the associated environmental dataset and the interference factor dataset to obtain the integrated dataset; Step S3: Based on the integrated dataset and machine learning, construct a condensation generation prediction model; based on the condensation generation prediction model, determine the actual condensation characteristics inside the high-voltage switch room, thereby generating an early warning notification. Step S4: Estimate the discharge risk status of the high-voltage switch room based on the actual condensation characteristics; implement matched fuzzy PID environmental factor control based on the discharge risk status.

[0005] Optionally, in step S1, a simulation environment for the high-voltage switch room is built, and the simulation environment is divided into several subspaces, including: The structure and equipment layout of the high-voltage switchgear room are obtained; wherein, the structure includes the internal structural layout of the high-voltage switchgear room; and the equipment layout includes the spatial location, three-dimensional shape, and power parameters of the equipment inside the high-voltage switchgear room. Based on the aforementioned structure and equipment layout, the airflow characteristics inside the high-voltage switch chamber are determined; wherein, the airflow characteristics include the shape and distribution of airflow channels; Based on the structure of the high-voltage switch room, a matching simulation environment is built; based on the airflow characteristics, the simulation environment is divided into several subspaces.

[0006] Optionally, in step S1, real-time multi-source environmental parameters of the high-voltage switchgear room are collected, and useful multi-source environmental parameters are identified, including: Real-time multi-source environmental parameters of the indoor and outdoor spaces of the high-voltage switchgear are collected; based on the real-time multi-source environmental parameters, the fluctuation value of the indoor environmental parameters of the high-voltage switchgear caused by the difference between the indoor and outdoor spaces for each environmental parameter is determined. A permissible fluctuation range for the internal environmental parameters of the high-voltage switch room is set. If the fluctuation value of the internal environmental parameters of the high-voltage switch room matches the permissible fluctuation range, the corresponding environmental parameter is determined to be a useless environmental parameter; otherwise, the corresponding environmental parameter is determined to be a useful environmental parameter, thereby obtaining useful multi-source environmental parameters.

[0007] Optionally, in step S2, based on the useful multi-source environmental parameters, a condensation generation associated environmental dataset for each subspace is obtained, including: The environmental state data contained in the useful multi-source environmental parameters are subjected to noise reduction preprocessing and timestamp label preprocessing. The data portion associated with condensation formation is extracted from the preprocessed environmental state data. Temporal variation features and spatial feature identification are performed on the data portion to obtain the condensation formation associated environmental dataset for each subspace. The condensation formation associated environmental dataset includes data reflecting temperature and humidity.

[0008] Optionally, in step S2, the device operation data within each subspace is obtained to get a dataset of interference factors, including: Obtain the operating mode and operating conditions of the equipment within each subspace, and estimate the local environmental disturbance data directly generated during equipment operation based on the operating mode and operating conditions; The local environmental disturbance data is subjected to time-domain calibration preprocessing and spatial calibration preprocessing to obtain a disturbance factor dataset; wherein, the disturbance factor dataset includes the subspace temperature and / or airflow disturbance distribution data caused by equipment operation.

[0009] Optionally, in step S2, the associated environment dataset and the interference factor dataset are merged to obtain an integrated dataset, including: Based on the spatiotemporal identifiers of the data, the data reflecting temperature and humidity included in the associated environmental dataset and the data on temperature and / or airflow interference distribution in the subspace caused by equipment operation included in the interference factor dataset are fused and aligned in the temporal and spatial domains to obtain the physical field distribution of each subspace. Temperature and humidity data with spatiotemporal labels are extracted from the physical field distribution to obtain an integrated dataset; wherein the spatiotemporal labels include timestamps and spatial coordinates.

[0010] Optionally, in step S3, a condensation generation prediction model is constructed based on the integrated dataset and machine learning, including: The integrated dataset is divided into a training dataset and a test dataset. A machine learning algorithm is used to train a condensation prediction model using the training dataset. The trained condensation generation prediction model was validated and its parameters were optimized using the test dataset to obtain the final usable condensation generation prediction model.

[0011] Optionally, in step S3, based on the condensation generation prediction model, the actual condensation characteristics inside the high-voltage switch room are determined to generate an early warning notification, including: The real-time temperature and humidity distribution data of each subspace inside the high-voltage switch room are obtained, the real-time temperature and humidity distribution data are converted into feature vectors and input into the final usable condensation generation prediction model, so as to obtain the actual condensation characteristics of each subspace; wherein, the actual condensation characteristics include the spatial distribution of condensation particle size and the spatial distribution of condensation quantity. Based on the actual characteristics of the condensation, the trend of condensation filling rate change in each subspace is estimated to generate an early warning notification.

[0012] Optionally, based on the actual characteristics of the condensation, the actual electric field distribution around the condensation is estimated to determine whether the triggering conditions for the condensation discharge event are met. Based on the determination result of the triggering condition, the discharge risk status of each subspace in the high-voltage switch room is estimated; wherein, the discharge risk status includes the expected occurrence event and the expected intensity of the condensation power generation event.

[0013] Optionally, in step S4, according to the discharge risk state, matching fuzzy PID environmental factor control is implemented, including: Based on the discharge risk status, a condensation generation suppression target is generated for each subspace in the high-voltage switch chamber; Based on the difference between the condensation suppression target and the actual shape of the condensation in the subspace, fuzzy PID control is applied to the condensation suppression devices arranged in the subspace until the condensation suppression target is achieved.

[0014] The beneficial effects of the above-mentioned technical solutions provided in the embodiments of the present invention include at least the following: This invention provides a condensation early warning method based on fuzzy PID control and multi-source environmental parameter fusion. This method constructs a simulation environment of a high-voltage switchgear room and collects multi-source environmental parameters. It also considers the interference caused by the operation of equipment within the switchgear room on the condensation formation environment, forming an integrated dataset that combines a condensation generation-related environmental dataset and an interference factor dataset. Using this dataset, a condensation generation prediction model is constructed in the simulation environment through machine learning. Real-time temperature and humidity data from the high-voltage switchgear room are used to determine the actual condensation characteristics inside the switchgear room and generate an early warning notification. Based on the actual condensation characteristics, the discharge risk state of the switchgear room is estimated, and matched fuzzy PID environmental factor control is implemented. By constructing a prediction model based on the multi-source environmental parameters of the high-voltage switchgear room and the environmental interference caused by the operation of electrical components as corrections, early condensation generation identification and monitoring are performed, timely and efficient suppression of condensation generation is achieved, and the operational safety of the high-voltage switchgear room is ensured.

[0015] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion provided in this embodiment of the invention.

[0018] Figure 2 This refers to the internal structure of the high-voltage switchgear room.

[0019] Figure 3 This shows the condensation distribution output by the condensation formation prediction model. Detailed Implementation

[0020] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0021] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," "far," "near," "front," and "rear," etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings and are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0022] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0023] Please see Figure 1 As shown, an embodiment of this application provides a condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion. This condensation early warning method based on fuzzy PID control includes: Step S1: Build a simulation environment for the high-voltage switch room and divide the simulation environment into several subspaces; collect real-time multi-source environmental parameters of the high-voltage switch room and identify the useful multi-source environmental parameters. Step S2: Based on useful multi-source environmental parameters, obtain the condensation generation associated environmental dataset for each subspace; acquire the equipment operation data within each subspace to obtain the interference factor dataset; merge the associated environmental dataset and the interference factor dataset to obtain the integrated dataset; Step S3: Based on the integrated dataset and machine learning, construct a condensation generation prediction model; based on the condensation generation prediction model, determine the actual condensation characteristics inside the high-voltage switch room, thereby generating an early warning notification. Step S4: Estimate the discharge risk status of the high-voltage switch room based on the actual condensation characteristics; implement matched fuzzy PID environmental factor control based on the discharge risk status.

[0024] This condensation early warning method based on fuzzy PID control, which integrates multi-source environmental parameters, constructs a predictive model that uses multi-source environmental parameters of the high-voltage switchgear room as a basis and environmental interference generated by the operation of electrical components as a correction. It then identifies and monitors condensation generation in advance, suppresses condensation generation in a timely and efficient manner, and ensures the operational safety of the high-voltage switchgear room.

[0025] In another embodiment, in step S1, a simulation environment for the high-voltage switch room is established, and the simulation environment is divided into several subspaces, including: Obtain the structure and equipment layout of the high-voltage switchgear room; wherein, the structure includes the internal structural layout of the high-voltage switchgear room; the equipment layout includes the spatial location, three-dimensional shape and power parameters of the equipment inside the high-voltage switchgear room; Based on the structure and equipment layout, determine the airflow characteristics inside the high-voltage switchgear room; among which, the airflow characteristics include the shape and distribution of airflow channels; Based on the structure of the high-voltage switch room, a matching simulation environment was built; based on the airflow characteristics, the simulation environment was divided into several subspaces.

[0026] Please see Figure 2The high-voltage switchgear room, as a functional device in a high-voltage substation, mainly comprises different electrical modules such as busbar room, protection room, cable room, and circuit breaker room. Each electrical module contains various types of electrical components, with different placement locations and operating power. Each electrical module is housed as a relatively independent component in a specific area within the switchgear room, and these specific areas are separated by partitions, thus forming multiple relatively independent isolated spaces within the switchgear room. Understandably, electrical components need to be interconnected to function properly, and heat is generated during operation. To meet the wiring connection requirements between electrical components and to dissipate heat effectively, channels are provided between different isolated spaces, and cooling fans are installed at different locations within the switchgear room. It is understandable that the high-voltage switchgear room houses multiple electrical modules, each with its own number and power of electrical components, and each module forms a relatively independent isolated space. This results in each module having a relatively independent internal atmospheric environment (such as temperature and humidity), making the temperature and humidity conditions of each module's isolated space relatively independent and difficult to be affected by the isolated spaces of other electrical modules. Furthermore, different electrical components generate different amounts of heat during operation, causing varying degrees of temperature change in their respective isolated spaces. Therefore, while the relative humidity and temperature of the isolated space corresponding to one electrical module may meet the conditions for condensation formation, this does not mean that the relative humidity and temperature of the isolated spaces corresponding to other electrical modules also meet the conditions. Thus, spatial differentiation and identification are necessary within the high-voltage switchgear room to accurately determine the condensation formation situation in different areas.

[0027] In practice, high-voltage switchgear rooms can be simulated to create a simulation environment that matches the structure and thermal effects of electrical components. This facilitates spatial zoning of the atmospheric environment within the simulation environment, allowing for the extraction of multi-source environmental parameters that substantially influence condensation formation in different areas of the switchgear room. Specifically, the internal structural layout of the switchgear room (e.g., the shape and size of each isolation space, partition distribution, and the location and dimensions of channels between different isolation spaces) is obtained, along with the spatial location, three-dimensional shape, and power parameters of all electrical components. Since the electrical components generate heat during operation, the amount of heat is related to their power parameters, and the heat radiation efficiency and range are related to the three-dimensional shape of their spatial location. Based on the switchgear room's structure and equipment layout, the shape and distribution of airflow channels within the switchgear room are determined. These airflow channels refer to the gaps within the switchgear room that allow for effective airflow (airflow velocity greater than a preset velocity threshold) and their three-dimensional spatial distribution. A matching simulation environment is then built based on the switchgear room's structure, corresponding to the actual three-dimensional structure of the switchgear room. The aforementioned airflow characteristics are then superimposed onto the simulation environment, thereby dividing the simulation environment into several subspaces. It can be understood that each subspace is in a state of relatively independent airflow characteristics, and the probability of airflow interaction between different subspaces is less than a preset probability threshold, which facilitates subsequent prediction of condensation generation in the high-voltage switch room by subspace as a unit.

[0028] In another embodiment, in step S1, real-time multi-source environmental parameters of the high-voltage switchgear room are collected, and useful multi-source environmental parameters are identified, including: Collect real-time multi-source environmental parameters of the indoor and outdoor spaces of the high-voltage switchgear; based on the real-time multi-source environmental parameters, determine the fluctuation value of the indoor environmental parameters of the high-voltage switchgear caused by the difference between the indoor and outdoor spaces for each environmental parameter; Set the allowable fluctuation range of the internal environmental parameters of the high-voltage switch room. If the fluctuation value of the internal environmental parameters of the high-voltage switch room matches the allowable fluctuation range, the corresponding environmental parameter is determined to be a useless environmental parameter; otherwise, the corresponding environmental parameter is determined to be a useful environmental parameter, thus obtaining useful multi-source environmental parameters.

[0029] High-voltage switchgear rooms are typically located outdoors, and the temperature and humidity of the outdoor environment can affect the internal temperature and humidity of the switchgear room to varying degrees. Understandably, changes in outdoor temperature or humidity can disrupt the existing temperature or humidity balance inside the switchgear room, leading to changes in temperature or humidity. However, not every change in outdoor temperature or humidity will cause a change in the internal temperature or humidity of the switchgear room. It is only when the temperature difference between the outdoor environment and the internal temperature of the switchgear room is sufficiently large, and the amplitude of the outdoor temperature change is sufficiently large, that a change in temperature will occur inside the switchgear room. Similarly, only when the humidity difference between the outdoor environment and the internal humidity of the switchgear room is sufficiently large, and the amplitude of the outdoor humidity change is sufficiently large, will a change in humidity will occur inside the switchgear room. Therefore, the real-time multi-source environmental parameters collected for the switchgear room (such as the real-time temperature and humidity parameters of both the interior and exterior environments) are not directly related to condensation formation inside the switchgear room. Thus, it is necessary to screen and identify the collected real-time multi-source environmental parameters to obtain a subset of multi-source environmental parameters that have a substantial impact on condensation formation inside the switchgear room (i.e., useful multi-source environmental parameters).

[0030] Specifically, real-time multi-source environmental parameters of the high-voltage switchgear room are compared and analyzed using neural networks. This yields the fluctuation values ​​of the internal environmental parameters (temperature or humidity) caused by the differences between the internal and external spaces for each environmental parameter (temperature or humidity). If these fluctuation values ​​fall within the preset allowable fluctuation range for the internal environmental parameters of the high-voltage switchgear room, the corresponding environmental parameter is determined to be a useless environmental parameter. Otherwise, the corresponding environmental parameter is determined to be a useful environmental parameter, and all useful environmental parameters are integrated into useful multi-source environmental parameters, providing a data foundation for the subsequent generation of a correlated environmental dataset.

[0031] In another embodiment, in step S2, the condensation generation associated environmental dataset for each subspace is obtained based on useful multi-source environmental parameters, including: Denoising preprocessing and timestamp label preprocessing are performed on all environmental state data contained in useful multi-source environmental parameters; The data portion associated with condensation formation is extracted from the preprocessed environmental state data. Temporal variation features and spatial feature labels are then extracted from this data portion to obtain the condensation formation associated environmental dataset for each subspace. The condensation formation associated environmental dataset includes data reflecting temperature and humidity.

[0032] Useful multi-source environmental parameters include temperature and humidity parameters that affect condensation formation. To accurately extract the temperature and humidity parameters related to condensation formation, noise reduction and timestamp labeling (adding a generation timestamp to each environmental state data) are first performed on all environmental state data (i.e., temperature and humidity data) included in the above-mentioned useful multi-source environmental parameters, thus achieving standardized processing of useful multi-source environmental parameters. Furthermore, the data portion related to condensation formation is extracted from the preprocessed environmental state data, and time-domain variation features and spatial feature labels are extracted from this data portion to obtain the condensation formation-related environmental dataset for each subspace. In this way, the aforementioned condensation formation-related environmental dataset accurately reflects the temperature and humidity data affecting condensation formation, providing a data foundation for subsequent condensation formation prediction.

[0033] In another embodiment, in step S2, device operation data within each subspace is acquired to obtain a dataset of interfering factors, including: Obtain the operating mode and operating conditions of the equipment within each subspace, and estimate the local environmental disturbance data directly generated during equipment operation based on the operating mode and operating conditions; The local environmental disturbance data is preprocessed by time-domain calibration and spatial-domain calibration to obtain the disturbance factor dataset; the disturbance factor dataset includes the distribution data of subspace temperature and / or airflow disturbances caused by equipment operation.

[0034] High-voltage switchgear rooms house numerous electrical components. These components generate heat during operation, which dissipates and creates localized environmental disturbances within the switchgear room. These disturbances may include, but are not limited to, localized temperature increases and humidity decreases caused by the heat dissipation from electronic components. Furthermore, when strong airflow is present around these electronic components, the dissipated heat can spread over a wider area, resulting in greater temperature increases and humidity decreases. Therefore, the heat generated by electrical components and other equipment in high-voltage switchgear rooms during operation significantly impacts condensation formation, and these disturbances must be considered when predicting condensation formation. Specifically, the operating modes and conditions of the equipment in each subspace of the high-voltage switch room are first obtained (such as the power consumption and operating temperature of the equipment). This is used to estimate the local environmental disturbance data directly generated during equipment operation (such as local environmental temperature interference and the local environmental airflow interference it causes; the aforementioned local environmental temperature interference exhibits spatial gradient changes, which in turn generate airflow interference). Then, the aforementioned local environmental disturbance data is subjected to time-domain calibration preprocessing and spatial calibration preprocessing to obtain the subspace temperature and / or airflow interference distribution data caused by equipment operation, providing a disturbance correction basis for subsequent prediction of condensation formation.

[0035] In another embodiment, in step S2, the associated environmental dataset and the interference factor dataset are fused to obtain an integrated dataset, including: Based on the spatiotemporal identifiers of the data, the data reflecting temperature and humidity in the associated environmental dataset and the data on temperature and / or airflow disturbances in the subspace caused by equipment operation in the disturbance factor dataset are fused and aligned in the temporal and spatial domains to obtain the physical field distribution of each subspace. Temperature and humidity data with spatiotemporal labels are extracted from the physical field distribution to obtain an integrated dataset; the spatiotemporal labels include timestamps and spatial coordinates.

[0036] As discussed above, the temperature and humidity data in the associated environmental dataset and the temperature and / or airflow disturbance distribution data in the interfering factors dataset, caused by equipment operation, jointly influence condensation formation. To achieve comprehensive condensation prediction, it is necessary to integrate these two datasets. Specifically, based on the temporal and spatial identifiers of the data, the temperature and humidity data in the associated environmental dataset and the temperature and / or airflow disturbance distribution data in the interfering factors dataset are fused and aligned in both the temporal and spatial domains to obtain the physical field distribution of each subspace. This physical field distribution represents the temperature and humidity distribution at each location point within each subspace at a series of time points. Temperature and humidity data with timestamps and spatial coordinates are then extracted from the physical field distribution to obtain the integrated dataset, providing a training and testing data foundation for subsequently building a condensation prediction model.

[0037] In another embodiment, in step S3, a condensation generation prediction model is constructed based on the integrated dataset and machine learning, including: The integrated dataset is divided into a training dataset and a test dataset. A predictive model for condensation is trained using a machine learning algorithm on the training dataset. The training model for predicting condensation formation was validated and its parameters were optimized using a test dataset to obtain a final, usable model for predicting condensation formation.

[0038] In practice, the integrated dataset is divided into a training dataset and a test dataset according to a preset ratio (e.g., 3:7). A condensation generation prediction model is trained using the training dataset through a machine learning algorithm (e.g., a convolutional neural network algorithm). The test dataset is then used to verify the performance and optimize the parameters of the condensation generation prediction model, resulting in a final usable condensation generation prediction model. The performance verification and parameter optimization described above are standard practices in this field and will not be described in detail here.

[0039] In another embodiment, in step S3, the actual condensation characteristics inside the high-voltage switchgear room are determined based on the condensation generation prediction model, thereby generating an early warning notification, including: Real-time temperature and humidity distribution data of each subspace inside the high-voltage switch room are acquired, and the real-time temperature and humidity distribution data are converted into feature vectors and input into the final usable condensation generation prediction model to obtain the actual condensation characteristics of each subspace; wherein, the actual condensation characteristics include the spatial distribution of condensation particle size and the spatial distribution of condensation quantity. Based on the actual characteristics of condensation, the trend of condensation filling rate change in each subspace is estimated to generate an early warning notification.

[0040] Understandably, the aforementioned final usable condensation formation prediction model is adapted to the internal structure of the high-voltage switchgear room. At this point, real-time temperature and humidity distribution data for each subspace within the high-voltage switchgear room are acquired. This real-time temperature and humidity distribution data is then semantically vectorized to obtain feature vectors. These feature vectors are then input into the condensation formation prediction model, thereby outputting the actual condensation characteristics of each subspace. Please refer to [link / reference]. Figure 3 The condensation generation test model analyzes and processes the aforementioned feature vectors to output the condensation distribution. This condensation distribution characterizes the spatial distribution of condensation particle size and quantity generated in the subspace. It is understandable that condensation generation within the subspace is a dynamic process. Under the condition of satisfying condensation generation requirements, the particle size and quantity of condensation generated in the subspace gradually increase, causing the subspace to be continuously filled with condensation. When the proportion of condensation volume filling the subspace reaches a preset threshold, it indicates a significant risk of insulation failure in the subspace. At this point, a corresponding early warning notification is issued to facilitate timely condensation suppression operations in the high-voltage switchgear room.

[0041] In another embodiment, step S4, estimating the discharge risk status of the high-voltage switchgear room based on the actual condensation characteristics, includes: Based on the actual characteristics of condensation, estimate the actual electric field distribution around the condensation to determine whether the triggering conditions for a condensation discharge event are met. Based on the determination of the triggering conditions, the discharge risk status of each subspace in the high-voltage switch room is estimated; wherein, the discharge risk status includes the expected occurrence and intensity of condensation power generation events.

[0042] As a conductive material, condensation, when suspended and exposed in the environment inside a high-voltage switchgear room, adsorbs electrical charges. When the amount of charge accumulated inside the condensation reaches a certain level, it forms a high-voltage electric field and triggers external power generation, thereby damaging electrical components. The more condensation formed inside the high-voltage switchgear room, the larger the condensation particle size, the more charge accumulated, and the higher the corresponding electric field strength. To accurately predict discharge events caused by charge accumulation during condensation formation, the amount of charge accumulated in the condensation is estimated based on the spatial distribution of condensation particle size and quantity. This determines the actual electric field distribution around the condensation (i.e., the electric field strength distributed around the condensation). If the electric field strength is greater than a preset strength threshold, it is determined that the electric field around the condensation will trigger a discharge event; otherwise, it is determined that the electric field around the condensation will not trigger a discharge event. Based on the judgment of whether the condensation discharge event is triggered, combined with the increasing trend of the electric field strength around the condensation, the expected occurrence event and expected intensity of condensation power generation events in each subspace are estimated, providing a basis for subsequent adjustments to the condensation generation suppression operation.

[0043] In another embodiment, in step S4, matching fuzzy PID environmental factor control is implemented according to the discharge risk state, including: Based on the discharge risk status, condensation generation suppression targets are generated for each sub-space within the high-voltage switch chamber; Based on the difference between the condensation suppression target and the actual shape of the condensation in the subspace, fuzzy PID control is implemented on the condensation suppression equipment arranged in the subspace until the condensation suppression target is achieved.

[0044] The size and quantity of condensation particles generated inside the high-voltage switchgear directly determine the level of discharge risk. To prevent damage from discharge breakdown, timely condensation suppression operations (such as heating and drying) are necessary. In practice, based on the predicted occurrence and intensity of condensation generation events in each subspace, condensation suppression targets are generated for each subspace within the high-voltage switchgear. These targets may include, but are not limited to, reduction targets for condensation density and particle size within the high-voltage switchgear. Then, based on the difference between the condensation suppression targets and the actual shape of the condensation in the subspace, fuzzy PID control is applied to the condensation suppression equipment within the subspace until the condensation suppression targets are achieved. This fuzzy PID control refers to the application of fuzzy PID closed-loop control to the heating equipment within the subspace, effectively suppressing condensation and ensuring the operational safety of the high-voltage switchgear.

[0045] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. This disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims. Thus, if these modifications and variations of the invention fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A condensation early warning method based on multi-source environmental parameter fusion using fuzzy PID control, characterized in that, include: Step S1: Build a simulation environment for the high-voltage switch room and divide the simulation environment into several subspaces; collect real-time multi-source environmental parameters of the high-voltage switch room and identify the useful multi-source environmental parameters therein; Step S2: Based on the useful multi-source environmental parameters, obtain the condensation generation associated environmental dataset for each subspace; obtain the equipment operation data within each subspace to obtain the interference factor dataset; By merging the associated environment dataset and the interference factor dataset, an integrated dataset is obtained; Step S3: Based on the integrated dataset and machine learning, construct a condensation generation prediction model; Based on the condensation generation prediction model, the actual condensation characteristics inside the high-voltage switch room are determined, thereby generating an early warning notification; Step S4: Estimate the discharge risk status of the high-voltage switch room based on the actual characteristics of the condensation. Based on the discharge risk status, a matching fuzzy PID environmental factor control is implemented.

2. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 1, characterized in that: In step S1, a simulation environment for the high-voltage switch room is built, and the simulation environment is divided into several subspaces, including: The structure and equipment layout of the high-voltage switchgear room are obtained; wherein, the structure includes the internal structural layout of the high-voltage switchgear room; and the equipment layout includes the spatial location, three-dimensional shape, and power parameters of the equipment inside the high-voltage switchgear room. Based on the aforementioned structure and equipment layout, the airflow characteristics inside the high-voltage switch chamber are determined; wherein, the airflow characteristics include the shape and distribution of airflow channels; Based on the structure of the high-voltage switch room, a matching simulation environment is built; based on the airflow characteristics, the simulation environment is divided into several subspaces.

3. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 2, characterized in that: In step S1, real-time multi-source environmental parameters of the high-voltage switch room are collected, and useful multi-source environmental parameters are identified, including: Real-time multi-source environmental parameters of the indoor and outdoor spaces of the high-voltage switchgear are collected; based on the real-time multi-source environmental parameters, the fluctuation value of the indoor environmental parameters of the high-voltage switchgear caused by the difference between the indoor and outdoor spaces corresponding to each environmental parameter is determined; A permissible fluctuation range for the internal environmental parameters of the high-voltage switch room is set. If the fluctuation value of the internal environmental parameters of the high-voltage switch room matches the permissible fluctuation range, the corresponding environmental parameter is determined to be a useless environmental parameter; otherwise, the corresponding environmental parameter is determined to be a useful environmental parameter, thereby obtaining useful multi-source environmental parameters.

4. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 1, characterized in that: In step S2, based on the useful multi-source environmental parameters, a condensation generation associated environmental dataset for each subspace is obtained, including: The environmental state data contained in the useful multi-source environmental parameters are subjected to noise reduction preprocessing and timestamp label preprocessing. The data portion associated with condensation formation is extracted from the preprocessed environmental state data. Temporal variation features and spatial feature identification are performed on the data portion to obtain the condensation formation associated environmental dataset for each subspace. The condensation formation associated environmental dataset includes data reflecting temperature and humidity.

5. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 4, characterized in that: In step S2, the device operation data within each subspace is acquired to obtain the interference factor dataset, including: Obtain the operating mode and operating conditions of the equipment within each subspace, and estimate the local environmental disturbance data directly generated during equipment operation based on the operating mode and operating conditions; The local environmental disturbance data is subjected to time-domain calibration preprocessing and spatial calibration preprocessing to obtain a disturbance factor dataset; wherein, the disturbance factor dataset includes the subspace temperature and / or airflow disturbance distribution data caused by equipment operation.

6. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 5, characterized in that: In step S2, the associated environment dataset and the interference factor dataset are merged to obtain an integrated dataset, including: Based on the spatiotemporal identifiers of the data, the data reflecting temperature and humidity included in the associated environmental dataset and the data on temperature and / or airflow interference distribution in the subspace caused by equipment operation included in the interference factor dataset are fused and aligned in the temporal and spatial domains to obtain the physical field distribution of each subspace. Temperature and humidity data with spatiotemporal labels are extracted from the physical field distribution to obtain an integrated dataset; wherein the spatiotemporal labels include timestamps and spatial coordinates.

7. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 1, characterized in that: In step S3, based on the integrated dataset and machine learning, a condensation generation prediction model is constructed, including: The integrated dataset is divided into a training dataset and a test dataset. A machine learning algorithm is used to train a condensation prediction model using the training dataset. The trained condensation generation prediction model was validated and its parameters were optimized using the test dataset to obtain the final usable condensation generation prediction model.

8. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 7, characterized in that: In step S3, based on the condensation generation prediction model, the actual condensation characteristics inside the high-voltage switch room are determined to generate an early warning notification, including: The real-time temperature and humidity distribution data of each subspace inside the high-voltage switch room are obtained, the real-time temperature and humidity distribution data are converted into feature vectors and input into the final usable condensation generation prediction model, so as to obtain the actual condensation characteristics of each subspace; wherein, the actual condensation characteristics include the spatial distribution of condensation particle size and the spatial distribution of condensation quantity. Based on the actual characteristics of the condensation, the trend of condensation filling rate change in each subspace is estimated to generate an early warning notification.

9. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 1, characterized in that: In step S4, based on the actual condensation characteristics, the discharge risk status of the high-voltage switchgear room is estimated, including: Based on the actual characteristics of the condensation, estimate the actual electric field distribution around the condensation to determine whether the triggering conditions for the condensation discharge event are met. Based on the determination result of the triggering condition, the discharge risk status of each subspace in the high-voltage switch room is estimated; wherein, the discharge risk status includes the expected occurrence event and the expected intensity of the condensation power generation event.

10. The condensation early warning method based on fuzzy PID control for multi-source environmental parameter fusion as described in claim 9, characterized in that: In step S4, based on the discharge risk state, a matching fuzzy PID environmental factor control is implemented, including: Based on the discharge risk status, a condensation generation suppression target is generated for each subspace in the high-voltage switch chamber; Based on the difference between the condensation suppression target and the actual shape of the condensation in the subspace, fuzzy PID control is applied to the condensation suppression devices arranged in the subspace until the condensation suppression target is achieved.

Images