A method and system for dehumidification and condensation prevention in an indoor fully cabled substation.

Abstract

This invention relates to a method and system for dehumidification and anti-condensation in an indoor fully cabled substation. The method includes: collecting temperature and humidity data within the substation and predicting temperature and humidity trends based on a deep learning model to establish a dynamic temperature and humidity change model; establishing an adaptive thermal regulation phase change material (PCM) anti-condensation unit; deploying the PCM anti-condensation unit at anti-condensation locations in the substation based on information about condensation-prone areas identified in the dynamic temperature and humidity change model and the operating status data of the ventilation system, thereby obtaining an anti-condensation network and corresponding anti-condensation effect data; establishing an active anti-condensation protection mechanism based on a PTC heating element according to the dynamic temperature and humidity change model and the anti-condensation effect data; and establishing a comprehensive control system through a multi-objective optimization algorithm to dehumidify and prevent condensation in the substation. This invention reduces energy consumption while improving the anti-condensation effect through dynamic temperature and humidity prediction, PCM anti-condensation, and active heating technology.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a method and system for dehumidification and anti-condensation in an indoor fully cabled substation. Background Technology

[0002] With the development of power systems, indoor fully cabled substations have been widely used in urban power distribution networks. These substations typically employ a sealed structure, containing numerous cable terminations, joints, and power distribution equipment. During operation, due to temperature and humidity variations between the inside and outside of the substation, condensation easily forms on the equipment surfaces, posing a serious threat to the safe operation of electrical equipment.

[0003] Traditional dehumidification and anti-condensation technologies mainly include ventilation dehumidification, heating dehumidification, and desiccant moisture absorption. Ventilation dehumidification reduces humidity in the station building by introducing dry air; heating dehumidification uses heaters to keep the surface temperature of equipment above the dew point temperature; desiccant moisture absorption uses materials such as silica gel to absorb moisture from the air; in some areas, air conditioning and dehumidifiers are also used for station building dehumidification.

[0004] However, existing technologies often lack intelligent control, making it difficult to cope with complex and changing environmental conditions. Sometimes, they may even introduce humid air into the station building, exacerbating condensation problems. Conventional heating methods for preventing condensation are energy-intensive, and continued use significantly increases station operating costs and carbon emissions. Desiccants require frequent replacement, resulting in heavy maintenance workloads and unstable dehumidification effects. While air conditioning and dehumidifiers are more effective, they are expensive, energy-intensive, and pose risks of noise pollution and refrigerant leakage during operation. Furthermore, most existing anti-condensation and dehumidification technologies are single-method solutions, lacking systematic integration and intelligent control strategies. This makes it difficult to provide targeted protection based on the condensation risk characteristics of different areas within the station building, leading to uneven anti-condensation effects and the possibility of condensation problems in some areas. Summary of the Invention

[0005] This invention provides a method and system for dehumidification and anti-condensation in indoor fully cabled substations, which addresses the shortcomings of existing technologies.

[0006] This invention provides a method for dehumidification and condensation prevention in an indoor fully cabled substation, comprising:

[0007] S1: Collect temperature and humidity data in the substation room, obtain environmental parameter data of the substation room, and predict the temperature and humidity change trend in the substation room based on the time series deep learning model, and establish a dynamic temperature and humidity change model.

[0008] S2: Based on the dynamic temperature and humidity change model and the station building environmental parameter data, establish an adaptive thermal regulation phase change material anti-condensation unit;

[0009] S3: Based on the information of condensation-prone areas identified in the dynamic temperature and humidity change model and the operating status data of the ventilation system, the phase change material anti-condensation unit is deployed at the anti-condensation location of the power distribution room to obtain an anti-condensation network, and the working status of the anti-condensation network is monitored in real time to obtain anti-condensation effect data.

[0010] S4: Based on the dynamic temperature and humidity change model and the anti-condensation effect data, establish an active anti-condensation protection mechanism based on PTC heating elements;

[0011] S5: Based on the data flow information of the dynamic temperature and humidity change model, the ventilation system, the anti-condensation network, and the active anti-condensation protection mechanism, a comprehensive control system is established through a multi-objective optimization algorithm, and the substation is dehumidified and prevented from condensing through the comprehensive control system.

[0012] According to the present invention, a method for dehumidification and condensation prevention in an indoor fully cabled substation is provided, wherein step S1 further includes:

[0013] S11: Real-time monitoring of key locations including the surface of distribution cabinets, cable joints, condensation risk areas on walls, and corners of the station building to obtain station building environmental parameter data;

[0014] S12: Store and index the station building environmental parameter data through a time-series database to obtain a historical temperature and humidity dataset;

[0015] S13: Based on the future meteorological change prediction data and the historical temperature and humidity dataset, predict the temperature and humidity change trend and obtain the temperature and humidity change trend prediction result.

[0016] S14: Analyze the predicted results of temperature and humidity change trends using the condensation risk assessment algorithm to obtain dynamic identification information of areas prone to condensation within the station building, and obtain a complete dynamic temperature and humidity change model.

[0017] According to the present invention, a method for dehumidification and anti-condensation of an indoor fully cabled substation is provided, wherein step S13 further includes:

[0018] S131: Use an LSTM neural network to extract features from the historical temperature and humidity dataset to obtain time-series features of temperature and humidity changes;

[0019] S132: Obtain weather forecast information through external meteorological data API to obtain future weather change prediction data;

[0020] S133: The time series characteristics are fused with the future meteorological change prediction data using multivariate time series analysis to obtain comprehensive prediction input parameters;

[0021] S134: The comprehensive prediction input parameters are learned by a deep learning prediction model to obtain the prediction results of temperature and humidity change trends.

[0022] According to the present invention, a method for dehumidification and condensation prevention in an indoor fully cabled substation is provided, wherein step S2 further includes:

[0023] S21: Based on the dynamic temperature and humidity change model and the station building environmental parameter data, select a basic phase change material suitable for the station building temperature range;

[0024] S22: Add nanoscale modifiers to the basic phase change material to optimize its performance and obtain a modified phase change material;

[0025] S23: The modified phase change material is encapsulated in microcapsules with a multi-layer structure design to obtain a phase change material anti-condensation unit.

[0026] According to the present invention, a method for dehumidification and anti-condensation in an indoor fully cabled substation is provided. In step S3, the step of deploying the phase change material anti-condensation unit at the anti-condensation location in the substation, based on the information of condensation-prone areas identified in the dynamic temperature and humidity change model and the operating status data of the ventilation system, to obtain an anti-condensation network, further includes:

[0027] S31: Perform data mining on the temperature and humidity distribution data in the dynamic temperature and humidity change model to identify areas prone to condensation and obtain a location map of these areas.

[0028] S32: Computational fluid dynamics simulation is used to process the operating status data of the ventilation system, generate the airflow path and velocity distribution model in the station building, and obtain the air circulation dead zone identification results;

[0029] S33: Based on the risk rating method, the location map of the condensation-prone area and the identification results of the air circulation dead zone are comprehensively overlaid and analyzed to determine the high-risk area, medium-risk area and low-risk area, and obtain a graded anti-condensation layout scheme;

[0030] S34: According to the graded anti-condensation arrangement scheme, the phase change material anti-condensation unit is arranged at the anti-condensation position of the power distribution room to obtain an anti-condensation network.

[0031] According to the present invention, a method for dehumidification and anti-condensation in an indoor fully cabled substation is provided. Step S3, which involves real-time monitoring of the operating status of the anti-condensation network, specifically includes:

[0032] The surface temperature of the phase change material anti-condensation unit is monitored by a miniature temperature sensor to obtain the operating temperature of the anti-condensation unit.

[0033] Infrared thermal imaging technology was used to scan the temperature distribution of the phase change material anti-condensation unit to obtain data on heat diffusion effect;

[0034] By measuring the heat transfer between the phase change material anti-condensation unit and the environment using a heat flow sensor, information on energy exchange during the phase change process can be obtained.

[0035] A surface humidity detector is used to detect condensation on the protected surface and obtain anti-condensation surface condition data.

[0036] According to the present invention, a method for dehumidification and condensation prevention in an indoor fully cabled substation, S4 further includes:

[0037] S41: Based on the dynamic temperature and humidity change model and the anti-condensation effect data, integrate PTC heating elements in high-risk condensation areas to obtain a basic heating element layout;

[0038] S42: Combining the regional temperature distribution based on thermal imaging technology, optimize the layout of the basic heating element to obtain an optimized heating element layout;

[0039] S43: Power control is applied to the PTC heating elements in the optimized heating element layout to form an active anti-condensation protection mechanism based on the PTC heating elements.

[0040] According to the present invention, a method for dehumidification and condensation prevention in an indoor fully cabled substation is provided, wherein step S5 further includes:

[0041] S51: By integrating the data stream information of the dynamic temperature and humidity change model, the ventilation system, the anti-condensation network, and the active anti-condensation protection mechanism through the industrial controller, a system-level control platform is obtained;

[0042] S52: Set optimization objectives including minimizing energy consumption, maximizing dehumidification efficiency, and minimizing temperature uniformity, and construct objective functions. Obtain the optimal control strategy through a multi-objective optimization algorithm.

[0043] S53: The integrated control system is controlled by the optimal control strategy to dehumidify and prevent condensation in the power distribution room.

[0044] According to the dehumidification and anti-condensation method for an indoor fully cabled substation provided by the present invention, the expression for the optimization objective of the multi-objective optimization algorithm in step S52 is:

[0045]

[0046] Where F(x) is the optimization objective, t0 is the starting point of the time range, and t fLet λ be the endpoint of the time range, λ be the exponential decay coefficient, V(t) be the time series of the ventilation system wind speed regulation coefficient, Q(t) be the time series of the auxiliary electric heating system power regulation coefficient, x1 be the decision variable for the ventilation system wind speed regulation coefficient, x3 be the decision variable for the heating system power regulation coefficient, H(t) be the humidity time series, P(t) be the temperature distribution time series of the anti-condensation network, x2 be the decision variable for the phase change material arrangement density, and T(t) be the temperature time series. avg This represents the average temperature value.

[0047] This invention also provides an indoor fully cabled substation dehumidification and anti-condensation system, comprising:

[0048] Prediction module: Used to collect temperature and humidity data in the power distribution room, obtain environmental parameter data of the room, and predict the temperature and humidity change trend in the room based on a time series deep learning model, and establish a dynamic temperature and humidity change model.

[0049] The first construction module is used to establish an adaptive thermal regulation phase change material anti-condensation unit based on the dynamic temperature and humidity change model and the station building environmental parameter data.

[0050] Monitoring module: Used to monitor the working status of the anti-condensation network deployed at the anti-condensation location in the substation in real time and obtain anti-condensation effect data;

[0051] The second construction module is used to establish an active anti-condensation protection mechanism based on PTC heating elements, according to the dynamic temperature and humidity change model and the anti-condensation effect data.

[0052] Control module: Used to control the integrated control system including the ventilation system, the anti-condensation network and the active anti-condensation protection mechanism through a multi-objective optimization algorithm, so as to dehumidify and prevent condensation in the power distribution room.

[0053] This invention provides a method and system for dehumidification and anti-condensation in an indoor fully cabled power distribution station. By utilizing a time-series deep learning model to construct a dynamic temperature and humidity change model, the system can accurately predict future temperature and humidity trends and condensation risk distribution within the station, solving the problem that traditional methods cannot achieve proactive anti-condensation. Simultaneously, it fully leverages the temporal characteristics of temperature and humidity data, significantly improving prediction accuracy and providing a reliable basis for the precise deployment of subsequent anti-condensation measures. Secondly, this invention lays the foundation for precise deployment of anti-condensation units by generating dynamic identification information for condensation-prone areas. Furthermore, by combining data mining technology with computational fluid dynamics simulation, it not only accurately identifies condensation-prone areas but also discovers ventilation dead zones through airflow path and velocity distribution models. The risk rating method incorporates these two pieces of information. The resulting tiered anti-condensation layout scheme, developed through comprehensive analysis, significantly improves the efficiency and targeted application of anti-condensation materials. During network operation, a multi-dimensional monitoring network provides comprehensive oversight of the network's status, offering abundant data for evaluating anti-condensation effectiveness. Based on this data, a PTC heating element integration scheme, combined with thermal imaging technology to optimize heating element layout, achieves precise control of active anti-condensation, avoiding energy waste. Finally, the entire system integrates multi-source data streams through an industrial controller and employs a multi-objective optimization algorithm with three objectives: minimizing energy consumption, maximizing dehumidification efficiency, and minimizing temperature uniformity. This algorithm generates the optimal control strategy, balancing various indicators to ensure rapid response to environmental changes while maintaining system stability, demonstrating strong adaptability and stability. The overall solution of this invention enables all-weather, fully automatic, and low-energy-consumption operation of anti-condensation dehumidification in power distribution rooms, significantly improving the safety and reliability of power distribution equipment and reducing maintenance costs. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0055] Figure 1 This is a schematic diagram of a dehumidification and anti-condensation method for an indoor fully cabled substation provided in an embodiment of the present invention;

[0056] Figure 2 This is a schematic diagram of a dehumidification and anti-condensation system for an indoor fully cabled power distribution room, provided as an embodiment of the present invention.

[0057] Figure label:

[0058] 100. Prediction module; 200. First construction module; 300. Monitoring module; 400. Second construction module; 500. Control module. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, embodiments of this invention, and should not be construed as limiting the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. In the description of this invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0060] The embodiments of the present invention are described below with reference to the figures.

[0061] like Figure 1 As shown, the present invention provides a method for dehumidification and condensation prevention in an indoor fully cabled substation, comprising:

[0062] S1: Collect temperature and humidity data inside the power distribution station room, obtain environmental parameter data of the station room, and predict the temperature and humidity change trend inside the station room based on a time series deep learning model, and establish a dynamic temperature and humidity change model.

[0063] Furthermore, step S1 mainly includes the collection, storage, processing, and predictive analysis of temperature and humidity data within the substation, ultimately establishing a dynamic temperature and humidity change model. During the operation of the substation, condensation easily occurs on the surface of electrical equipment due to factors such as equipment heating, ambient temperature differences, and humidity changes, posing a threat to the safe operation of the equipment. To solve this problem, it is necessary to accurately monitor and predict the changing trends of temperature and humidity within the substation, thereby taking targeted anti-condensation measures.

[0064] Step S1 further includes:

[0065] S11: Real-time monitoring of key locations, including the surface of distribution cabinets, cable joints, condensation risk areas on walls, and corners of the station building, to obtain station building environmental parameter data.

[0066] Furthermore, in step S11, high-precision temperature and humidity sensors are first installed on locations such as the surface of the distribution cabinet, cable joints, condensation risk areas on the walls, and corners of the station building to collect temperature and humidity data in these areas. The surface of the distribution cabinet, as the main outer casing of electrical equipment, is easily affected by the heat generated by internal equipment and the external environment, leading to temperature differences. Cable joints, due to their complex structure, are often key areas for cold bridge formation. Condensation risk areas on the walls typically refer to areas near the exterior walls that are significantly affected by the external environment. Corners of the station building are locations with poor air circulation and where moisture easily accumulates. SHT3x series temperature and humidity sensors are installed at these locations, with accuracies of ±0.2℃ and ±1.5%RH, ensuring the accuracy of the collected data. These sensors transmit data to the central data acquisition unit via an RS485 bus using the Modbus RTU protocol, forming a multi-point real-time monitoring network. The data acquisition frequency is dynamically adjusted according to the rate of environmental change, typically once every 5 minutes in stable environments, and increased to once every 1 minute during drastic environmental changes to ensure the capture of key temperature and humidity changes. In addition to temperature and humidity, auxiliary parameters such as surface temperature and air velocity are also collected simultaneously to calculate dew point temperature and airflow distribution. After preliminary screening to remove obvious outliers, these raw data form a standardized dataset of station environmental parameters.

[0067] S12: Store and index the station building environmental parameter data through a time-series database to obtain a historical temperature and humidity dataset.

[0068] Furthermore, unlike traditional relational databases, the time-series database of this invention is optimized for time-series data, enabling efficient processing of large amounts of timestamp data. This invention uses InfluxDB as the time-series database, storing the station environmental parameter data obtained in step S11 in chronological order. Each data record includes a collection timestamp, collection location identifier, temperature value, humidity value, calculated dew point temperature value, and other auxiliary data. Data storage employs compression algorithms to reduce storage space, and a dynamic downsampling strategy based on time windows is designed. Recent data maintains its original sampling rate, while earlier data undergoes appropriate downsampling. For example, data from the last 24 hours maintains its original frequency of once every 5 minutes, data from 1-7 days ago is reduced to once every 15 minutes, and data from 7-30 days ago is reduced to once every hour, ensuring both the granularity of data during critical periods and optimizing storage efficiency. The data index adopts a multi-level index structure, first divided by station area, then grouped by sensor location, and finally a B+ tree index built by timestamp, achieving millisecond-level data retrieval performance. Through the above design, the system can quickly extract historical temperature and humidity data for specific time periods and locations, facilitating subsequent analysis. In addition, the time-series database also implements data integrity checks and interpolation processing. It repairs data gaps that may occur during the collection process through linear interpolation or spline interpolation, ensuring data continuity and ultimately forming a complete historical temperature and humidity dataset.

[0069] S13: Based on the predicted future meteorological changes and the historical temperature and humidity dataset, predict the temperature and humidity change trends to obtain the predicted temperature and humidity change trends.

[0070] Step S13 further includes:

[0071] S131: Use an LSTM neural network to extract features from the historical temperature and humidity dataset to obtain time-series features of temperature and humidity changes.

[0072] Furthermore, in step S131, this invention employs an LSTM (Long Short-Term Memory) neural network to extract features from the historical temperature and humidity dataset. LSTM is a variant of recurrent neural networks, specifically designed to handle and predict long-term dependencies in time series data. In this invention, the input to the LSTM network is a standardized historical temperature and humidity time series, including the temperature, humidity, and calculated dew point temperature for each monitoring point. The network structure includes a 128-neuron LSTM layer, a 64-neuron dense layer, and an output layer. The LSTM layer is responsible for capturing time series features, extracting temperature and humidity change patterns, periodic features, and abrupt changes from the historical data. Specifically, the historical data is divided into sliding segments with a 24-hour time window, sliding forward one hour at a time, using the data from the past 24 hours to predict the value at the next moment, and training the network to learn the temperature and humidity change patterns.

[0073] S132: Obtain weather forecast information through external meteorological data API to obtain future weather change prediction data.

[0074] Step S132 obtains weather forecast information via an external meteorological data API. Temperature and humidity changes within the substation are influenced not only by the operating status of internal equipment but also by external meteorological conditions. Therefore, accurately predicting temperature and humidity trends within the substation requires the introduction of external meteorological data. This invention obtains 1-3 day weather forecast data for the substation's location by calling a meteorological service API, including elements such as temperature, humidity, air pressure, precipitation, wind speed, and wind direction. The API interface adopts a RESTful architecture, obtaining structured meteorological data in JSON format via HTTP requests. The system automatically updates the forecast data every 3 hours to obtain the latest weather forecast information. The obtained raw meteorological data undergoes format conversion and time alignment processing to maintain consistency with the timestamps of the monitoring data within the substation, forming a standardized dataset for predicting future weather changes, preparing for subsequent data fusion.

[0075] S133: The time series characteristics are fused with the future meteorological change prediction data using multivariate time series analysis to obtain comprehensive prediction input parameters.

[0076] Furthermore, step S133 aims to combine the features extracted from the historical data inside the station with external meteorological forecast data to form a more comprehensive prediction basis. Specifically, the Dynamic Time Warping (DTW) algorithm is used to analyze the time delay relationship between internal and external data, determining the time lag of the impact of external meteorological changes on the station's internal environment. Based on the calculated DTW distance matrix, a time-series correlation model between external meteorological variables and internal temperature and humidity parameters is established to quantitatively analyze the influence coefficients of external factors such as temperature and humidity on corresponding parameters within the station. Simultaneously, power distribution load data is introduced as an additional feature to establish a multi-dimensional feature space. Principal component analysis (PCA) is used to reduce the dimensionality of the feature space, removing redundant features and retaining key information. Finally, the processed historical feature vectors are combined with future meteorological forecast data aligned to the time axis according to a determined fusion rule to form comprehensive prediction input parameters for forecasting. The fusion rule uses a weighted average method, allocating the contribution of different features according to the influence weights of each factor in the historical data; the weights are optimized using gradient descent.

[0077] S134: The comprehensive prediction input parameters are learned by a deep learning prediction model to obtain the prediction results of temperature and humidity change trends.

[0078] Further, step S134 uses a deep learning prediction model to learn the comprehensive prediction input parameters, ultimately obtaining the predicted results of temperature and humidity change trends. This step employs a sequence-to-sequence (Seq2Seq) prediction model, which consists of an encoder-decoder architecture specifically designed for time series prediction. The encoder uses a bidirectional LSTM network, capable of simultaneously considering past and future contextual information; the decoder uses an LSTM network with an attention mechanism, able to allocate more attention to important parts of the input sequence. The model input is the comprehensive prediction input parameters generated in step S133, and the output is the predicted temperature and humidity sequences for each monitoring point within the station for the next 24 hours. Model training uses supervised learning with historical datasets, employing mean squared error (MSE) as the loss function, and updating parameters using the Adam optimizer. To enhance the model's generalization ability, dropout regularization and early stopping strategies are introduced during training to prevent overfitting. After model training, the prediction accuracy is verified using a test set, with the mean absolute error for temperature prediction controlled within 0.5℃ and the mean absolute error for humidity prediction controlled within 3%. The final output includes the predicted temperature and humidity values ​​for each monitoring point over the next 24 hours and their 95% confidence intervals, forming a prediction result of temperature and humidity change trends.

[0079] S14: Analyze the predicted results of temperature and humidity change trends using the condensation risk assessment algorithm to obtain dynamic identification information of areas prone to condensation within the station building, and obtain a complete dynamic temperature and humidity change model.

[0080] Step S14 utilizes a condensation risk assessment algorithm to analyze the predicted temperature and humidity trends, obtaining dynamic identification information of condensation-prone areas within the station building and constructing a complete dynamic temperature and humidity change model. The condensation risk assessment algorithm is based on dew point temperature theory, assessing condensation risk by calculating the difference between the surface temperature and the air dew point temperature at each monitoring point (temperature difference). When the surface temperature is below or close to the dew point temperature, condensation easily occurs on the surface. The algorithm first calculates the dew point temperature change curve for the next 24 hours based on the predicted temperature and humidity data, then compares it with the predicted surface temperature curve to calculate the minimum temperature difference and its occurrence time. Based on the size of the temperature difference, the risk level is divided into four levels: safe (temperature difference > 5℃), caution (2℃ < temperature difference ≤ 5℃), warning (0℃ < temperature difference ≤ 2℃), and danger (temperature difference ≤ 0℃). For each monitoring point, the algorithm generates a risk level time series diagram for the next 24 hours and identifies the peak risk period. Through a spatial interpolation algorithm, the risk assessment results of discrete monitoring points are extended to the entire station building space, generating a three-dimensional risk distribution heat map that visually displays condensation-prone areas and their risk change trends. Finally, based on the risk analysis results and historical condensation event records, a risk prediction model was established using machine learning classification algorithms (such as random forests) to further improve the accuracy of risk identification. These analysis results together constitute a complete dynamic temperature and humidity change model, including core elements such as the spatiotemporal distribution information of temperature and humidity, trend prediction, and condensation risk assessment.

[0081] In a specific embodiment using a power distribution station as an example, the station has an area of ​​approximately 40 square meters and contains 6 sets of high-voltage cable joints, 4 medium-voltage switchgear cabinets, and 2 distribution transformers. First, temperature and humidity sensors are installed at 20 key locations within the station, collecting data every 5 minutes. Next, the collected data is stored in a time-series database. After 30 days of accumulation, a historical temperature and humidity dataset containing 17,280 records (20 points × 12 times / hour × 24 hours × 3 days) is formed. Subsequently, the LSTM network used in step S131 is applied to extract features from this dataset. During network training, after 200 iterations, the loss value decreased from the initial 0.025 to 0.006, successfully capturing the diurnal variation patterns of temperature and humidity (high temperature and low humidity during the day, and low temperature and high humidity at night) and the impact patterns of equipment load changes on the environment. Step S132 obtains the weather forecast for the area for the next 48 hours. The forecast indicates that a cold air mass will pass through, causing the external temperature to drop by 8°C and the relative humidity to increase by 20% within 12 hours. Step S133 fuses the extracted time-series features with meteorological forecast data. DTW algorithm analysis determines that the time lag of external meteorological changes on the station's internal temperature is approximately 3 hours, and the time lag on humidity is approximately 2 hours. Step S134 uses a Seq2Seq model to predict temperature and humidity changes at various points within the station over the next 24 hours based on the fused data. The prediction results show that under the influence of cold air, the surface temperature of the cable joint area will drop to 12℃ between 2 AM and 6 AM the following day, while the humidity in the same area will rise to 85% during the same period, resulting in a dew point temperature of approximately 9.5℃. Finally, the condensation risk assessment algorithm calculates that the minimum difference between the surface temperature and dew point temperature in the cable joint area is 2.5℃, occurring at 4 AM, with a risk level of "Caution." The minimum surface temperature difference near the distribution cabinet close to the external wall is only 1.3℃, with a risk level of "Warning." The risk distribution heatmap generated through spatial interpolation clearly shows that the area near the external wall in the northeast corner of the station has the highest condensation risk and requires key protection. The obtained complete dynamic temperature and humidity change model provides precise guidance for subsequent anti-condensation measures.

[0082] S2: Based on the dynamic temperature and humidity change model and the station building environmental parameter data, establish an adaptive thermal regulation phase change material anti-condensation unit.

[0083] Step S2 further includes:

[0084] S21: Based on the dynamic temperature and humidity change model and the station building environmental parameter data, select a basic phase change material suitable for the station building temperature range.

[0085] Furthermore, phase change material (PCM) is a functional material that can absorb or release a large amount of latent heat within a specific temperature range, achieving heat storage and release through a phase change process. In step S21 of this invention, the key to selecting the basic PCM lies in determining its phase change temperature, which should match the environmental characteristics of the power distribution station. Specifically, firstly, temperature fluctuation range data for each area within the station are extracted from a dynamic temperature and humidity change model, and the annual minimum, maximum, and average temperatures are calculated. Then, surface temperature and dew point temperature data for easily condensing areas within the station are extracted, and the distribution of the difference between the two is calculated to determine the temperature regulation range required for condensation prevention. Next, a temperature-condensation risk correlation model is established through condensation frequency analysis to determine the optimal phase change temperature range. In addition to the phase change temperature, the selection criteria also consider thermophysical parameters such as latent heat value, density, specific heat capacity, and thermal conductivity. By establishing a material performance-environmental requirement matching score model, candidate materials are comprehensively scored to select the most suitable basic phase change material. The scoring model adopts a weighted summation method, and the weight allocation during scoring is based on the importance of the indicators. Specifically, in this embodiment, the phase change temperature matching degree has the highest weight, accounting for 40%, the latent heat value accounts for 25%, and other indicators account for 35%. Based on the above data analysis, this invention selects organic phase change materials with a phase change temperature between 15℃ and 30℃ as the basic material, and finally selects fatty acid compounds (palmitic acid) as the basic phase change material.

[0086] S22: Add nanoscale modifiers to the basic phase change material to optimize its performance and obtain a modified phase change material.

[0087] Specifically, in step S22, 5-10% of nano-carbon material is first added to the basic phase change material to obtain a thermally optimized phase change material; then, 3-5% of crosslinking agent is added to the thermally optimized phase change material to obtain a structurally stable phase change material; finally, 1-3% of flame retardant is added to the structurally stable phase change material to obtain a modified phase change material.

[0088] Furthermore, nano-carbon materials possess extremely high thermal conductivity, and their addition to phase change materials can significantly improve their thermal conductivity. Crosslinking agents are compounds that can form chemical bonds between molecules, connecting molecular chains into a three-dimensional network structure. Commonly used crosslinking agents include polyfunctional alcohols, polyfunctional acids and their anhydrides, epoxy resins, etc. The purpose of adding crosslinking agents to thermally optimized phase change materials is to enhance the shape stability of the material and prevent material loss during the liquid phase change state. Flame retardants are additives that can inhibit or delay the combustion of materials. Commonly used flame retardants include halogenated flame retardants, phosphorus-based flame retardants, inorganic flame retardants, etc. Considering the special environment of power distribution substations, this invention selects low-smoke, halogen-free, and low-toxicity environmentally friendly flame retardants. This embodiment selects a nitrogen-phosphorus synergistic flame retardant to minimize the impact on phase change performance while ensuring flame retardant effect.

[0089] S23: The modified phase change material is encapsulated in microcapsules with a multi-layer structure design to obtain a phase change material anti-condensation unit.

[0090] Step S23 involves microencapsulating the modified phase change material using a multi-layered structure design to obtain an anti-condensation unit. Microencapsulation refers to encapsulating the phase change material in tiny capsules to form a core-shell structure, thereby preventing leakage of the liquid phase change material and improving its stability and service life. The encapsulation process employs interfacial polymerization. First, the modified phase change material is dispersed in a continuous phase to form an emulsion. Then, a polymerization reaction occurs at the interface between the dispersed and continuous phases to form a polymer shell encapsulating the phase change material. The selected shell material must meet characteristics such as high strength, durability, good sealing, and thermal conductivity. In this invention, polyacrylate is selected as the shell material.

[0091] The aforementioned multi-layer structure design refers to the microcapsule employing a multi-shell structure, from the inside out: the inner layer, which directly contacts the phase change material, uses a high-density polyethylene film to provide basic sealing; the middle layer uses aluminum foil to improve thermal conductivity and barrier properties; and the outer layer uses moisture-proof and flame-retardant engineering plastics to provide mechanical strength and protection. The multi-layer structure is achieved through a layer-by-layer assembly process. The inner layer is formed using an emulsification-solvent evaporation method, the middle layer is prepared using a vacuum aluminizing process, and the outer layer is completed using a dip-coating or spray-coating process.

[0092] S3: Based on the information of condensation-prone areas identified in the dynamic temperature and humidity change model and the operating status data of the ventilation system, the phase change material anti-condensation unit is deployed at the anti-condensation location of the power distribution room to obtain an anti-condensation network, and the working status of the anti-condensation network is monitored in real time to obtain anti-condensation effect data.

[0093] In step S3, the step of deploying the phase change material anti-condensation unit at the anti-condensation location of the power distribution room based on the information of condensation-prone areas identified in the dynamic temperature and humidity change model and the operating status data of the ventilation system to obtain the anti-condensation network further includes:

[0094] S31: Perform data mining on the temperature and humidity distribution data in the dynamic temperature and humidity change model to identify areas prone to condensation and obtain a location map of these areas.

[0095] Further, in step S31, areas prone to condensation are first identified, and a location map of these areas is obtained. In this step, firstly, the three-dimensional information of each monitoring point, along with the corresponding temperature, humidity, and calculated dew point temperature data, are extracted from the dynamic temperature and humidity change model to form a multi-dimensional data matrix. Secondly, the difference between the surface temperature and the dew point temperature (temperature difference) is calculated, denoted as ΔT. When ΔT is close to or less than zero, it indicates a high risk of condensation in the area. Thirdly, the K-means clustering algorithm is applied to analyze the temperature difference data. The core idea of ​​this algorithm is to divide n data points into k clusters, with each data point belonging to the cluster represented by its nearest cluster center. The specific calculation process is as follows: randomly select k initial cluster centers; calculate the distance from each data point to each cluster center and assign it to the nearest cluster; recalculate the center point of each cluster; repeat the above steps until the cluster centers no longer change significantly. Through cluster analysis, the area within the station is divided into different categories based on the condensation risk characteristics.

[0096] After completing the above analysis, this invention uses spatial interpolation technology, specifically Kriging interpolation, to extend the risk assessment results of discrete monitoring points to the entire station space, and finally generates a complete heat map of condensation risk distribution within the station, namely, a condensation-prone area location map, which can represent the condensation risk level of different areas and intuitively display the spatial distribution of condensation-prone areas.

[0097] S32: Computational fluid dynamics simulation is used to process the operating status data of the ventilation system, generate the airflow path and velocity distribution model in the station building, and obtain the results of air circulation dead zone identification.

[0098] Step S32 uses computational fluid dynamics (CFD) simulation to process the operating status data of the ventilation system, generating an airflow path and velocity distribution model within the station building, and obtaining the results of air circulation dead zone identification. Specifically, in this step, a three-dimensional geometric model of the power distribution station building is first established, including information such as the station building structure, equipment layout, and ventilation outlet locations. Then, a computational mesh is generated based on the geometric model, using either a structured or unstructured mesh, with the mesh density increased in areas of drastic flow changes. Next, boundary conditions and initial conditions are set. Boundary conditions include inlet air velocity, outlet pressure, and no wall slip, while initial conditions are the initial temperature field and velocity field.

[0099] CFD simulation can obtain the three-dimensional airflow field inside the station building, including the velocity vector field, pressure field, and temperature field. Based on the simulation results, the airflow path and velocity distribution are calculated to identify air circulation dead zones. A velocity threshold is defined; in this embodiment, it is set to 20% of the average velocity in the station building. When the velocity in a region is below the threshold and the flow direction changes frequently, it is determined to be an air circulation dead zone. The mathematical expression is as follows:

[0100]

[0101] Where D represents the set of dead zones, (x,y,z) are the coordinates of a point in space, and v(x,y,z) is the velocity of the flow at that point in space. Velocity divergence represents the expansion or compression rate of a fluid. When it is close to zero, it means that the fluid volume is almost constant, corresponding to a stagnant flow state. The dead zone areas of air circulation identified by the above method are used to generate a dead zone distribution map, which marks the areas where airflow is not smooth in the station building. These areas are difficult to expel moisture and are potential high-risk areas for condensation.

[0102] S33: Based on the risk rating method, the location map of the condensation-prone area and the identification results of the air circulation dead zone are comprehensively overlaid and analyzed to determine the high-risk area, medium-risk area and low-risk area, and obtain a graded anti-condensation layout scheme.

[0103] Step S33 involves comprehensively overlaying the condensation-prone area location map with the air circulation dead zone identification results using a risk rating method to determine high-risk, medium-risk, and low-risk areas, thus obtaining a graded anti-condensation layout plan. Risk rating refers to the process of classifying risks based on their probability of occurrence and the severity of their consequences. In this step, the risk matrix method is used for risk rating. This method locates risk factors in a matrix according to their probability of occurrence and degree of impact, thereby determining the risk level.

[0104] S34: According to the graded anti-condensation arrangement scheme, the phase change material anti-condensation unit is arranged at the anti-condensation position of the power distribution room to obtain an anti-condensation network.

[0105] Furthermore, based on risk zoning, a layout strategy for phase change material anti-condensation units was formulated. High-risk areas adopted high-density layout with a coverage rate of over 90%; medium-risk areas adopted medium-density layout with a coverage rate of 50-70%; and low-risk areas adopted low-density layout with a coverage rate of approximately 30%. Simultaneously, appropriate forms of anti-condensation units were selected according to the characteristics of different areas: panel-shaped units were used on walls, flexible wrapping units were used at cable joints, and box-shaped units were used at the bottom of distribution cabinets, forming a comprehensive and reasonable hierarchical anti-condensation layout scheme. Finally, according to the hierarchical anti-condensation layout scheme, the phase change material anti-condensation units were deployed at the anti-condensation locations in the substation.

[0106] During installation, it is essential to ensure good thermal contact between the anti-condensation unit and the protected surface. If necessary, thermally conductive silicone grease should be used to fill the contact gaps. For areas where direct contact is not possible, such as internal equipment spaces, heat conduction channels should be designed to transfer the cooling effect of the anti-condensation unit to the target area. After installation, the locations of the anti-condensation units should be recorded and marked for easy monitoring and maintenance later. Through these installation steps, an anti-condensation network covering key areas of the station building is formed. This network can automatically absorb or release heat according to temperature changes, actively regulating the local environment and preventing condensation.

[0107] Specifically, step S3, which involves real-time monitoring of the working status of the anti-condensation network, includes:

[0108] The surface temperature of the phase change material anti-condensation unit is monitored by a miniature temperature sensor to obtain the operating temperature of the anti-condensation unit; the temperature distribution of the phase change material anti-condensation unit is scanned using infrared thermal imaging technology to obtain data on heat diffusion effect; the heat transfer between the phase change material anti-condensation unit and the environment is measured by a heat flow sensor to obtain energy exchange information of the phase change process; and the condensation condition of the protected surface is detected by a surface humidity detector to obtain anti-condensation surface status data.

[0109] Miniature temperature sensors are used to monitor the surface temperature of the phase change material (PCM) anti-condensation unit to obtain its operating temperature. Thermistor or thermocouple temperature sensors with dimensions smaller than 5mm × 5mm are selected and installed at critical locations on the PCM surface. The number and location of sensors are determined based on the size and importance of the PCM unit; typically, 2-3 sensors are installed on each high-risk area, 1-2 on medium-risk areas, and fewer on low-risk areas. The sensor sampling frequency is set to once every 5 minutes. When a rapid temperature change is detected (rate of change exceeding 0.5℃ / min), the sampling frequency is automatically increased to once every 1 minute. Temperature data is transmitted wirelessly to a central monitoring system, which records the temperature change curve of the PCM unit in real time. Based on the phase change temperature characteristics of the PCM material, it is determined whether the unit is in a phase change state: if the unit temperature remains near the phase change temperature (±0.5℃) for a period of time, it indicates that the unit is absorbing or releasing latent heat and is in an active operating state.

[0110] Specifically, in this invention, an infrared thermal imager is used to periodically scan the anti-condensation network. The thermal imager has a resolution of no less than 320×240 pixels, a temperature sensitivity better than 0.05℃, and a scanning frequency of once per hour, increasing to once every 30 minutes when the ambient temperature changes rapidly. After radiometric and geometric correction, the thermal image data is converted into a temperature distribution map. The characteristics of the temperature distribution map are analyzed using image processing algorithms. Finally, the temperature distribution maps of two consecutive scans are compared to calculate the temperature change rate field and evaluate the dynamic response performance of the anti-condensation unit.

[0111] Heat flux sensors are used to measure the heat transfer between the anti-condensation unit of a phase change material and the environment, obtaining information on energy exchange during the phase change process. A heat flux sensor is a device used to measure heat flux density (heat flow rate per unit area), and common types include thermoelectric and thermal gradient types. Heat flux sensors are installed on both the inner surface (closer to the protected equipment) and the outer surface (facing the environment) of the anti-condensation unit. The sensor size is 10mm × 10mm, and the sensitivity is better than 0.01W / (m²). 2 • K). The sensor data acquisition frequency is synchronized with the temperature sensor. By measuring the heat flux density value and combining it with the sensor area, the heat flow rate is calculated. The difference in heat flow rate between the inner and outer surfaces is the heat stored or released by the anti-condensation unit. A positive value indicates that the unit is absorbing heat (during the ambient temperature rise phase), and a negative value indicates that the unit is releasing heat (during the ambient temperature fall phase). Finally, the total heat exchange within a complete cycle is calculated by integration to evaluate the energy regulation capability of the anti-condensation unit.

[0112] A surface humidity detector is a sensing device that directly detects the condensation state of a surface. This invention involves installing a small surface humidity detector (less than 10mm x 10mm) on the surface of equipment prone to condensation to minimize interference with equipment operation. The detector outputs a continuous signal (surface humidity percentage), with a sampling frequency set to once every 10 minutes. When a humidity change trend is detected, the frequency is automatically increased to once every 2 minutes. The surface humidity data directly verifies the anti-condensation effect and identifies areas with insufficient anti-condensation measures.

[0113] In one specific embodiment, the station building has an area of ​​60 square meters and contains 8 sets of high-voltage cable joints, 6 medium-voltage switchgear cabinets, and 3 distribution transformers. First, through data mining of a dynamic temperature and humidity change model, three main condensation-prone areas were identified: the northeast corner near the outer wall, where the minimum temperature difference is 0.8℃, occurring daily between 3-5 AM; the cable inlet / outlet, where the minimum temperature difference is 1.5℃, occurring daily between 4-6 AM and during rainy weather; and the bottom area of ​​the distribution cabinets, where the minimum temperature difference is 2.2℃, occurring irregularly and mainly related to equipment load changes. A condensation-prone area location map generated by Kriging interpolation shows that approximately 15% of the station building is a high-condensation-risk area. Then, in step S32, CFD simulation was used to analyze the airflow within the station building under ventilation conditions with a wind speed of 0.5 m / s. Simulation results show that a significant airflow dead zone formed in the northeast corner, with a flow velocity below 0.1 m / s; the narrow space between the back of the equipment and the distribution cabinet also exhibited poor airflow, with an average flow velocity of 0.15 m / s; while the central passage of the station building had unobstructed airflow, with a flow velocity of 0.4-0.5 m / s. Subsequently, condensation risk and ventilation risk were comprehensively scored, with weights set at 0.6 and 0.4 respectively. The calculation results show that the northeast corner area received a comprehensive risk score of 8.5, classifying it as a high-risk area; the cable inlet / outlet scored 7.2, also classifying it as a high-risk area; the bottom of the distribution cabinet scored 5.8, classifying it as a medium-risk area; and the central passage of the station building scored 2.1, classifying it as a low-risk area. Based on this classification result, an anti-condensation deployment plan was formulated: In the high-risk area (12 square meters), 24 500mm×500mm plate-shaped anti-condensation units were deployed, achieving a coverage rate of 95%; in the medium-risk area (20 square meters), 28 anti-condensation units of the same specification were deployed, achieving a coverage rate of 70%; and in the low-risk area (28 square meters), 14 anti-condensation units of the same specification were deployed, achieving a coverage rate of 30%. In step S34, the anti-condensation units were installed according to the plan. For the northeast corner wall, light steel keel was used to fix the plate-shaped units; for cable joints, flexible wrapping units were used to tightly adhere to the joints; and for the bottom of the distribution cabinet, box-shaped units were placed. After installation, 60 miniature temperature sensors, 2 infrared thermal imagers, 24 heat flow sensors, and 30 surface humidity detectors were deployed, forming a comprehensive monitoring network. Operational data showed that during a cold air intrusion, when the ambient temperature dropped from 25°C to 12°C, the surface temperature of the anti-condensation unit dropped to 18.2°C and stabilized for about 3 hours, proving that the phase change material was releasing latent heat; the peak heat release power measured by the heat flow sensor was 45W per square meter; the surface humidity detector did not detect any condensation phenomenon throughout the process, verifying the effectiveness of the anti-condensation network.

[0114] S4: Based on the dynamic temperature and humidity change model and the anti-condensation effect data, establish an active anti-condensation protection mechanism based on PTC heating elements.

[0115] Step S4 primarily involves establishing an active anti-condensation mechanism based on PTC heating elements, as a supplement to the passive anti-condensation network, to ensure effective prevention of condensation even under extreme conditions. PTC (Positive Temperature Coefficient) heating elements are thermistor elements whose resistance increases with temperature, giving them a self-limiting temperature function. This prevents overheating and enables precise temperature control, making them particularly suitable for anti-condensation and dehumidification applications.

[0116] S4 further includes:

[0117] S41: Based on the dynamic temperature and humidity change model and the anti-condensation effect data, integrate PTC heating elements in high-risk condensation areas to obtain the basic heating element layout.

[0118] In step S41, based on the dynamic temperature and humidity change model and anti-condensation effect data, PTC heating elements are integrated in high-risk condensation areas to obtain the basic heating element layout. First, key data is extracted from the dynamic temperature and humidity change model, especially the difference between surface temperature and dew point temperature (referred to as temperature difference). Temperature difference is a direct indicator for judging condensation risk; when the temperature difference is less than or equal to zero, condensation will occur on the surface. The data processing procedure statistically analyzes the temperature difference data of all monitoring points within the station, including the minimum temperature difference value, the frequency of temperature differences below the threshold, and the temporal distribution characteristics of the temperature difference. A temperature difference of 2℃ is typically set as the risk threshold; values ​​below this value are considered to indicate a significant risk of condensation.

[0119] The risk index (RI) for each monitoring point is calculated. The risk index is a composite indicator that comprehensively considers both temperature difference and duration. In the calculation, the reciprocal of the temperature difference at each time point is multiplied by its weighting coefficient, and then summed. The risk index increases rapidly when the temperature difference approaches or falls below zero. Critical periods, such as the early morning low-temperature period, are given higher weights to reflect their greater condensation risk. Based on the calculated risk index, all areas within the station are ranked to determine the area with the highest condensation risk.

[0120] Then, the anti-condensation effect data was analyzed to assess the match between the operating status of existing phase change material anti-condensation units and anti-condensation requirements. The anti-condensation effect data included the anti-condensation unit's operating temperature, heat diffusion effect data, energy exchange information during the phase change process, and anti-condensation surface condition data. Energy exchange data measured by heat flow sensors was used to calculate the anti-condensation capacity gap for each region. The calculation method involved integrating the difference between the heat flow required to maintain the surface temperature above the dew point temperature and the actual heat flow provided by the phase change material during the insufficient anti-condensation period, yielding the energy gap value in joules.

[0121] Based on risk index and energy deficit data, areas requiring PTC heating element installation are identified. Priority is given to areas meeting the following criteria: risk index within the top 30% of the station building, significant energy deficit, high equipment importance, or severe condensation consequences. For each selected area, the required PTC heating element power density is calculated by dividing the energy deficit by the area area and the cumulative time of the deficit, then multiplying by a safety factor (typically 1.2-1.5) to obtain the required power per unit area (W / m²). 2 Select a suitable PTC heating element based on the calculation results, taking into account factors such as rated power, operating voltage, size and shape, and protection level.

[0122] Next, this invention designs the basic heating element layout, adhering to the following principles: matching the heat load distribution, with higher heating density in high-risk areas; maintaining heating uniformity to avoid localized overheating or underheating; avoiding sensitive parts of electrical equipment to reduce electromagnetic interference; facilitating installation and maintenance; and optimizing cable arrangement to reduce wire consumption. After the layout design is completed, the heating effect is verified through numerical thermal simulation to ensure that the anti-condensation requirements are met. The numerical simulation uses the finite element analysis method to establish a composite heat conduction model including the substrate, phase change material, and PTC element, solving for the temperature field distribution and checking for any areas with excessively low temperatures or uneven temperature distribution.

[0123] S42: Combining the regional temperature distribution based on thermal imaging technology, optimize the layout of the basic heating element to obtain an optimized heating element layout.

[0124] Step S42 combines regional temperature distribution based on thermal imaging technology to optimize the layout of basic heating elements, resulting in an optimized heating element layout. Thermal imaging technology is a non-contact temperature measurement method that can visualize the temperature distribution on an object's surface. In this step, an infrared thermal imager is used to scan the substation with the anti-condensation network installed to obtain a temperature distribution image under actual operating conditions. The raw data collected by the thermal imager is an infrared radiation intensity matrix, which needs to be converted into an accurate temperature matrix through a series of processing steps.

[0125] The data processing flow first performs geometric correction to eliminate the effects of lens distortion and shooting angle, ensuring that the spatial position in the image matches the actual position. Geometric correction uses a reference point matching method, marking reference points with known coordinates in the thermal image and correcting the image through a coordinate transformation matrix. Next, radiometric correction is performed to compensate for the effects of environmental radiation and differences in material emissivity. Radiometric correction must consider the surface emissivity values ​​of different materials: typically 0.1-0.3 for metal surfaces and 0.8-0.95 for insulating material surfaces.

[0126] Then, the corrected thermal image is processed for image analysis to extract key features. The processing steps include: image filtering, using Gaussian filtering or median filtering to reduce noise; temperature gradient calculation, using Sobel or Prewitt operators to calculate the temperature gradient field; region segmentation, using threshold segmentation or watershed algorithms to segment the image into different temperature zones; and feature point extraction, identifying temperature extreme points, rapidly changing regions, and temperature stable regions.

[0127] For thermal images acquired at multiple time points, time series analysis is performed, including: calculating the temperature-time curve for each pixel; Fourier analysis to extract the periodic characteristics of temperature changes; trend analysis to identify regions with rising or falling temperature trends; and anomaly detection to identify regions with abnormal temperature fluctuations. The purpose of time series analysis is to identify dynamic thermal characteristics, distinguish regions with different levels of thermal inertia, and provide a basis for optimizing heating control strategies.

[0128] The thermal imaging analysis results are compared with the basic heating element layout in step S41 to identify any mismatches. Common mismatches include: hot spots, i.e., areas in the thermal image where the temperature is significantly higher than the surrounding area, which may indicate an excessive number of heat sources and an overly dense arrangement of heating elements; cold spots, i.e., areas where the temperature remains consistently low, indicating insufficient heating element coverage; areas with excessive temperature gradients, where unreasonable element spacing leads to uneven heat distribution; and areas with drastic temperature changes, which may require special control strategies.

[0129] Based on comparative analysis, the layout of the basic heating elements was optimized and adjusted. The adjustment methods included: reducing the number of heating elements or lowering the power density in hot spots; increasing the element density or selecting higher-power elements in cold spots; adjusting the element spacing in areas with large temperature gradients to achieve a more uniform heat distribution; and using special layouts, such as gradient or layered arrangements, in areas with drastic temperature changes. The optimization process employed an iterative method, verifying the effect through thermal simulation after each adjustment until the optimal layout was achieved.

[0130] The optimized layout considers not only spatial distribution but also control zoning. Regions with similar thermal characteristics are grouped into the same control zone, with each zone configured with an independent temperature sensor and control loop to achieve refined control. Control zone division is based on cluster analysis, using K-means or hierarchical clustering algorithms to group regions with similar temperature feature vectors into one category. Temperature feature vectors include multi-dimensional features such as average temperature, temperature fluctuation amplitude, heating rate, and cooling rate.

[0131] S43: Power control is applied to the PTC heating elements in the optimized heating element layout to form an active anti-condensation protection mechanism based on the PTC heating elements.

[0132] Step S43 performs power control on the PTC heating elements in the optimized heating element layout to form an active anti-condensation protection mechanism based on the PTC heating elements. Power control is the key to ensuring the efficient and accurate operation of the heating system, involving control strategy design, regulation method selection and safety protection mechanism establishment.

[0133] First, define the control objective: maintain the temperature of the protected surface above the dew point temperature by a certain margin (usually 2-3℃) while minimizing energy consumption. This can be expressed as a mathematical optimization problem, namely, minimizing the time integral of the power function under the constraint that the surface temperature is always above the dew point temperature plus a safety margin.

[0134] The control strategy employs a hierarchical structure, comprising three levels: prediction, decision-making, and execution. The prediction level utilizes a dynamic temperature and humidity change model to predict the trends in ambient temperature and humidity, as well as surface temperature, over a future period (typically 1-6 hours). Based on the prediction results and the current state, the decision-making level formulates the optimal control plan, including start-up and shutdown timings, power curves, and control modes. The execution level is responsible for implementing the control plan, adjusting heating power in real time, and responding to environmental changes.

[0135] The specific control algorithm adopts the Model Predictive Control (MPC) framework. MPC is an advanced control strategy capable of handling multivariable systems, satisfying constraints, and optimizing control objectives. The core idea of ​​MPC is to predict the system response over a future period using a system model in each control cycle, solve for the optimal control sequence, but only execute the first control action in the sequence. Then, the prediction window slides, and the process is repeated. The control objective function includes two weighted terms: temperature tracking accuracy and energy consumption. The anti-condensation effect and energy consumption are balanced by adjusting the weighting coefficients.

[0136] To achieve precise power control, PWM (Pulse Width Modulation) technology is employed. PWM controls the average power output by adjusting the on / off time ratio (duty cycle) of the voltage. The basic parameters of PWM control include the carrier frequency and duty cycle resolution. The carrier frequency is selected within the 20-40kHz range to avoid the range of human hearing and reduce electromagnetic interference; the duty cycle resolution is set to 10 bits (0.1%) to ensure accurate power regulation. The PWM control signal is output to the PTC heating element through a solid-state relay or MOSFET drive circuit.

[0137] Based on power control, a multi-level safety protection mechanism is established: over-temperature protection, which automatically reduces power or cuts off power when the temperature of the PTC element or the protected surface exceeds a preset safety threshold; over-current protection, which monitors the operating current of the PTC element in real time through a current sensor, and immediately cuts off power and alarms when an abnormal current is detected; leakage protection, which is equipped with a leakage current protector, which cuts off the power supply within 30ms when the system grounding current exceeds a safe value (usually 30mA); and abnormal operation detection, which determines whether the PTC element is working properly by comparing the actual temperature rise with the theoretical temperature rise.

[0138] The PTC heating system works in synergy with the passive anti-condensation network and ventilation system. The control strategy employs a three-tiered anti-condensation scheme: the first tier is ventilation and dehumidification, suitable when the outside air humidity is lower than the indoor air humidity; the second tier is passive temperature regulation using PTC, suitable when temperature fluctuations fall within the PTC's operating range; and the third tier is active PTC heating, activated only when the first two tiers fail to meet anti-condensation requirements. This tiered strategy is intelligently switched via a control algorithm, whose decisions are based on real-time environmental data and predictive models, comprehensively considering both anti-condensation effectiveness and energy consumption.

[0139] In one specific embodiment, the station building has an area of ​​70 square meters and contains 10 sets of high-voltage cable joints, 8 medium-voltage switchgear cabinets, and 4 distribution transformers. In step S41, analysis of dynamic temperature and humidity change model data revealed that the minimum temperature difference in the cable joint area dropped to 0.6℃, with a risk exposure time ratio of 25%; the minimum temperature difference at the bottom of the distribution cabinets was 1.2℃, with a risk exposure time ratio of 18%; while the temperature difference near the outer wall frequently fell below 2℃, with a risk exposure time ratio as high as 32%. These data were obtained through statistical analysis of temperature and humidity sensor data collected every 5 minutes for 30 consecutive days. Analysis of the anti-condensation effect data and calculation of the energy gap revealed that the energy gap in the cable joint area reached 35 Wh / m² under extreme weather conditions, 22 Wh / m² at the bottom of the distribution cabinets, and 42 Wh / m² near the outer wall. These energy gap data were obtained by integrating the difference between the actual heat flow measured by the heat flow sensor and the theoretically required heat flow during periods with a temperature difference below 2℃.

[0140] Based on the above data, the required power for the PTC heating elements was calculated as follows: 200 watts / square meter for the cable joint area, 150 watts / square meter for the bottom of the distribution cabinet, and 250 watts / square meter for the exterior wall area. PTC heating elements with a rated power of 50 watts and dimensions of 100mm × 200mm were selected. An initial layout was designed according to the area shape, with a total of 48 PTC heating elements covering a total area of ​​approximately 15 square meters. In step S42, thermal imaging analysis revealed three areas of uneven heat distribution in the initial layout: a cold spot of approximately 2°C appeared at the corner of the exterior wall; a thermal bridge formed between two adjacent cable joints, with a temperature 3.5°C higher than the surrounding area; and a large temperature gradient at the bottom of the distribution cabinet, with a maximum difference of 4°C. These thermal imaging data were collected by an infrared thermal imager with a resolution of 640×480 pixels at three time points: 6:00 AM, 12:00 PM, and 10:00 PM, and were processed after geometric and radiometric correction.

[0141] To address the identified issues, the layout was adjusted as follows: two heating elements were added at the corners of the exterior walls; the number of elements in the thermal bridge area was reduced from four to two, and their arrangement was changed to a spaced-out configuration; a variable-density layout was adopted at the bottom of the distribution cabinet, increasing the element density in the cold zone, changing the original evenly distributed 10 elements to 8 in the cold zone and 2 in the hot zone. The optimized layout uses a total of 50 PTC heating elements. Verification using a finite element thermal analysis model showed a significant improvement in temperature uniformity, with simulation results showing the maximum temperature difference reduced from 4℃ to approximately 1.5℃. In step S43, a three-zone independent control system was designed, with two temperature sensors and two humidity sensors configured in each zone. An improved PID control algorithm was used, and the optimal parameter combination was obtained through genetic algorithm optimization. The control system employs PWM modulation technology with a 25kHz carrier frequency to achieve power regulation with an accuracy of 0.1%.

[0142] In actual operation tests, under extreme conditions where the outdoor temperature plummeted from 15℃ to -5℃, the PTC system successfully maintained the surface temperature of critical areas at 2.5℃ above the dew point, with no condensation occurring throughout the entire process. The average power consumption was 3.2 kWh / day, and it only activated between 2 AM and 6 AM when the temperature was lowest. The test results verified the effectiveness of this active anti-condensation protection mechanism and the energy efficiency optimization design goals, providing a reliable guarantee for the safe operation of indoor fully cabled substations.

[0143] S5: Based on the data flow information of the dynamic temperature and humidity change model, the ventilation system, the anti-condensation network, and the active anti-condensation protection mechanism, a comprehensive control system is established through a multi-objective optimization algorithm, and the substation is dehumidified and prevented from condensing through the comprehensive control system.

[0144] Step S5 further includes:

[0145] S51: By integrating the data flow information of the dynamic temperature and humidity change model, the ventilation system, the anti-condensation network, and the active anti-condensation protection mechanism through the industrial controller, a system-level control platform is obtained.

[0146] In step S51, the data flow information from the dynamic temperature and humidity change model, ventilation system, anti-condensation network, and active anti-condensation protection mechanism is integrated through the industrial controller to obtain a system-level control platform. The industrial controller features strong anti-interference capability, high reliability, and good real-time performance. In this invention, a Siemens S7-1500 series PLC is selected as the core controller, combined with an industrial Ethernet communication network and a SCADA (Supervisory and Data Acquisition) system to construct a system-level control platform. Through the data flow integration process, a unified system-level control platform is formed, providing the data foundation and execution environment for the next step of multi-objective optimization control.

[0147] S52: Set optimization objectives including minimizing energy consumption, maximizing dehumidification efficiency, and minimizing temperature uniformity, and construct the objective function. Obtain the optimal control strategy through a multi-objective optimization algorithm.

[0148] The expression for the optimization objective of the multi-objective optimization algorithm in step S52 is as follows:

[0149]

[0150] Where F(x) is the optimization objective, t0 is the starting point of the time range, and t f Let λ be the endpoint of the time range, λ be the exponential decay coefficient, V(t) be the time series of the ventilation system wind speed regulation coefficient, Q(t) be the time series of the auxiliary electric heating system power regulation coefficient, x1 be the decision variable for the ventilation system wind speed regulation coefficient, x3 be the decision variable for the heating system power regulation coefficient, H(t) be the humidity time series, P(t) be the temperature distribution time series of the anti-condensation network, x2 be the decision variable for the phase change material arrangement density, and T(t) be the temperature time series. avg This represents the average temperature value.

[0151] Furthermore, the objective functions include energy consumption minimization, dehumidification efficiency maximization, and temperature uniformity minimization. The integrated control system must also satisfy the following constraints:

[0152]

[0153] Where G(x) is the set of constraints, and T max H is the upper limit of temperature. min P is the lower limit of humidity. max This is the upper limit for energy consumption.

[0154] This invention employs the NSGA-II algorithm to solve the aforementioned multi-objective optimization problem. NSGA-II is a highly efficient multi-objective optimization algorithm capable of generating a well-distributed Pareto optimal solution in a single run, making it suitable for handling complex optimization problems with multiple objectives and constraints. The final output control strategies include: the ventilation system control timing (time functions of fan speed and valve opening), the PTC heating system control timing (time functions of heating power in each zone), and a phase change material density distribution map. These control strategies serve as input instructions for the next control execution step.

[0155] S53: The integrated control system is controlled by the optimal control strategy to dehumidify and prevent condensation in the power distribution room.

[0156] In one specific embodiment, the station building has an area of ​​80 square meters and contains 12 sets of high-voltage cable joints, 10 medium-voltage switchgear cabinets, and 5 distribution transformers. Firstly, based on step S51, a Siemens S7-1515 PLC is used as the central controller, along with an ET200SP distributed I / O station and a TP1200 touchscreen, to construct a system-level control platform. Sixty-four data acquisition points are set up, including 32 temperature and humidity acquisition points, 16 heat flow acquisition points, 8 airflow acquisition points, and 8 electrical parameter acquisition points, forming a comprehensive data acquisition network. The data acquisition cycle is set as follows: temperature and humidity data once every 5 minutes, fan speed and valve opening once every 1 minute, and PTC heating power once every 30 seconds. The collected raw data is preprocessed and stored in a time-series database, generating approximately 28,800 data records per day. Through data analysis, a temperature and humidity distribution model was established inside the station building, identifying three typical condensation risk areas: the cable joint area (minimum temperature difference 0.8℃), the bottom area of ​​the distribution cabinet (minimum temperature difference 1.5℃), and the area near the outer wall (minimum temperature difference 0.6℃).

[0157] Next, the optimization objective weights were set as follows: energy consumption minimization w1 = 0.4, dehumidification efficiency maximization w2 = 0.35, and temperature uniformity minimization w3 = 0.25. Using the NSGA-II algorithm, with a population size of 100, a maximum number of iterations of 500, a crossover probability of 0.8, and a mutation probability of 0.1, the optimization calculation was run. After approximately 300 iterations, the algorithm converged, yielding 20 Pareto optimal solutions. These solutions each had a different emphasis on the three objectives. For example, solution A was optimal in terms of energy consumption (saving 25% of energy) but had poorer temperature uniformity; solution B had the best dehumidification efficiency but slightly higher energy consumption; and solution C balanced the three objectives. Considering the humid climate environment of the station, solution B, which prioritized dehumidification efficiency, was selected as the final control strategy.

[0158] The subsequent optimization of the control strategy will translate into equipment control commands. For the ventilation system, the optimal operating speed of the fan was calculated to be 1200 rpm, accounting for 75% of the rated speed; the opening degrees of the inlet and outlet valves were 60% and 80%, respectively. For the PTC heating system, the heating power for the three risk areas was determined as follows: 180 W / m for the cable joint area. 2 120W / m at the bottom of the distribution cabinet 2 220W / m near the exterior wall area 2 Using PWM modulation technology, a carrier frequency of 25kHz was set, and the PWM duty cycle for each region was calculated based on power requirements. During the monitoring process, real-time monitoring of temperature and humidity changes within the station and equipment response was conducted to ensure the system operated according to the optimization strategy.

[0159] Throughout a complete day-night operating cycle, the system automatically adjusts its control strategy based on changes in the external environment: during the day when external humidity is low, it increases ventilation to fully utilize natural ventilation for dehumidification; in the evening when temperatures drop, it gradually reduces ventilation to enhance the temperature-regulating effect of the phase change material; and in the early morning when temperatures are lowest, it activates PTC heating in high-risk areas to ensure that the surface temperature remains above the dew point. Through this intelligent and coordinated control, the temperature inside the substation is maintained within the range of 18-25℃, and the relative humidity is controlled between 50-65%, with no condensation occurring. Simultaneously, the average daily energy consumption is only 4.5 kWh, saving 65% of energy compared to traditional all-time heating methods. This example verifies the effectiveness and efficiency of the multi-objective optimization-based integrated control strategy of this invention in dehumidification and condensation prevention in indoor fully cabled substations.

[0160] like Figure 2 As shown, the present invention also provides an indoor fully cabled substation dehumidification and anti-condensation system, comprising:

[0161] Prediction module 100: Used to collect temperature and humidity data in the power distribution room, obtain environmental parameter data of the room, and predict the temperature and humidity change trend in the room based on a time series deep learning model, and establish a dynamic temperature and humidity change model.

[0162] First construction module 200: used to establish an adaptive thermal regulation phase change material anti-condensation unit based on the dynamic temperature and humidity change model and the station building environmental parameter data;

[0163] Monitoring module 300: Used to monitor the working status of the anti-condensation network deployed at the anti-condensation location in the substation in real time and obtain anti-condensation effect data;

[0164] The second construction module 400 is used to establish an active anti-condensation protection mechanism based on the PTC heating element according to the dynamic temperature and humidity change model and the anti-condensation effect data.

[0165] Control module 500: Used to control the integrated control system including the ventilation system, the anti-condensation network and the active anti-condensation protection mechanism through a multi-objective optimization algorithm, so as to dehumidify and prevent condensation in the power distribution room.

[0166] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0167] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0168] This invention aims to address the problems of air conditioners and traditional dehumidifiers consuming large amounts of electricity, especially during high-humidity seasons, resulting in high operating costs and increasing the economic burden on power companies. Traditional equipment struggles to accurately adjust to changes in ambient humidity, easily leading to over- or under-dehumidification. Over-dehumidification wastes energy, while under-dehumidification fails to meet the stringent humidity requirements of fully cabled substations, posing safety hazards. Traditional dehumidification methods struggle to maintain uniform temperature and humidity, easily causing condensation in localized areas, such as cable trenches and equipment surfaces, potentially threatening equipment insulation and lifespan. Furthermore, traditional equipment lacks intelligent control mechanisms, failing to achieve humidity monitoring, early warning, and automatic adjustment functions, requiring frequent manual operation, increasing maintenance burden, and hindering refined management. This invention comprehensively applies dynamic temperature and humidity prediction, phase change material anti-condensation, and active heating technologies, using a multi-objective optimization algorithm to achieve precise all-weather anti-condensation dehumidification in substations, significantly improving anti-condensation effectiveness while reducing energy consumption and extending equipment lifespan.

[0169] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An indoor full-cable power distribution station house dehumidification and condensation prevention method, characterized by, The method comprises the following steps: S1: Collecting the temperature and humidity data in the power distribution station, obtaining the environmental parameter data of the station, and predicting the temperature and humidity change trend based on the time series deep learning model to establish a dynamic temperature and humidity change model; S2: According to the dynamic temperature and humidity change model and the station environmental parameter data, an adaptive heat regulation phase change material condensation prevention unit is established; S3: According to the condensation prone area information identified in the dynamic temperature and humidity change model and the operation state data of the ventilation system, the phase change material condensation prevention unit is arranged at the condensation prevention position of the power distribution station to obtain a condensation prevention network, and the working state of the condensation prevention network is monitored in real time to obtain condensation prevention effect data; S4: According to the dynamic temperature and humidity change model and the condensation prevention effect data, an active condensation prevention guarantee mechanism based on PTC heating element is established; S5: Based on the data flow information of the dynamic temperature and humidity change model, the ventilation system, the condensation prevention network and the active condensation prevention guarantee mechanism, a comprehensive control system is established through a multi-objective optimization algorithm, and the power distribution station is dehumidified and condensation prevented through the comprehensive control system.

2. The dehumidification and condensation prevention method for an indoor full-cable power distribution station room according to claim 1, characterized in that, Step S1 further comprises: S11: Real-time monitoring of key positions including the surface of power distribution cabinet, cable joint, wall condensation risk area and corner of station to obtain station environmental parameter data; S12: Storing and indexing the station environmental parameter data through time series database to obtain historical temperature and humidity data set; S13: Predicting the temperature and humidity change trend based on future meteorological change prediction data and the historical temperature and humidity data set to obtain temperature and humidity change trend prediction result; S14: Analyzing the temperature and humidity change trend prediction result by using condensation risk assessment algorithm to obtain dynamic identification information of condensation prone area in the station, and obtaining a complete dynamic temperature and humidity change model.

3. The method according to claim 2, wherein the method is characterized by, Step S13 further comprises: S131: Using LSTM neural network to extract features from the historical temperature and humidity data set to obtain time series features of temperature and humidity change; S132: Obtaining weather forecast information through external meteorological data API to obtain future meteorological change prediction data; S133: Using multivariate time series analysis method to fuse the time series features with the future meteorological change prediction data to obtain comprehensive prediction input parameters; S134: Learning the comprehensive prediction input parameters through deep learning prediction model to obtain temperature and humidity change trend prediction result.

4. The method of claim 1, wherein the method is characterized by, Step S2 further comprises: S21: Selecting a basic phase change material suitable for the temperature range of the station according to the dynamic temperature and humidity change model and the station environmental parameter data; S22: Adding nano-scale modifier to the basic phase change material for performance optimization to obtain modified phase change material; S23: Multi-layer structure design of microcapsule packaging of the modified phase change material to obtain phase change material condensation prevention unit.

5. The method of claim 1, wherein the method is characterized by, In step S3, according to the condensation prone area information identified in the dynamic temperature and humidity change model and the operation state data of the ventilation system, the phase change material condensation prevention unit is arranged at the condensation prevention position of the power distribution station to obtain a condensation prevention network. S31: data mining is performed on the temperature and humidity distribution data in the dynamic temperature and humidity change model, an easy-condensation area is identified, and an easy-condensation area positioning map is obtained; S32: the operation state data of the ventilation system is processed by using computational fluid dynamics simulation, a model of air flow path and flow velocity distribution in the station house is generated, and an air circulation dead angle identification result is obtained; S33: the easy-condensation area positioning map and the air circulation dead angle identification result are comprehensively superimposed and analyzed according to a risk rating method, high-risk areas, medium-risk areas and low-risk areas are determined, and a hierarchical anti-condensation arrangement scheme is obtained; S34: according to the hierarchical anti-condensation arrangement scheme, the phase change material anti-condensation unit is arranged at the anti-condensation position of the power distribution station house, and an anti-condensation network is obtained.

6. The method of claim 1, wherein the method is an indoor full-cable power distribution station house dehumidification and condensation prevention method, characterized by, The step of monitoring the working state of the anti-condensation network in step S3 specifically includes: monitoring the surface temperature of the phase change material anti-condensation unit by using a micro temperature sensor to obtain the working temperature of the anti-condensation unit; scanning the temperature distribution of the phase change material anti-condensation unit by using infrared thermal imaging technology to obtain heat diffusion effect data; measuring the heat transfer between the phase change material anti-condensation unit and the environment by using a heat flow sensor to obtain phase change process energy exchange information; detecting the condensation situation of the protected surface by using a surface humidity detector to obtain anti-condensation surface state data.

7. The method of claim 1, wherein the method is an indoor full-cable power distribution station dehumidification and condensation prevention method, characterized by, S4 further includes: S41: integrating a PTC heating element in the high-risk condensation area according to the dynamic temperature and humidity change model and the anti-condensation effect data, and obtaining a basic heating element layout; S42: optimizing the basic heating element layout in combination with the regional temperature distribution based on thermal imaging technology, and obtaining an optimized heating element layout; S43: performing power control on the PTC heating element in the optimized heating element layout to form an active anti-condensation guarantee mechanism based on the PTC heating element.

8. The method of claim 1, wherein the method is an indoor full-cable power distribution station dehumidification and condensation prevention method, characterized by, Step S5 further includes: S51: integrating the data stream information of the dynamic temperature and humidity change model, the ventilation system, the anti-condensation network and the active anti-condensation guarantee mechanism by using an industrial controller to obtain a system-level control platform; S52: setting optimization objectives including energy consumption minimization, dehumidification efficiency maximization and temperature uniformity minimization and constructing a target function, and obtaining an optimal control strategy by using a multi-objective optimization algorithm; S53: controlling the comprehensive control system by using the optimal control strategy to dehumidify and prevent condensation in the power distribution station house.

9. The method according to claim 8, wherein the method is characterized by, The expression of the optimization objectives of the multi-objective optimization algorithm in step S52 is: wherein F(x) is the optimization objective, t0 is the start of the time range, t f is the end of the time range, λ is the exponential decay coefficient, V(t) is the time series of the ventilation system air speed adjustment coefficient, Q(t) is the time series of the auxiliary electric heating system power adjustment coefficient, x1 is the ventilation system air speed adjustment coefficient decision variable, x3 is the heating system power adjustment coefficient decision variable, H(t) is the humidity time series, P(t) is the temperature distribution time series of the anti-condensation network, x2 is the phase change material arrangement density decision variable, T(t) is the temperature time series, T avg is the average temperature value.

10. An indoor full-cable power distribution station house dehumidification and condensation prevention system, characterized in that, including: a prediction module for collecting temperature and humidity data in the power distribution station house to obtain station house environment parameter data, and predicting temperature and humidity change trends in the station house based on a time series deep learning model to establish a dynamic temperature and humidity change model; a first construction module for establishing a self-adaptive thermal regulation phase change material anti-condensation unit according to the dynamic temperature and humidity change model and the station house environment parameter data; a monitoring module for monitoring the working state of the anti-condensation network arranged at the anti-condensation position of the power distribution station house in real time to obtain anti-condensation effect data; The second construction module is used for establishing an active anti-condensation guarantee mechanism based on the PTC heating element according to the dynamic temperature and humidity change model and the anti-condensation effect data; The control module is used for controlling a comprehensive control system including a ventilation system, the anti-condensation network and the active anti-condensation guarantee mechanism by a multi-objective optimization algorithm, so as to dehumidify and prevent condensation of the power distribution station.

Images