Intelligent induction type anti-condensation system and method for power equipment

Abstract

The invention belongs to the technical field of condensation prevention of power equipment, and discloses an intelligent induction type condensation prevention system and method for the power equipment, and the system comprises a multivariate data collection unit, an intelligent control unit and an execution unit, and achieves the comprehensive real-time monitoring of the internal and external environments of the equipment through the integration of the multivariate data collection unit; the intelligent control unit can accurately predict the condensation risk based on real-time and historical data by using a pre-trained machine learning model, and dynamically adjust the condensation prevention strategy according to the condensation risk, thereby ensuring that condensation can be effectively prevented under various complex environmental conditions; the accurate execution of the execution unit further guarantees the condensation prevention effect, the operation stability and safety of the power equipment are integrally improved, the equipment fault risk caused by condensation is reduced, the service life of the equipment is prolonged, and the unnecessary maintenance cost is reduced; the system has self-learning and self-adaptive capabilities, can continuously optimize a control strategy to adapt to different environmental conditions, and ensures an optimal anti-condensation effect.

Description

Technical Field

[0001] The present invention belongs to the technical field of anti-condensation for electric equipment, and in particular relates to an intelligent induction anti-condensation system and method for electric equipment. Background Art

[0002] With the rapid development of the power industry, the power network is constantly expanding and upgrading, and the operating environment of power equipment is facing unprecedented challenges. While pursuing efficient and stable power supply, the natural environment conditions of the equipment are becoming increasingly complex and changeable, which puts higher requirements on the reliability and durability of power equipment. Especially in areas with high humidity or significant temperature difference between day and night, the condensation phenomenon inside power equipment has become a key problem that needs to be solved urgently. The condensation phenomenon is mainly caused by the temperature difference between the inside and outside of the equipment, which causes the water vapor in the air to condense into liquid water on the surface with lower temperature. Inside the power equipment, this condensation may not only directly adhere to the insulating materials, metal parts or electronic components, reducing their insulation performance and increasing the risk of leakage, but also penetrate into the equipment through penetration, causing more serious consequences. The condensation phenomenon is often accompanied by complex environmental factors such as temperature fluctuations and humidity changes. The superposition of these factors makes the harm of condensation to power equipment more difficult to predict and control.

[0003] At present, although the existing anti-condensation devices can play a certain dehumidification role, they have certain limitations, which are specifically manifested in the following aspects: First, most of the existing anti-condensation devices rely on manually set thresholds for control. When facing complex and changeable environmental conditions, it is difficult to achieve the optimal anti-condensation effect because the working parameters cannot be dynamically adjusted according to real-time data; Second, most of the existing anti-condensation devices only use a single sensor. When the temperature and humidity are unevenly distributed inside the power equipment, the measurement accuracy of the single sensor is often difficult to guarantee, resulting in a high false alarm or missed alarm rate; Third, the existing anti-condensation devices mostly use heating dehumidification to reduce the relative humidity by increasing the internal temperature of the equipment, thereby preventing condensation. However, the simple heating method only increases the unsaturation degree of water vapor in the air, does not actually discharge the water, and has high energy consumption; and when the external environment humidity is high, it is difficult to achieve the ideal dehumidification effect by heating alone.

[0004] It can be seen that when facing complex and changing environmental conditions, the existing anti-condensation devices are unable to dynamically adjust working parameters according to real-time data, resulting in difficulty in achieving the optimal anti-condensation effect. Summary of the invention

[0005] The present invention provides an intelligent induction anti-condensation system and method for power equipment to solve the technical problem that the existing anti-condensation device cannot dynamically adjust the working parameters according to real-time data when facing complex and changeable environmental conditions, resulting in difficulty in achieving the optimal anti-condensation effect.

[0006] In order to achieve the above object, the present invention adopts the following technical contents: An intelligent induction anti-condensation system for electric power equipment, comprising: A multivariate data collection unit, used for collecting real-time multivariate data to be predicted and historical multivariate data; An intelligent control unit is used to obtain a condensation risk prediction result based on the real-time multivariate data to be predicted and a pre-trained condensation risk prediction model; and output an anti-condensation instruction based on the condensation risk prediction result; wherein the condensation risk prediction model adopts a machine learning model and is trained based on historical multivariate data; The execution unit is used to execute the anti-condensation operation according to the anti-condensation instruction to prevent condensation from occurring inside the power equipment.

[0007] Furthermore, the multivariate data acquisition unit adopts a multivariate sensor array; the multivariate sensor array includes a temperature and humidity sensor and a dew point sensor installed in the power equipment cabinet, a temperature and humidity sensor and an air pressure sensor installed in the power equipment room, and a temperature and humidity sensor and a micrometeorological sensor installed outdoors the power equipment; the multivariate sensor array is electrically connected to the intelligent control unit through a high-speed communication interface.

[0008] Furthermore, the real-time multivariate data to be predicted and the historical multivariate data include at least: temperature inside the cabinet, humidity inside the cabinet, dew point inside the cabinet, indoor temperature, indoor humidity, indoor dew point, air pressure, weather forecast parameters, outdoor temperature, outdoor humidity and micrometeorological parameters.

[0009] Furthermore, the intelligent control unit includes a model training module, a condensation risk prediction module and an anti-condensation instruction output module; The model training module is used to train the machine learning model according to historical multivariate data to obtain a condensation risk prediction model; The condensation risk prediction module is used to input the collected real-time multivariate data to be predicted into the condensation risk prediction model to obtain the condensation risk prediction result; The anti-condensation instruction output module is used to judge the condensation risk prediction result and the preset risk threshold, and when the condensation risk prediction result is greater than or equal to the preset risk threshold, the anti-condensation instruction is transmitted to the execution unit.

[0010] Further, the execution unit includes a heating module, a dehumidification module and a ventilation module; The heating module is used to heat the interior of the power equipment to increase the air temperature and reduce the relative humidity; The dehumidification module is used to dehumidify the interior of the power equipment to remove moisture from the air; The ventilation module is used to introduce external dry air or promote internal air circulation to accelerate the discharge of moisture; The anti-condensation instruction includes working parameters, and the working parameters include start and stop time, heating power and dehumidification rate.

[0011] Furthermore, the anti-condensation system also includes a sleep unit, which is used to control the multivariate data acquisition unit, the intelligent control unit and the execution unit to suspend operation when the condensation risk is judged to disappear according to the condensation risk prediction result, and to re-awaken the multivariate data acquisition unit, the intelligent control unit and the execution unit after a preset time.

[0012] Furthermore, the anti-condensation system also includes a user interaction monitoring unit; the user interaction monitoring unit is electrically connected to the multivariate data acquisition unit, the intelligent control unit and the execution unit, respectively, and is used to display real-time multivariate data collected by the multivariate data acquisition unit, to display and store condensation risk prediction results output by the intelligent control unit, and to display and warn fault diagnosis results of the multivariate data acquisition unit and the execution unit.

[0013] An intelligent inductive anti-condensation method for electric power equipment, based on the above-mentioned intelligent inductive anti-condensation system for electric power equipment, comprises: The multivariate data collection unit collects real-time multivariate data to be predicted and historical multivariate data; wherein the real-time multivariate data to be predicted and the historical multivariate data both include at least: temperature inside the cabinet, humidity inside the cabinet, dew point inside the cabinet, indoor temperature, indoor humidity, indoor dew point, air pressure, weather forecast parameters, outdoor temperature, outdoor humidity and micro-meteorological parameters; The intelligent control unit obtains a condensation risk prediction result according to the real-time multivariate data to be predicted and the pre-trained condensation risk prediction model; and outputs an anti-condensation instruction according to the condensation risk prediction result; wherein the condensation risk prediction model adopts a machine learning model and is trained based on historical multivariate data; The execution unit executes the anti-condensation operation according to the anti-condensation instruction to prevent condensation from being generated inside the electric equipment.

[0014] Furthermore, the condensation risk prediction model is constructed based on a forest algorithm.

[0015] Furthermore, the condensation risk prediction result is judged against the preset risk threshold. If the condensation risk prediction result is greater than or equal to the preset risk threshold, the anti-condensation instruction is transmitted to the execution unit; if the condensation risk prediction result is less than the preset risk threshold, the operation of the multivariate data acquisition unit, the intelligent control unit and the execution unit are suspended to enter a sleep state.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides an intelligent inductive anti-condensation system for power equipment. The system includes a multivariate data acquisition unit, an intelligent control unit and an execution unit. By integrating the multivariate data acquisition unit, comprehensive real-time monitoring of the internal and external environment of the equipment is realized; the intelligent control unit uses a pre-trained machine learning model to accurately predict the condensation risk based on real-time and historical data, and dynamically adjusts the anti-condensation strategy accordingly, ensuring that condensation can be effectively prevented under various complex environmental conditions; at the same time, the precise execution of the execution unit further guarantees the anti-condensation effect, improves the overall operating stability and safety of the power equipment, reduces the risk of equipment failure caused by condensation, extends the equipment service life, and reduces unnecessary maintenance costs; the system has self-learning and adaptive capabilities, can continuously optimize the control strategy to adapt to different environmental conditions, and ensures the optimal anti-condensation effect.

[0017] Preferably, in the present invention, by arranging sensors at different positions inside and outside the power equipment, multivariate data affecting condensation formation are comprehensively collected, thereby improving the accuracy and comprehensiveness of data collection and providing a reliable data basis for condensation risk prediction.

[0018] Preferably, in the present invention, the real-time multivariate data and historical multivariate data to be predicted include at least data such as cabinet temperature, cabinet humidity, cabinet dew point, indoor temperature, indoor humidity, indoor dew point, air pressure, weather forecast parameters, outdoor temperature, outdoor humidity and micrometeorological parameters, covering multiple key factors affecting condensation formation, making condensation risk prediction more accurate and comprehensive, and enabling the system to better cope with complex and changeable environmental conditions.

[0019] Preferably, in the present invention, by dividing the three modules of model training, condensation risk prediction and anti-condensation instruction output, the automation and intelligence of condensation risk prediction and prevention and control are realized, and the response speed and accuracy of the system are improved.

[0020] Preferably, in the present invention, the humidity and dew point inside the power equipment are effectively reduced through various means such as heating, dehumidification and ventilation, the generation of condensation is prevented, and the normal operation of the equipment is guaranteed.

[0021] Preferably, in the present invention, a sleep unit is added to the system. When the condensation risk is low, each unit is controlled to enter a sleep state, thereby reducing the energy consumption of the system and extending the service life of the equipment, while retaining the ability to quickly respond to condensation risks.

[0022] Preferably, in the present invention, the system provides intuitive data display and fault diagnosis functions by adding a user interactive monitoring unit, which is convenient for users to monitor the operating status and condensation risk of the equipment in real time and take timely countermeasures, thereby improving the maintainability of the system and user experience.

[0023] The present invention also provides an intelligent inductive anti-condensation method for power equipment. Based on the above-mentioned intelligent inductive anti-condensation system for power equipment, the method realizes comprehensive, accurate and real-time prevention and control of condensation risks inside power equipment by integrating three links: multivariate data collection, intelligent risk prediction and anti-condensation execution. The method is based on detailed real-time and historical data (including temperature and humidity inside and outside the cabinet, dew point, air pressure, weather forecast and micro-meteorological parameters, etc.), and uses a pre-trained machine learning model to predict condensation risks, ensuring the accuracy and reliability of the prediction results. The intelligent control unit quickly outputs anti-condensation instructions according to the prediction results, and the execution unit quickly responds and executes the corresponding heating, dehumidification or ventilation operations, effectively preventing the generation of condensation, thereby ensuring the stable operation of the power equipment, extending the equipment life, and reducing the failure rate and maintenance cost caused by condensation. The use of this method effectively improves the intelligence level of anti-condensation of power equipment and enhances the safety and reliability of the equipment.

[0024] Preferably, in the present invention, the condensation risk prediction model is constructed based on a forest algorithm. The forest algorithm has high prediction accuracy and robustness, can more accurately predict the condensation risk, and improves the overall performance of the system.

[0025] Preferably, in the present invention, the condensation risk prediction results are judged against the preset risk threshold, thereby achieving a graded response to the condensation risk, which not only ensures the timeliness of prevention and control, but also avoids unnecessary waste of resources; at the same time, when the condensation risk is low, the system operation is suspended to enter a dormant state, thereby further reducing energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic structural diagram of an intelligent inductive anti-condensation system for electric power equipment provided by an embodiment of the present invention; Figure 2 A schematic diagram of the layout of a multi-sensor array of an intelligent inductive anti-condensation system for electric power equipment provided by an embodiment of the present invention; Figure 3 A working diagram of an intelligent control unit of an intelligent inductive anti-condensation system for electric power equipment provided by an embodiment of the present invention; Figure 4 A schematic diagram of the working mechanism of an actuator of an intelligent inductive anti-condensation system for electric power equipment provided by an embodiment of the present invention Figure 5 A schematic diagram of a user interaction interface of an intelligent inductive anti-condensation system for electric power equipment provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail in the following specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0031] In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "horizontal", "inner", etc. indicate an orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the invention is usually placed when in use, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0032] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", which does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0033] In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0034] Example 1 As mentioned in the background technology, although the existing anti-condensation devices can play a certain dehumidification role, they have certain limitations, which are specifically manifested in the following aspects: First, most of the existing anti-condensation devices rely on manually set thresholds for control. When facing complex and changeable environmental conditions, it is difficult to achieve the optimal anti-condensation effect due to the inability to dynamically adjust the working parameters according to real-time data; Second, most of the existing anti-condensation devices only use a single sensor. When the temperature and humidity are unevenly distributed inside the power equipment, the measurement accuracy of the single sensor is often difficult to guarantee, resulting in a high false alarm or missed alarm rate; Third, the existing anti-condensation devices mostly use heating dehumidification to reduce the relative humidity by increasing the internal temperature of the equipment, thereby preventing condensation. However, the simple heating method only increases the unsaturation degree of water vapor in the air, does not actually discharge the water, and has high energy consumption; and when the external environment humidity is high, it is difficult to achieve the ideal dehumidification effect by heating alone.

[0035] The defects of the prior art are more specifically reflected in the following aspects: 1. The single sensor detection method has great limitations: At present, some anti-condensation devices on the market only use a single temperature and humidity sensor for detection. However, humidity is a function of temperature, and temperature changes have a decisive influence on the humidity measurement results. When the temperature and humidity are unevenly distributed inside the power equipment, the measurement accuracy of a single sensor is often difficult to guarantee, resulting in a high false alarm or missed alarm rate.

[0036] 2. Heating dehumidification is inefficient and consumes a lot of energy: Traditional anti-condensation devices mostly use heating dehumidification, which reduces relative humidity by increasing the internal temperature of the equipment to prevent condensation; however, this method only increases the unsaturation of water vapor in the air, does not actually discharge moisture, and consumes a lot of energy. In addition, when the external humidity is high, it is difficult to achieve the ideal dehumidification effect by simply heating.

[0037] 3. Lack of support for intelligent control algorithms: Most existing anti-condensation devices rely on manually set thresholds for control and lack the support of intelligent control algorithms. This causes the device to be unable to cope with complex and changing environmental conditions and is unable to dynamically adjust working parameters according to real-time data to achieve the optimal anti-condensation effect.

[0038] 4. Low system integration and high maintenance cost: Some anti-condensation devices are not designed with full consideration of system integration, resulting in complicated installation, commissioning and maintenance processes; this not only increases the user's cost of use, but also reduces the reliability and ease of use of the device.

[0039] In order to solve the above problems, this embodiment provides an intelligent sensing anti-condensation system for power equipment. The system can accurately judge whether the power equipment needs anti-condensation operation in real time, and has high working efficiency, high intelligence, and good anti-condensation effect, which is of great significance for ensuring the stable operation of the power system.

[0040] The present invention is further described in detail below with reference to the accompanying drawings and embodiments: like Figure 1 As shown, this embodiment provides an intelligent inductive anti-condensation system for electric power equipment, including a multivariate data acquisition unit, an intelligent control unit and an execution unit, and the specific structure and functions are as follows: Combination Figure 1 and Figure 2 As shown, in this embodiment, the multivariate data acquisition unit uses a multivariate sensor array to collect real-time multivariate data to be predicted and historical multivariate data; The multi-sensor array includes temperature and humidity sensors and dew point sensors installed in the power equipment cabinet, temperature and humidity sensors and air pressure sensors installed in the power equipment room, and temperature and humidity sensors and micro-meteorological sensors installed outdoors; the above-mentioned high-precision sensors are used to comprehensively monitor the environmental parameters inside and around the power equipment.

[0041] The above sensors are designed to collect environmental parameters inside and outside the power equipment at regular intervals to ensure comprehensive monitoring of the operating environment of the power equipment. The collected parameters cover multiple aspects such as temperature, humidity, dew point, air pressure, and meteorology, also known as multivariate data; the real-time multivariate data (data to be predicted) is collected immediately, while the multivariate data used to build and train machine learning models is historical multivariate data; the multivariate data collected above provides a detailed data foundation for subsequent condensation risk prediction and intelligent control.

[0042] In this embodiment, the sensors are reasonably arranged according to the internal structure of the power equipment and the airflow characteristics to ensure that the temperature, humidity and dew point status of each area can be accurately reflected; Specifically, the outdoor part: install micro-meteorological sensors; the indoor part: install temperature and humidity sensors; the power equipment cabinet part: install temperature and humidity sensors and dew point sensors; the power equipment cable trench part: install temperature sensors and airflow sensors. This layout strategy not only improves the comprehensiveness of data collection, but also effectively avoids monitoring blind spots caused by improper sensor locations.

[0043] Data transmission: The collected multivariate data needs to be transmitted to the intelligent control unit through a stable and efficient communication interface. To achieve this goal, this embodiment adopts a unified high-speed communication interface, such as RS485 MODBUS, CAN bus, etc. These communication interfaces have the advantages of fast transmission speed, high stability, and strong anti-interference ability, which can ensure the accuracy and real-time performance of data during transmission. After receiving the data, the intelligent control unit will further process and analyze it to achieve intelligent control of power equipment and accurate prediction of condensation risks.

[0044] In this embodiment, the intelligent control unit includes a core processor, which uses a high-performance microprocessor or an embedded system as the core processor, and is responsible for data processing, algorithm operation and instruction sending; specifically, it can obtain condensation risk prediction results based on the real-time multivariate data to be predicted and the pre-trained condensation risk prediction model; and output anti-condensation instructions based on the condensation risk prediction results; the core processor can execute the following intelligent control algorithm: This control algorithm can integrate the multivariate data collected by the multivariate sensor array, analyze the environmental change trend, and predict the condensation risk. According to the condensation risk prediction results, dynamically adjust the working parameters of the actuator, such as heating power, dehumidification rate, operating time, etc.

[0045] Specifically, this control algorithm has self-learning and self-adaptation capabilities, and can continuously optimize the control strategy according to the actual operation effect to adapt to different environmental conditions and equipment status. This algorithm first collects multivariate data from multiple sensors, such as temperature, humidity and air pressure, and performs data preprocessing, including cleaning outliers and noise, and normalization to ensure that data of different dimensions can be used in the same model. Then extract features, such as the rate of change of temperature, the rate of change of humidity, and other related features, such as dew point temperature and absolute humidity; then use weighted average, Kalman filtering or neural network methods to fuse the data of multiple sensors, and the weights can be dynamically adjusted according to the accuracy and reliability of the sensors. In the condensation risk prediction stage, a machine learning model (such as support vector machine, random forest or neural network) is applied for prediction. The condensation risk prediction model can be expressed as a function of condensation risk as temperature, humidity, air pressure, rate of change of temperature, rate of change of humidity and other features. Finally, according to the condensation risk prediction results, an anti-condensation instruction is output, and the working parameters such as heating power and dehumidification rate of the actuator are dynamically adjusted according to the anti-condensation instruction.

[0046] In this embodiment, the prediction effect of the control algorithm is better, which is mainly attributed to the application of multivariate data fusion, dynamic feature extraction, machine learning model and real-time adjustment capabilities.

[0047] Among them, the condensation risk prediction model can be expressed as: condensation risk = f(T, H, P, dT / dt, dH / dt, other characteristics) Among them, f is the functional relationship learned through training data, temperature (T), humidity (H), air pressure (P), temperature change rate (dT / dt), humidity change rate (dH / dt).

[0048] In this embodiment, the intelligent inductive anti-condensation system is also equipped with a large-capacity memory for storing historical multivariate data and condensation risk prediction models, which is convenient for direct call of subsequent prediction operations. At the same time, it supports remote communication functions and can transmit data to the background monitoring system through wired or wireless means to realize remote monitoring and fault diagnosis.

[0049] The following is a further description of the working process of the intelligent control unit in conjunction with the accompanying drawings: The intelligent control unit uses the built-in intelligent control algorithm to analyze the pre-processed real-time multivariate data, identify environmental change trends, and predict condensation risks. The intelligent control algorithm uses advanced technologies such as machine learning and fuzzy logic to improve the accuracy of predictions.

[0050] like Figure 3 As shown, the specific working process of the intelligent control unit is as follows: This intelligent control unit integrates the temperature, humidity, dew point, air pressure and meteorological data (wind speed, wind direction, precipitation, light, etc.) obtained by the multi-sensor array to predict real-time multi-data, combines weather forecasts (meteorological data) and historical multi-data, predicts the condensation phenomenon in the power equipment, and accurately controls the operation of the dehumidification, heating and ventilation modules of the execution unit to achieve anti-condensation effect and maximize energy saving. The specific steps are as follows: A. Algorithm feature selection: Select features that have an important impact on condensation, including cabinet internal and external temperature, outdoor temperature, cabinet internal and external humidity, outdoor humidity, dew point, air pressure, wind speed, wind direction, precipitation, light, etc.

[0051] B. Model training: Use machine learning algorithms (neural networks) to build a condensation prediction model (also called a condensation risk prediction model), and use historical data and weather forecast data (including the above-mentioned features that have a significant impact on condensation phenomena) to train and optimize the model.

[0052] C. Condensation risk prediction: Input real-time data and weather forecast data into the trained condensation prediction model to obtain the condensation risk prediction results.

[0053] D. Algorithm implementation: In the condensation prediction model, Y = f(X) represents a functional relationship, where X is the input feature vector, which includes various environmental factors that affect the condensation phenomenon, such as cabinet temperature, outdoor temperature, cabinet humidity, outdoor humidity, dew point, air pressure, wind speed, wind direction, precipitation, light, etc.; and Y is the condensation risk prediction result output by the model.

[0054] The following is a detailed description of the condensation prediction model: 1. Feature selection and processing: a. Feature selection: From the data obtained from the multi-sensor array and weather forecast service, select the features that have an important impact on the condensation phenomenon (temperature, humidity, dew point) as input X.

[0055] b. Feature processing: preprocessing of selected features, including data cleaning (removing outliers and noise), data interpolation (processing missing values), feature scaling (normalization or standardization), etc.

[0056] 2. Model selection and training: a. Model selection: Random Forest algorithm.

[0057] b. Model training: Use historical data and weather forecast data as training sets to train the selected machine learning algorithm. During the training process, the prediction performance of the model on the training set is optimized by adjusting the algorithm parameters and optimizing the objective function.

[0058] 3. Model prediction and optimization: a. Model prediction: The real-time data and weather forecast data are used as input X and input into the trained model to obtain the prediction result Y of condensation risk.

[0059] The forest algorithm itself is an integrated learning method that makes a final prediction by constructing multiple decision trees and combining their prediction results. When constructing each decision tree, this embodiment not only performs random column sampling, but also dynamically adjusts the feature selection strategy according to the structure of the current tree and the data distribution. For example, features can be weighted and selected according to the importance or relevance of the features, or different feature subsets can be used at different tree levels. At the same time, according to the characteristics of the specific problem, new splitting criteria are designed to optimize the construction of the decision tree. For example, for data sets such as temperature and humidity, custom splitting criteria based on distance, density or similarity are designed to better capture the relationships and patterns between data.

[0060] Its prediction process can be expressed as: Y=f(X;θ); in: Y is the condensation risk predicted by the model.

[0061] X is the input feature vector, which contains various environmental factors that affect the condensation phenomenon.

[0062] f is the function defined by the machine learning algorithm, which describes the mapping relationship between input features and predicted results.

[0063] θ is the parameter of the model, which is adjusted by the optimization algorithm during the training process to optimize the prediction performance of the model on the training set.

[0064] For the random forest algorithm, f is actually an integrated model composed of multiple decision trees, each of which has its own parameters and structure. When given an input feature vector X, the random forest algorithm will make each decision tree make a prediction, and then combine the average results of these predictions to get the final prediction result Y.

[0065] In order to train the model, in this embodiment, the error term ε needs to be minimized. Generally, this embodiment uses the mean square error (MSE) as the loss function:

[0066] Where n is the number of samples, y i is the actual target value, x i T θ is the predicted value of the model.

[0067] In this embodiment, the parameter θ that minimizes the loss function can be found through optimization algorithms such as gradient descent.

[0068] The specific implementation method is as follows: First, a dataset is loaded using the Pandas library. Then, three features are selected from the dataset: ‘Temperature’, ‘Humidity’, and ‘DewPoint’ as the input variables of the model and three target variables: ‘E_dehumidifier’, ‘E_heater’, and ‘E_ventilation’ as the target variable y for three different linear regression models.

[0069] Next, use the train_test_split function to divide the data set into a training set and a test set, where the test set accounts for 20% of the total data and random_state=42; this ensures the randomness of the data division and repeatability.

[0070] A linear regression model was trained for each target variable and the three linear regression models were used to predict the energy consumption of dehumidifier, heater and ventilation system respectively.

[0071] b. Model optimization: Further optimize the model according to the needs of actual application scenarios.

[0072] The real-time data and weather forecast data in the test set are used as input X and input into the trained random forest model to obtain the prediction result Y of condensation risk.

[0073] The model is further optimized through cross-validation, grid search and other methods to improve the prediction performance and generalization ability.

[0074] Condensation risk assessment and decision making: According to the condensation risk prediction result Y output by the model, determine whether there is condensation risk in the current environment.

[0075] According to the judgment results of condensation risk, formulate corresponding control strategies, such as starting the dehumidification module, heating module or keeping the equipment in sleep mode.

[0076] 4. Condensation risk assessment and decision making: a. Risk judgment: Based on the condensation risk prediction result Y output by the model, determine whether there is condensation risk in the current environment.

[0077] Decision making: According to the condensation risk prediction results, formulate corresponding control strategies. For example, when the condensation risk is predicted to be high, start the dehumidification module and / or the heating module; when the condensation risk is predicted to be low, keep the system in a dormant state to save energy.

[0078] b. Decision making: By real-time monitoring of ambient temperature and humidity, combined with historical data and the above algorithm, the condensation risk is predicted. When the condensation risk is predicted to be high (≧80%), the dehumidification module or heating module is automatically started to reduce humidity or increase temperature, thereby effectively preventing the formation of condensation; when the condensation risk is predicted to be low (<80%), the system remains in a dormant state to save energy. This strategy ensures that the equipment runs in time when needed and maintains energy saving when not needed, thereby achieving efficient and intelligent anti-condensation dehumidification management.

[0079] Combination Figure 1 and Figure 4 As shown, in this embodiment, the execution unit includes a heating module, a dehumidification module and a ventilation module; the specific functions are as follows: Heating module: According to the instructions of the intelligent control unit, the internal part of the power equipment is heated to increase the air temperature and reduce the relative humidity, thereby preventing condensation. The heating module can be made of high-efficiency and energy-saving materials such as PTC thermistors and heating films.

[0080] Dehumidification module: When the humidity is high or the risk of condensation is high, the dehumidification module is started to perform dehumidification. The dehumidification module can use a variety of methods such as condensation dehumidification and adsorption dehumidification, which can be selected and optimized according to actual needs.

[0081] Ventilation module: Under appropriate conditions, ventilation modules can be used to introduce dry external air or promote internal air circulation to accelerate the discharge of moisture and reduce the risk of condensation.

[0082] In this embodiment, the specific working principle of each module of the execution unit is also provided: Dehumidification module: When the humidity is high or the risk of condensation is high, the dehumidification module will be activated. This module uses electronic condensation dehumidification, adsorption dehumidification and other technologies to effectively remove moisture from the air and further reduce the risk of condensation.

[0083] Start condition: Start when the ambient humidity exceeds the preset threshold (such as 70%RH) or the above prediction model shows high risk.

[0084] Start-up ratio: adjust the working intensity according to the degree to which the humidity exceeds the threshold and the condensation risk level. For example, for every 5% RH increase in humidity, increase the working power by 10% until the maximum dehumidification capacity is reached.

[0085] Stop condition: Stop when the humidity drops to a safe range (such as 50%-60%RH) and the risk of condensation is low.

[0086] Standalone operation: When humidity is the main problem and other conditions (such as temperature) are suitable, it can be operated alone.

[0087] Combined operation: works in conjunction with the heating module, especially when the temperature is also low, to accelerate the dehumidification effect.

[0088] Heating module: When condensation risk is predicted, the intelligent control unit may first start the heating module to heat the inside of the power equipment, increase the air temperature, reduce the relative humidity, and circulate the air through the ventilation system to suppress the generation of condensation. The heating power can be dynamically adjusted according to real-time data and prediction results to minimize energy consumption.

[0089] Startup conditions: Startup when condensation risk is predicted, especially when the temperature is close to or below the dew point temperature.

[0090] Start-up ratio: Dynamically adjust the heating power according to the temperature difference between the temperature and the dew point and the humidity. The smaller the temperature difference and the higher the humidity, the greater the heating power.

[0091] Stop condition: Stop after the internal temperature rises to a safe range and the humidity is effectively controlled.

[0092] Single operation: When temperature is the main factor in condensation formation, it can be used alone to increase the ambient temperature.

[0093] Combined operation: Usually used in conjunction with a dehumidification module and a ventilation module to fully reduce the risk of condensation.

[0094] Ventilation modules: In some cases, ventilation modules will be used to introduce dry air from outside or promote internal air circulation to accelerate the discharge of moisture and reduce the risk of condensation. The ventilation rate and direction can be adjusted according to actual needs.

[0095] Startup conditions: Start when the internal humidity is high and the external air is relatively dry, or when it is necessary to accelerate the internal air circulation.

[0096] Start-up ratio: The ventilation rate and direction can be adjusted according to the humidity difference between inside and outside and the wind speed. The drier the outside, the more appropriate the ventilation rate can be.

[0097] Stop condition: When the internal humidity reaches the ideal level, or the external humidity becomes so high that it is no longer conducive to ventilation and dehumidification.

[0098] Standalone operation: It can be operated standalone when the external environment is dry and only slight dehumidification is required inside.

[0099] Combined operation: often works together with dehumidification module and heating module to promote the exchange of moisture and heat.

[0100] It can be seen that this embodiment provides an intelligent inductive anti-condensation system for electric power equipment, which has the following advantages: First, multi-data fusion monitoring: Integrate multiple sensors (including temperature and humidity sensors, dew point sensors, air pressure sensors and meteorological sensors, etc.) to monitor the internal and external environmental parameters of power equipment in real time, and improve measurement accuracy and reliability through data fusion algorithms.

[0101] Second, intelligent control algorithm: using advanced intelligent control algorithm, according to real-time monitoring data, recent weather forecast data and historical data, dynamically adjust the working parameters (such as start and stop time, heating power, dehumidification rate, etc.), predict condensation in advance, so as to achieve the best anti-condensation effect. At the same time, the algorithm also has self-learning and self-adaptive capabilities, and can continuously optimize the control strategy to adapt to different environmental conditions.

[0102] Third, easy installation and maintenance: The device is modularly designed, and different modules are selected according to different application scenarios. The structure is compact and reasonable, which is convenient for users to install and debug. At the same time, it provides a friendly user interface and remote monitoring function to facilitate users to carry out daily maintenance and troubleshooting.

[0103] Fourth, by optimizing the dehumidification method and improving the system integration, the energy consumption of the device can be reduced and the operating efficiency can be improved.

[0104] Example 2 This embodiment also provides another intelligent inductive anti-condensation system for power equipment, the basic structure of which is the same as that of Embodiment 1, and further optimization and improvement are made on the basis of Embodiment 1. The specific optimization scheme is as follows: This embodiment provides an intelligent inductive anti-condensation system for electric power equipment, and further provides a sleep unit, that is, provides an intelligent sleep mechanism, which is used to control the multivariate data acquisition unit, the intelligent control unit and the execution unit to suspend operation when the condensation risk is judged to disappear according to the condensation risk prediction result, and wake up the multivariate data acquisition unit, the intelligent control unit and the execution unit again after a preset time. Specifically: The system makes a judgment based on the condensation risk prediction results of real-time monitoring. When it is predicted that the condensation is about to disappear, all modules are automatically stopped and enter a dormant state.

[0105] During sleep, the system periodically wakes up to perform brief data collection and prediction to ensure timely response to condensation risks.

[0106] Startup conditions: Start when all monitoring indicators show that the condensation risk is extremely low and the prediction model shows that there will be no condensation risk in the future.

[0107] During sleep: Wake up periodically (e.g. every hour) to perform short-term data collection and analysis to ensure timely response to possible condensation risks.

[0108] Stop sleep: Once any signs of condensation risk are found, the relevant modules will be immediately awakened and put into working state.

[0109] It can be seen that in this embodiment, by adding a sleep unit, an intelligent sleep mechanism is proposed, which can further reduce the standby power consumption, thereby meeting the current requirements of energy saving and environmental protection.

[0110] like Figure 5 As shown, this embodiment provides an intelligent inductive anti-condensation system for power equipment, and further provides a user interaction interface, which is electrically connected to the multi-dimensional data acquisition unit, the intelligent control unit and the execution unit respectively, and can display the data collected by the multi-dimensional data acquisition unit; it can switch between automatic and manual control, and has an operation button to realize manual control operation; it can control each module of the execution unit in real time, which is convenient for the user's manual adjustment and intervention, and improves the flexibility and control experience of the entire system.

[0111] This user interaction interface can be integrated into the console to facilitate manual operation and monitoring by the user. The specific functions are as follows: Local display: Equipped with local display devices such as color LCD screens or LED indicators to display key parameters such as temperature, humidity, dew point, etc. inside the power equipment and the working status of the anti-condensation device in real time.

[0112] Remote monitoring: Supports remote monitoring and configuration via a Web browser or dedicated client software. Users can view device status, receive alarm information, and perform remote operations anytime and anywhere.

[0113] Fault self-diagnosis and alarm: The device has a fault self-diagnosis function, which can automatically detect sensor failures, heating and dehumidification module abnormalities, etc., and promptly remind users to handle them through sound and light alarms, SMS notifications, etc.; it helps to ensure stable operation and timely maintenance of the equipment.

[0114] It can be seen that the user interaction monitoring unit added in this embodiment provides intuitive data display and fault diagnosis functions, which facilitates users to monitor the operating status and condensation risk of the equipment in real time, take countermeasures in time, and improve the maintainability of the system and user experience.

[0115] Example 3 This embodiment also provides another intelligent inductive anti-condensation system for electric power equipment, which is further improved on the basis of Embodiment 1 or Embodiment 2, as follows: In actual operation, the start and stop of each module should also consider factors such as energy efficiency, equipment life, and maintenance cost, assign weights to each factor, and make comprehensive decisions. For example, historical data can be learned through machine learning models to optimize the start-up strategy and power regulation of each module to achieve both efficient and economical operation results.

[0116] Therefore, in order to more accurately control the intelligent inductive anti-condensation system for power equipment, this embodiment designs an algorithm based on multi-factor weight decision-making. This algorithm comprehensively considers the humidity, temperature, dew point, air pressure, airflow and other data outdoors, indoors, inside the cabinet, and in the cable trench, and assigns different operation weights. In the early stage of model training, the following weights are assigned. Because the condensation conditions in different regions will be different, the weight ratios are adjusted according to the effects after a period of operation, as shown in Table 1.

[0117] Table 1 is a reference table for weight distribution

[0118] It can be seen that according to the preset weight values ​​of each multivariate data, the training effect of the condensation risk prediction model can be improved, so that the prediction accuracy of the condensation risk prediction model is higher, thereby ensuring the accurate judgment of the condensation risk prediction results, so as to achieve accurate control of the execution unit, thereby improving the work efficiency and discrimination ability of the system, and has good promotion and application value.

[0119] In summary, the present invention provides an intelligent inductive anti-condensation system and method for electric power equipment. The existing anti-condensation technology for electric power equipment has the following advantages: This system realizes comprehensive monitoring and efficient prevention of condensation risks inside power equipment by integrating key functions such as multi-dimensional data collection, intelligent prediction control, precise execution of operations, and optimized energy consumption management. This system uses a multi-sensor array to widely collect internal and external environmental data, combines machine learning models to accurately predict condensation risks, and intelligently outputs anti-condensation instructions based on the prediction results. Through the synergistic effect of various means such as heating, dehumidification, and ventilation, it effectively prevents the occurrence of condensation and ensures the safe and stable operation of power equipment. At the same time, the system introduces a sleep unit and a user interactive monitoring unit, which not only reduces energy consumption at non-essential times, but also improves the system's operational convenience and fault diagnosis capabilities, providing strong support for the intelligent maintenance and management of power equipment. It can be seen that the use of this system has significantly improved the automation level and reliability of anti-condensation of power equipment, reduced operation and maintenance costs, and has significant economic and social benefits.

[0120] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention.

Claims

1. An intelligent induction anti-condensation system for power equipment, characterized in that: include: A multivariate data collection unit, used for collecting real-time multivariate data to be predicted and historical multivariate data; An intelligent control unit, used to obtain a condensation risk prediction result based on the real-time multivariate data to be predicted and a pre-trained condensation risk prediction model; and output an anti-condensation instruction based on the condensation risk prediction result; wherein the condensation risk prediction model adopts a machine learning model and is trained based on historical multivariate data; The execution unit is used to execute the anti-condensation operation according to the anti-condensation instruction to prevent condensation from occurring inside the power equipment.

2. The intelligent inductive anti-condensation system for electric power equipment according to claim 1, characterized in that: The multivariate data acquisition unit adopts a multivariate sensor array; the multivariate sensor array includes a temperature and humidity sensor and a dew point sensor arranged in the power equipment cabinet, a temperature and humidity sensor and an air pressure sensor arranged indoors, and a temperature and humidity sensor and a micro-meteorological sensor arranged outdoors; The multi-sensor array is electrically connected to the intelligent control unit via a high-speed communication interface.

3. The intelligent inductive anti-condensation system for electric power equipment according to claim 2, characterized in that: The real-time multivariate data to be predicted and the historical multivariate data include at least: temperature inside the cabinet, humidity inside the cabinet, dew point inside the cabinet, indoor temperature, indoor humidity, indoor dew point, air pressure, weather forecast parameters, outdoor temperature, outdoor humidity and micro-meteorological parameters.

4. The intelligent inductive anti-condensation system for electric power equipment according to claim 1, characterized in that: The intelligent control unit includes a model training module, a condensation risk prediction module and an anti-condensation instruction output module; The model training module is used to train the machine learning model according to historical multivariate data to obtain a condensation risk prediction model; The condensation risk prediction module is used to input the collected real-time multivariate data to be predicted into the condensation risk prediction model to obtain the condensation risk prediction result; The anti-condensation instruction output module is used to judge the condensation risk prediction result and the preset risk threshold, and when the condensation risk prediction result is greater than or equal to the preset risk threshold, the anti-condensation instruction is transmitted to the execution unit.

5. The intelligent inductive anti-condensation system for electric power equipment according to claim 1, characterized in that: The execution unit includes a heating module, a dehumidification module and a ventilation module; The heating module is used to heat the interior of the power equipment to increase the air temperature and reduce the relative humidity; The dehumidification module is used to dehumidify the interior of the power equipment to remove moisture from the air; The ventilation module is used to introduce external dry air or promote internal air circulation to accelerate the discharge of moisture; The anti-condensation instruction includes working parameters, and the working parameters include start and stop time, heating power and dehumidification rate.

6. The intelligent inductive anti-condensation system for electric power equipment according to claim 1, characterized in that: The anti-condensation system also includes a sleep unit, which is used to control the multivariate data acquisition unit, the intelligent control unit and the execution unit to suspend operation when the condensation risk is determined to disappear according to the condensation risk prediction result, and to wake up the multivariate data acquisition unit, the intelligent control unit and the execution unit again after a preset time.

7. The intelligent inductive anti-condensation system for electric power equipment according to claim 1, characterized in that: The anti-condensation system also includes a user interaction monitoring unit; the user interaction monitoring unit is electrically connected to the multivariate data acquisition unit, the intelligent control unit and the execution unit, respectively, and is used to display real-time multivariate data collected by the multivariate data acquisition unit, to display and store condensation risk prediction results output by the intelligent control unit, and to display and warn fault diagnosis results of the multivariate data acquisition unit and the execution unit.

8. An intelligent inductive anti-condensation method for electric power equipment, characterized in that: The intelligent inductive anti-condensation system for electric power equipment according to any one of claims 1 to 7 comprises: The multivariate data collection unit collects real-time multivariate data to be predicted and historical multivariate data; wherein the real-time multivariate data to be predicted and the historical multivariate data both include at least: temperature inside the cabinet, humidity inside the cabinet, dew point inside the cabinet, indoor temperature, indoor humidity, indoor dew point, air pressure, weather forecast parameters, outdoor temperature, outdoor humidity and micro-meteorological parameters; The intelligent control unit obtains a condensation risk prediction result according to the real-time multivariate data to be predicted and the pre-trained condensation risk prediction model; and outputs an anti-condensation instruction according to the condensation risk prediction result; wherein the condensation risk prediction model adopts a machine learning model and is trained based on historical multivariate data; The execution unit executes the anti-condensation operation according to the anti-condensation instruction to prevent condensation from being generated inside the electric equipment.

9. The intelligent inductive anti-condensation method for electric power equipment according to claim 8, characterized in that: The condensation risk prediction model is constructed based on the forest algorithm.

10. The intelligent inductive anti-condensation method for electric power equipment according to claim 8, characterized in that: The condensation risk prediction result is judged against the preset risk threshold. If the condensation risk prediction result is greater than or equal to the preset risk threshold, the anti-condensation instruction is transmitted to the execution unit; if the condensation risk prediction result is less than the preset risk threshold, the operation of the multivariate data acquisition unit, the intelligent control unit and the execution unit are suspended to enter a sleep state.

Images