Electrical cabinet active anti-condensation method and system based on condensation mechanism mathematical model

Abstract

The invention relates to the technical field of electrical cabinet monitoring, in particular to an electrical cabinet active anti-condensation method and system based on a condensation mechanism mathematical model. According to the invention, a mathematical model based on a condensation formation mechanism is constructed, and environmental meteorological parameters, station room environmental parameters, electrical cabinet internal microenvironment parameters and cable trench state parameters are monitored through a four-stage monitoring system; and fusing the multi-source heterogeneous data, inputting a condensation mechanism mathematical model to calculate a condensation risk index, and adopting a hierarchical response mechanism to cooperatively control an execution mechanism according to the obtained condensation risk index value to realize active defense of condensation. According to the method, the condensation phenomenon with the space-time dynamic characteristic is quantitatively described through a mass transfer and heat transfer coupling equation, a grading prevention and control strategy and a cooperative control algorithm are developed by calculating the condensation risk index, setting the threshold condition and judging the condensation formation risk level, the prevention and control intensity is dynamically adjusted according to the risk level, and the control accuracy is improved. And the risk response speed and the resource utilization efficiency under the complex working condition are obviously improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electrical cabinet monitoring, and in particular to an electrical cabinet active anti-condensation method and system based on a condensation mechanism mathematical model. BACKGROUND

[0002] During the operation of an electrical cabinet, condensation may occur due to the dynamic coupling of environmental temperature and humidity, and condensation is very harmful to the equipment in the electrical cabinet. Therefore, it has great practical value to study condensation prevention and control technology. In response to the risk of condensation, the engineering field has developed various anti-condensation technologies, but the current mainstream research direction is a dynamic and intelligent prevention and control system. In the intelligent prevention and control mode, the condensation risk level is determined by monitoring environmental parameters, and a heating and dehumidifying device is controlled to prevent condensation from forming. The determination of the condensation risk is an important part, and the accuracy of the determination is directly determined by the evaluation model, which directly affects whether the prevention measures are reasonable and timely.

[0003] The condensation risk evaluation model in the prior art has defects, which are mainly manifested as the disconnection of the "mechanism-monitoring-prevention and control" chain. The condensation risk evaluation model does not fully consider the formation mechanism of condensation, lacks multi-dimensional monitoring data support, and discrete prevention and control measures are disconnected from the dynamic environmental evolution law. This systematic fragmentation makes it difficult for the existing technology to meet the demand of the intelligent power grid for panoramic perception of the equipment state and active intervention of the risk. At the same time, in terms of prevention and control strategy, the existing prevention and control mode is relatively single, and cannot carry out graded prevention and control according to the condensation risk level, resulting in poor prevention and control effect. SUMMARY

[0004] To solve the above problems, the present application provides an electrical cabinet active anti-condensation method and system based on a condensation mechanism mathematical model, which constructs a condensation mechanism mathematical model based on multi-dimensional data, calculates a condensation risk index, and implements graded active regulation of an actuator, thereby improving the condensation prevention and control effect.

[0005] The technical scheme of the present application is as follows: An electrical cabinet active anti-condensation method based on a condensation mechanism mathematical model, comprising the following steps: S1, monitoring environmental meteorological parameters, station room environmental parameters, electrical cabinet internal micro-environment parameters and cable trench state parameters; S2, performing multi-source data fusion on the data collected in step S1, and realizing physical quantity collaborative analysis through a hierarchical architecture during the fusion process; S3, inputting the data obtained in step S2 into a pre-constructed condensation mechanism mathematical model to calculate a condensation risk index; wherein the construction process of the condensation mechanism mathematical model is as follows: The improved The dew point temperature is calculated by an empirical formula, and the formula is as follows: (1), In formula (1), Td is the dew point temperature in the electrical cabinet, Ta is the air temperature in the electrical cabinet, RH is the relative humidity in the electrical cabinet, P0 is the standard atmospheric pressure, P is the actual atmospheric pressure in the station room, and the constants 243.12 and 17.62 are empirical values based on experimental data fitting of thermodynamic principles; The transport equation of water vapor and the temperature field evolution equation are constructed to quantitatively describe the condensation phenomenon with spatiotemporal dynamic characteristics. The transport equation of water vapor is: (2), In formula (2), ρ is the water vapor density in the cabinet, is the partial derivative of the water vapor density in the cabinet with respect to time; is the effective diffusion coefficient of the comprehensive molecular diffusion and weak convection effect, which is calibrated by experiments for simplified calculation; the Laplace term describes the spatial gradient effect of water vapor diffusion, and reveals the high-risk area of condensation; characterizes the phase change rate constant determined by the surface roughness and material characteristics together; is the Heaviside function, H(T d - T) represents that when the temperature in the cabinet T is lower than the dew point temperature T d , it is 1, triggering condensation; other parameter meanings are the same as those in formula (1); The temperature field evolution equation is: (3), In formula (3), is the air density, is the specific heat capacity at constant pressure, is the partial derivative of the temperature with respect to time; is the latent heat of vaporization of water; the Laplace term describes the spatial gradient effect of heat conduction, and reveals the high-risk area of condensation; is the thermal conductivity; other parameter meanings are the same as those in formulas (1) and (2); Formulas (2) and (3) are the mathematical model of the condensation mechanism. The parameters obtained by applying the model are combined with formula (4) to output the condensation risk index CRI: (4), In formula (4), represents the integral region, i.e., the spatial volume in the electrical cabinet; represents the temperature gradient, The module representing the temperature gradient; the formula (4) is obtained by integrating the local conditions of each point in the electrical cabinet to obtain the global condensation risk index; S4, the condensation risk index value obtained according to step S3 adopts a hierarchical response mechanism to cooperatively control the execution mechanism, so as to realize the active defense of condensation.

[0006] Further, the multi-source data fusion of step S2 is divided into three steps of primary fusion, intermediate fusion and advanced fusion, the primary fusion is used to eliminate the noise of the same type of sensor; the intermediate fusion establishes a multivariate regression model of temperature-air pressure-wind speed, and the coefficients are calibrated through historical data to realize abnormal data checking and missing compensation; the advanced fusion introduces Kalman filter dynamic correction to correct the model deviation caused by environmental mutation.

[0007] Further, the model verification and parameter optimization process are further included after the model is constructed in step S3; the model verification is realized by comparing the simulation results with the experimental data; the parameter optimization process is first based on the water vapor transport equation and the temperature field evolution equation to construct an error evaluation function : (5), In formula (5), is the predicted value of the water vapor density in the cabinet, is the measured value of the water vapor density in the cabinet; and are the predicted value and the measured value of the air temperature in the cabinet, respectively; is a weight coefficient for balancing the contribution degree of humidity and temperature error, represents a time sampling point, is the number of sampling points; The effective diffusion coefficient D and the phase change rate constant β are solved by inversion optimization , and and and are solved by using formulas (2) and (3), and the evaluation function is solved by substituting formula (5), and the gradient descent method is used for iteration until the evaluation function reaches the minimum value, at which time the effective diffusion coefficient D and the phase change rate constant β are the optimal values.

[0008] The process of step S4 for taking a hierarchical response mechanism to control the execution mechanism is as follows: When the condensation risk index value is between 0.5 and 1.0, start the basic temperature control mode, use the heat generated by the device to maintain the temperature in the cabinet slightly higher than the dew point temperature, and inhibit the initial condensation; If the condensation risk index value decreases to the range of 0.2 to 0.5, activate the gradient heating and dehumidification mode, and perform zoned heating according to the humidity gradient distribution in the cabinet, while adjusting the heating intensity through closed-loop feedback to avoid local overheating; When the dew risk index value is lower than 0.2 or the humidity mutation rate exceeds 5%RH / min, a three-level prevention and control mode is triggered, that is, the heating module is linked with the micro-positive pressure maintaining system and the emergency ventilation device to form a trinity of "wet removal-pressure maintenance-temperature stabilization" for emergency prevention and control.

[0009] Preferably, the gradient heating and dehumidifying mode realizes precise temperature control through space zoning and power distribution, and the cabinet is divided into three temperature zones of bottom, middle and top, and a high-power heat source is applied to the bottom temperature zone where moisture gathers.

[0010] The control algorithm in step S4 adopts a mixed architecture of PID and fuzzy control, and the heat source power regulation follows the PID algorithm: (6), wherein, P P is the heat source power, T d Td is the dew point temperature, T Ta is the air temperature in the electrical cabinet, the proportional coefficient k p and the integral coefficient k i According to the dynamic optimization of the temperature zone characteristics, the proportional coefficient k p determines the response strength of the heating power to the temperature deviation, and the integral coefficient k i is used to eliminate steady-state error and suppress long-term temperature fluctuations; The input variables of the fuzzy controller include the dew risk index value, the humidity gradient and the air pressure difference, which are fuzzified into "low risk", "medium risk" and "high risk" levels through membership functions, and the output variable is a comprehensive control instruction integrating the heating power, the fan speed and the opening degree of the ventilation port.

[0011] An active anti-dew system for an electrical cabinet applying the above method, comprising: a data acquisition unit for acquiring station room environment parameters, cabinet micro-environment parameters and cable trench state parameters; an execution mechanism including a heating module, a micro-positive pressure maintaining system, an emergency ventilation device and a condenser located in the electrical cabinet, which operates according to the received control instructions to adjust the environment parameters in the electrical cabinet; an edge computing module arranged in the condenser of the electrical cabinet, which collects the environment data of the electrical cabinet, and transmits the preprocessed data to the anti-dew intelligent gateway; an anti-dew intelligent gateway corresponding to each substation, which is communicatively connected to the condenser of each electrical cabinet and each sensor in the substation through an RS485 bus, aggregates the environment data in the electrical cabinet and the substation, uploads the data to a distributed data platform through an Ethernet, and forwards the control instructions to the execution mechanism; The distributed data center receives data uploaded by the anti-condensation intelligent gateway, obtains weather forecast data and historical data of the substation, performs multi-source data fusion, inputs the fused data into a pre-constructed condensation mechanism mathematical model to calculate a condensation risk index, determines a control mode according to the condensation risk index value, generates a control instruction for an executing mechanism, and forwards the control instruction to each executing mechanism through the anti-condensation intelligent gateway.

[0012] Further, the data acquisition unit includes a temperature sensor, a humidity sensor, an air pressure sensor and a wind speed sensor, which are distributed around the substation, in the electrical cabinet and in the cable trench.

[0013] The micro-positive pressure maintenance system is a variable frequency fan installed on the electrical cabinet, which is used to adjust the pressure difference between the inside and outside of the electrical cabinet, maintain the micro-positive pressure environment in the electrical cabinet, and automatically increase the pressure to the set threshold after the cabinet door is opened or closed or the emergency ventilation is started, to prevent the invasion of instantaneous moisture.

[0014] The emergency ventilation linkage device includes a variable frequency fan installed on the electrical cabinet and a ventilation valve located at the ventilation port, the variable frequency fan is used to forcibly remove moisture, and the ventilation valve is used to adjust the opening and closing and opening degree of the ventilation port, to maintain the micro-positive pressure environment in the electrical cabinet.

[0015] The beneficial effects of the present application are: 1. The present application realizes dynamic perception of key environmental factors such as temperature, humidity and air pressure by constructing a four-level monitoring parameter system covering environmental meteorology, station room environment, cabinet microenvironment and cable trench state; and processes various types of sensing data through multi-source data fusion to meet the calculation requirements of condensation risk index.

[0016] 2. The present application constructs a mathematical model based on the condensation formation mechanism, quantitatively describes the condensation phenomenon with spatiotemporal dynamic characteristics through the coupling equation of mass transfer and heat transfer, judges the condensation formation risk level by calculating the condensation risk index and setting threshold conditions, develops a hierarchical prevention and control strategy and a collaborative control algorithm, dynamically adjusts the prevention and control intensity according to the risk level, and significantly improves the risk response speed and resource utilization efficiency under complex working conditions. The system of the present application can effectively inhibit the generation of condensation in both steady and dynamic environments, while reducing energy consumption and false alarm rate. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a flowchart of the method of the present application; Figure 2 is a system block diagram of the present application. DETAILED DESCRIPTION

[0018] The implementation process of the present application will be described in detail below in conjunction with the drawings and examples.

[0019] An electrical cabinet active anti-condensation system realizes full-chain function integration through a hardware and software collaborative architecture. The hardware layer includes a data acquisition unit composed of multiple types of sensors, an actuator for adjusting the environmental parameters in the electrical cabinet, an edge computing module responsible for data preprocessing, an anti-condensation intelligent gateway responsible for data and control instructions, and a distributed data platform responsible for data analysis and generation of condensation risk levels and control instructions. The arrangement of each part is as follows: (I) Data acquisition unit: The data acquisition unit includes temperature sensors, humidity sensors, air pressure sensors, and wind speed sensors. Selecting appropriate sensor models and arranging them reasonably is the key to monitoring. For temperature sensors, high-precision probes (error less than 0.1℃) are selected and installed at locations such as electrical cabinets, substations, and cable trenches. For humidity sensors, fast-responding types (updating data within 5 seconds) are selected and preferentially arranged in areas where moisture easily accumulates, such as cabinet door gaps and cable entrances. For air pressure sensors, they are installed at the junction of cabinet vents and cable trenches to monitor the internal and external air pressure difference and determine the direction of moisture flow. For wind speed sensors, they are deployed at power distribution room vents and cabinet top vents to monitor air flow intensity and prevent local air stagnation from causing humidity to rise. When installing various sensors, the cabinet structure characteristics should be considered, such as adding sensors in closed corners or increasing the number of sensors around high-heat generating equipment to ensure full coverage.

[0020] (II) Actuator: The actuator includes heating modules, micro-positive pressure maintenance systems, emergency ventilation linkage devices, and condensers. The electrical cabinet is divided into three temperature zones: the bottom, middle, and top. Heating plates are installed in each temperature zone to gradiently heat and dehumidify according to the hierarchical control strategy. The micro-positive pressure maintenance system is a frequency conversion fan on the electrical cabinet that can be used to adjust the internal and external air pressure difference. The emergency ventilation linkage device includes a frequency conversion fan on the electrical cabinet and a valve at the ventilation port that can be used to forcibly remove moisture and adjust the opening and closing of the ventilation port to maintain a micro-positive pressure environment. The condenser is used to reduce the humidity in the electrical cabinet when condensation occurs.

[0021] (III) Edge computing module: The edge computing module is installed in the condenser of the electrical cabinet and is communicatively connected with the sensors in the electrical cabinet. It is used to collect environmental data in the electrical cabinet and perform preliminary processing, and then transmit the data to the anti-condensation intelligent gateway through the RS485 bus.

[0022] (IV) Anti-condensation intelligent gateway: Each substation is installed with an anti-condensation intelligent gateway, which is connected with the condensers in each electrical cabinet in the substation through RS485 bus, collects the environmental data of each electrical cabinet in the substation, and is also connected with each sensor at the substation, cable trench and the like, and collects the environmental data of the substation and the cable trench. The anti-condensation intelligent gateway also has edge computing capability, can preliminarily analyze the data and generate some control instructions for the actuators, but the priority is lower than that of the control instructions of the distributed data hub. The anti-condensation intelligent gateway is connected with the distributed data hub through 4G wireless network, transmits the collected data to the distributed data hub, and can also receive the control instructions sent by the distributed data hub and forward them to the actuators in each electrical cabinet to control the actuators to operate according to the instructions.

[0023] (Five) Distributed data hub: The distributed data hub has strong data analysis and storage capabilities, is connected with the anti-condensation intelligent gateway through 4G wireless network, receives the pre-processed environmental data, and also obtains weather forecast data through networking, is responsible for data analysis and processing, analyzes the condensation risk level, and generates control instructions for the actuators, which are sent to each actuator through the anti-condensation intelligent gateway.

[0024] The system realizes efficient and stable data transmission through multi-level technical architecture. The anti-condensation intelligent gateway adopts multi-channel synchronous control technology, supports parallel reading of multiple sensor signals, and guarantees real-time data with millisecond-level refresh rate. Data transmission is divided into two modes of short-distance in the cabinet and remote communication: shielded twisted pair is used for anti-interference wired transmission in the cabinet to eliminate electromagnetic noise through differential signal; remote communication is based on low-power wide-area network technology, combined with adaptive power adjustment and frequency hopping mechanism to ensure reliable connection within a coverage range of thousands of meters. The system realizes end-to-end security protection through channel encryption and cyclic redundancy check to meet the requirements of data reliability and real-time performance in industrial scenarios.

[0025] Before using the above system to perform active anti-condensation control, a mathematical model of the condensation mechanism needs to be constructed.

[0026] Physical process of condensation formation: Condensation is a liquefaction phenomenon of water vapor in the air on the surface of an object when the temperature is lower than the dew point temperature. When air contacts a low-temperature surface, its temperature drops below the dew point, the water vapor partial pressure exceeds the saturation vapor pressure at that temperature, and the excess water molecules gather into liquid water droplets on the surface. This process is directly related to the moisture holding capacity of the air, which is affected by temperature, relative humidity and air pressure.

[0027] The condensation process involves three key physical concepts: relative humidity (RH), dew point temperature (Td) and atmospheric pressure (P). RH T d ​​P ). Relative humidity (RH RH ) represents the percentage of water vapor content in the current air to the maximum water vapor content of the air at this temperature, reflecting the degree of air humidity. The dew point temperature (Td T d ) refers to the critical temperature when the air is cooled to water vapor saturation at constant pressure, at which the relative humidity reaches 100%. Atmospheric pressure (P P ) represents the pressure of air molecules per unit area, which fluctuates and affects the air's moisture holding capacity.

[0028] The model construction process is as follows: First, the improved empirical formula is used to calculate the dew point temperature, and the formula is as follows: (1), In formula (1), is the dew point temperature in the electrical cabinet, is the air temperature in the electrical cabinet, is the relative humidity in the electrical cabinet, is the standard atmospheric pressure, is the actual atmospheric pressure in the station room; the constants 243.12 and 17.62 are empirical values based on experimental data fitting of thermodynamic principles, the former is related to the triple point temperature of water, which extends the low-temperature applicability, and the latter is related to the evaporation latent heat of water, which ensures the calculation accuracy in the medium and high temperature zones.

[0029] Then, the water vapor transport equation and the temperature field evolution equation are constructed to quantitatively describe the condensation phenomenon with spatiotemporal dynamic characteristics. The water vapor transport equation is: (2), In formula (2), is the water vapor density in the cabinet; is the partial derivative of the water vapor density in the cabinet with respect to time; is the effective diffusion coefficient of comprehensive molecular diffusion and weak convection effect, which is simplified by experimental calibration to avoid complex turbulence modeling; the Laplace term describes the spatial gradient effect of water vapor diffusion, revealing the high-risk area of condensation; characterizes the phase change rate constant determined by the surface roughness and material properties, which is optimized by condensation rate experiment inversion; is a function that approximates the phase change threshold characteristics in a step form, balancing model accuracy and computational efficiency. When the local temperature is less than the dew point temperature, the function value is 1, triggering the condensation process; otherwise, the function value is 0. Other parameter meanings are the same as formula (1).

[0030] The temperature field evolution equation is: (3), In formula (3), is the air density; is the specific heat capacity at constant pressure; is the partial derivative of temperature with respect to time; is the latent heat of vaporization of water, representing the heat released / absorbed during the phase change, taken ; Laplace term describes the spatial gradient effect of heat conduction, revealing high-risk areas of condensation; is the thermal conductivity; other parameter meanings are the same as formulas (1), (2).

[0031] Formulas (2) and (3) are the mathematical model of condensation mechanism. The parameters obtained by applying the model are combined with formula (4) to output the condensation risk index CRI: (4), In formula (4), represents the integral region, i.e., the spatial volume inside the electrical cabinet; represents the temperature gradient, represents the modulus of the temperature gradient, reflecting the degree of local temperature change; both are used to quantify the spatiotemporal distribution characteristics of condensation risk. represents when the temperature inside the cabinet is lower than the dew point temperature , it takes the value of 1, triggering condensation.

[0032] The integral process of formula (4) is to combine the temperature gradient influence to comprehensively calculate the condensation risk index of the entire electrical cabinet for all areas inside the cabinet that meet the condensation triggering condition .

[0033] In the above model, when the temperature drops to the modified dew point temperature , the condensation process is activated, the water vapor density decays due to condensation consumption, and the rate is controlled by ; the latent heat released by condensation reversely inhibits temperature drop, forming a self-regulating mechanism. The Laplace terms and describe the spatial gradient effects of water vapor diffusion and heat conduction, respectively, revealing high-risk areas of condensation (such as the bottom of the cabinet and ventilation dead corners), and air pressure disturbances indirectly affect the triggering condition in the equation through the modified dew point temperature . The time derivative terms and in the equation capture the minute-level dynamic process of condensation formation, such as temperature and humidity transients caused by equipment start-stop or environmental mutations.

[0034] Overall, the model fully describes the dynamic feedback mechanism and spatiotemporal distribution of condensation process, and can predict the probability of condensation occurrence in real time by inputting relevant parameters.

[0035] Finally, after building the model, the model is verified by comparing the simulation results with the test data, and an error evaluation function is constructed based on the water vapor transport equation and the temperature field evolution equation , and the parameters are optimized based on the error evaluation function: (5), In formula (5), is the predicted value of the water vapor density in the cabinet; is the measured value of the water vapor density in the cabinet; and are the predicted value and the measured value of the air temperature in the cabinet, respectively; is the weight coefficient for balancing the contribution of humidity and temperature errors; represents the time sampling point.

[0036] By inverse optimization of the effective diffusion coefficient and the phase change rate constant , the values of and are solved by using formulas (2) and (3), and the evaluation function is solved by substituting formula (5), and the gradient descent method is used for iteration until the evaluation function reaches the minimum value, at which time the effective diffusion coefficient and the phase change rate constant are the optimal values.

[0037] The active anti-condensation process is as follows: S1, through the sensors distributed in the substations, electrical cabinets and cable trenches and the Internet, the environmental meteorological parameters, station room environmental parameters, cabinet micro-environmental parameters and cable trench state parameters are monitored.

[0038] The environmental meteorological parameters include external air temperature, relative humidity, atmospheric pressure and wind speed, which are used to capture the potential influence of the external environment on the substation; the station room parameters focus on the temperature and humidity and air flow rate in the substation, reflecting the microclimate state of the cabinet group; the cabinet parameters cover the surface temperature, internal temperature and humidity and local air pressure difference, which are directly related to the core conditions of condensation formation; the cable trench parameters monitor the humidity and water depth in the trench, and identify the risk of moisture penetration. The parameters are coupled through physical correlation, for example, the external wind speed fluctuation drives the change of the station room air flow rate, the cabinet air pressure difference and the cable trench humidity jointly affect the local dew point temperature, forming a complete monitoring link from the macro environment to the micro cabinet.

[0039] S2, the data collected by the data acquisition unit are fused, and the physical quantities are collaboratively analyzed through a hierarchical architecture during the fusion process.

[0040] Multi-source data fusion is divided into three steps: primary fusion, intermediate fusion, and advanced fusion.

[0041] (1) Primary fusion is used to eliminate noise from sensors of the same type. The weights are dynamically allocated based on the sensor accuracy, and the formula is as follows: , in The temperature estimate after fusion is used to characterize... Spatial average temperature at any given time; For the first One sensor in j The original temperature measurement at that moment; For the first The standard deviation of each sensor's measurement, determined by calibration experiments, reflects the sensor's accuracy. The standard deviation of measurements for the same type of sensor group is used for dynamic weight allocation.

[0042] (2) The intermediate integration establishes a multivariate regression model of temperature-air pressure-wind speed, and uses historical data to calibrate coefficients to realize abnormal data verification and missing data compensation.

[0043] The relative humidity prediction model is as follows: , in Represents the predicted relative humidity. This represents the air temperature inside the electrical cabinet. This represents the air pressure inside the electrical cabinet. Represents the airflow velocity inside the electrical cabinet, coefficient , , These are regression coefficients, representing the weight of each parameter's contribution to relative humidity. This is the intercept term, calibrated by fitting historical data, and the optimal coefficients are solved using the least squares method to ensure that the prediction error is less than [a certain value]. .

[0044] Anomaly detection: Real-time calculation of measured relative humidity values Compared with the predicted value residual ,like ( If the standard deviation of the historical residual is used, the sensor is determined to be abnormal and a data rejection or calibration process is triggered. Missing value compensation: When the sensor fails, the predicted value is used. Temporary replacement, labeled as "model estimation data".

[0045] (3) Advanced fusion introduces Kalman filtering to dynamically correct model bias caused by sudden environmental changes.

[0046] To cope with environmental mutations (such as the sudden change of temperature and humidity caused by the opening of the cabinet door), the system uses Kalman Filter to dynamically correct the sensor data, ensuring the accuracy and robustness of the model input. The implementation process is divided into two stages: prediction and update: Prediction stage (update over time): Based on the state estimation value of the last time And the state transition matrix , predict the state of the current time : , Among them, the state variable , including the current temperature And relative humidity .

[0047] The state transition matrix Is determined by modeling the dynamic characteristics of the environment (such as temperature and humidity trends).

[0048] Update stage (update with measurement): When the new sensor measured value (contains noise) arrives, calculate the Kalman gain : , : State prediction covariance matrix, reflecting the uncertainty of prediction; : Observation matrix, mapping the state space to the measurement space (usually a unit matrix); : Sensor noise covariance matrix, determined by sensor calibration experiment; Use Kalman gain to correct the predicted value to get the optimal estimate : , The correction term is the residual between the measured value and the predicted value, and the Kalman gain dynamically adjusts the correction weight.

[0049] S3, input the fused data into the pre-built condensation mechanism mathematical model to calculate the condensation risk index CRI.

[0050] S4, according to the obtained condensation risk index value (CRI), adopt a hierarchical response mechanism, dynamically adjust the prevention and control intensity according to the risk level, and realize the active defense of condensation.

[0051] The process of hierarchical response mechanism control execution mechanism is as follows: When the condensation risk index value is between 0.5 and 1.0, the basic temperature control mode is activated, using the equipment's own heat to maintain the cabinet temperature slightly higher than the dew point temperature, thus suppressing the initial formation of condensation.

[0052] If the condensation risk index drops to the range of 0.2 to 0.5, the system activates the gradient heating dehumidification mode, which heats the humidity gradient distribution inside the cabinet in zones, prioritizing the application of high-power heat sources to high-risk areas such as the bottom and cable trench entrances, while adjusting the heating intensity through closed-loop feedback to avoid local overheating.

[0053] When the condensation risk index value is below 0.2 or the humidity change rate exceeds 5%RH / min, a level 3 prevention and control mode is triggered, that is, the heating module, the micro positive pressure maintenance system, and the emergency ventilation device work together to block moisture penetration through forced air replacement and air pressure balance.

[0054] The micro-positive pressure maintenance system regulates the air pressure difference between the inside and outside of the cabinet via a variable frequency fan. The target pressure difference is set at 20-50 Pa. The air pressure sensor provides real-time feedback data to drive the fan speed adjustment, with a response time of less than 3 seconds. After the cabinet door is opened or closed, or emergency ventilation is activated, the system automatically pressurizes to the set threshold to prevent instantaneous moisture intrusion.

[0055] The emergency ventilation linkage mechanism is activated when the CRI drops below 0.2 or when humidity suddenly changes, and the variable frequency fan operates accordingly. Forced dehumidification is achieved by simultaneously closing unnecessary vents to maintain a slightly positive pressure environment. The heating system switches to constant power mode to offset the risk of sudden temperature drops caused by ventilation, forming a three-pronged emergency control logic of "dehumidification-pressure maintenance-temperature stabilization".

[0056] The system also takes into account seasonal adaptability, focusing on steady-state control of heating power in winter and strengthening the dynamic coordination of air pressure gradient and ventilation efficiency in the rainy season to achieve full-condition coverage.

[0057] The actuator collaborative control algorithm adopts a hybrid architecture of PID and fuzzy control, and the heat source power regulation follows the PID algorithm: (6), in, P For heat source power, T d Dew point temperature, T The air temperature inside the electrical cabinet, proportionality coefficient k p With integral coefficient k i Based on the dynamic optimization of temperature zone characteristics, when the temperature of a single zone exceeds 60℃, the power reduction protection is automatically triggered, and adjacent zones are compensated with heating to prevent condensation transfer.

[0058] PID parameters are tuned using empirical formulas: , , in the formula This is the highest permissible temperature inside the cabinet, typically set to 90% of the equipment's temperature resistance limit. This is the minimum allowable temperature inside the cabinet, which can be dynamically adjusted based on historical environmental data. To address dynamic temperature deviation and ensure rapid convergence and steady-state accuracy, it is defined as follows: ,in Average temperature inside the cabinet; proportionality coefficient Determines the intensity of the heating power's response to temperature deviation; integral coefficient Used to eliminate steady-state errors and suppress long-term temperature fluctuations.

[0059] The fuzzy controller's input variables include the condensation risk index, humidity gradient, and air pressure difference. These are fuzzified into "low risk," "medium risk," and "high risk" levels using membership functions. The output variable is a comprehensive control command that integrates heating power, fan speed, and vent opening, solving nonlinear optimization problems under multi-objective coupling. When sensors fail, the system automatically switches to model predictive control (MPC) mode, estimating key parameters based on historical data and real-time models to ensure control robustness under extreme conditions.

[0060] To verify the operational effectiveness of the above system, an experimental platform capable of simulating the operating environment of a real electrical cabinet was built. The main body of the platform is a customized environmental simulation chamber, which is equipped with the following subsystems: (1) Temperature and humidity control system: Based on semiconductor cooling chip and ultrasonic humidifier, it realizes rapid adjustment of temperature from -10℃ to 50℃ and humidity from 10% to 95%RH, with fluctuation range ; (2) Air pressure disturbance device: The centrifugal fan and the air pressure balancing valve work together to simulate the inside and outside of the cabinet. Dynamic pressure difference; (3) Condensation monitoring network: Fiber optic temperature sensors and capacitive humidity sensors are installed on the inner wall of the simulation cabinet, cable trench, and surface of key equipment. The data acquisition frequency is [not specified]. Simultaneously deploy thermal imagers to capture the spatial distribution of condensation; (4) Actuator array: The heating elements are installed in gradient zones, and the variable frequency fan and ventilation valve are integrated on the top of the cabin, supporting remote program control and manual intervention.

[0061] The platform uses LabVIEW to develop a central control interface, enabling environmental parameter settings, data visualization, and automated management of the experimental process.

[0062] Performance verification experiment: Based on the experimental platform, comparative group experiments of steady-state environment, dynamic disturbance and extreme condensation are carried out to verify the system prevention and control efficiency. In the steady-state environment test, the constant temperature and humidity conditions (25℃ / 80%RH) are set, and the difference between the dew point temperature and the actual temperature in the cabinet is monitored after the system is started to verify whether the basic temperature control module can maintain the safety margin ; In the dynamic disturbance test, the sudden change condition in the rainy season is simulated, and the response time of the system activated gradient heating and emergency ventilation and the humidity drop rate are recorded; In the extreme condensation test, the cable trench leakage is simulated by forcibly injecting water at the bottom of the cabinet, and the blocking effect of the micro-positive pressure maintenance system on the moisture penetration and the dehumidification efficiency of the bottom heating sheet are monitored. In the experiment process, the safety threshold is set ; The temperature and humidity in the dynamic working condition are simulated; The test period of the whole experiment process is 72 hours.

[0063] The verification experiment data of the system are shown in Table 1: Table 1 Comparison of condensation prevention and control system performance verification experiment data .

Claims

1. An active anti-condensation method for electrical cabinets based on a mathematical model of the condensation mechanism, characterized in that it comprises the following steps: Comprising the following steps: S1, monitoring environmental meteorological parameters, station room environmental parameters, electrical cabinet micro-environmental parameters and cable trench state parameters; S2, multi-source data fusion is carried out on the data collected in step S1, and physical quantity collaborative analysis is realized through hierarchical architecture in the fusion process; S3, input the data obtained in step S2 into the pre-constructed condensation mechanism mathematical model to calculate the condensation risk index; wherein the construction process of the condensation mechanism mathematical model is as follows: The dew point temperature is calculated by using the improved Magnus empirical formula, and the formula is as follows: （1）， In formula (1), T d Tdp is the dew point temperature in the electrical cabinet, T Tair is the air temperature in the electrical cabinet, RH RH is the relative humidity in the electrical cabinet, P 0 P0 is the standard atmospheric pressure, P P is the actual atmospheric pressure in the station room, and the constants 243.12 and 17.62 are empirical values based on experimental data fitting of thermodynamic principles. The transport equation of water vapor and the temperature field evolution equation are constructed to quantitatively describe the condensation phenomenon with the space-time dynamic characteristics, and the transport equation of water vapor is: (2), In equation (2), is the water vapor density in the cabinet, is the partial derivative of the water vapor density with respect to time; is the effective diffusion coefficient, which is a combination of molecular diffusion and weak convection effect, and is calibrated by experiments; Laplace term describes the spatial gradient effect of water vapor diffusion, revealing the high-risk area of condensation; characterizes the phase change rate constant determined by the surface roughness and material properties; is the Heaviside function, H(T d - T) indicates that when the temperature in the cabinet T is lower than the dew point temperature T d , it takes the value 1, triggering condensation; The meanings of other parameters are the same as those in formula (1); The temperature field evolution equation is: (3), In equation (3), is the air density, is the specific heat capacity at constant pressure, is the partial derivative of temperature with respect to time; is the latent heat of vaporization of water; the Laplace term describes the spatial gradient effect of heat conduction, revealing the high risk area of condensation; k is the thermal conductivity; other parameter meanings are the same as equations (1), (2); The formulas (2) and (3) are the mathematical models of condensation mechanism, and the parameters obtained by applying the models are combined with the formula (4) to output a condensation risk index CRI: (4), In formula (4), represents the integral region, i.e. the volume of the space inside the electrical cabinet; represents the temperature gradient, represents the modulus of the temperature gradient; formula (4) obtains the global condensation risk index by integrating the local conditions of each point inside the electrical cabinet; S4, according to the condensation risk index value obtained in step S3, a hierarchical response mechanism is adopted to cooperatively control the actuator, so as to realize active defense of condensation.

2. The active anti-condensation method of electrical cabinet based on the condensation mechanism mathematical model according to claim 1, characterized in that: The multi-source data fusion of step S2 is divided into three steps of primary fusion, intermediate fusion and advanced fusion, the primary fusion is used to eliminate the noise of the same type of sensor; the intermediate fusion establishes a multiple regression model of temperature-air pressure-wind speed, and the coefficients are calibrated through historical data to realize abnormal data checking and missing compensation; the advanced fusion introduces Kalman filter dynamic correction to model deviation caused by environmental mutation.

3. The active condensation prevention method of electrical cabinet based on the condensation mechanism mathematical model according to claim 1, characterized in that: The model building in step S3 also has a model verification and parameter optimization process; the model verification is realized by comparing the simulation results with the experimental data; the parameter optimization process is first based on the water vapor transport equation and the temperature field evolution equation to construct an error evaluation function : (5), In formula (5), is the predicted value of the water vapor density in the cabinet, is the measured value of the water vapor density in the cabinet; and are the predicted value and the measured value of the air temperature in the cabinet, respectively; is a weight coefficient for balancing the contribution of humidity and temperature errors, represents a time sampling point, is the number of sampling points; Optimizing effective diffusion coefficient by inversion and phase transformation rate constant , using formula (2), (3) to solve and , substituting formula (5) to solve evaluation function, using gradient descent method to iterate until evaluation function reaches minimum value, at this time effective diffusion coefficient D and phase transformation rate constant are optimal values.

4. The active anti-condensation method of electrical cabinet based on the condensation mechanism mathematical model according to claim 1, characterized in that: The process of step S4 for controlling the actuator by adopting a hierarchical response mechanism is as follows: When the condensation risk index value is between 0.5 and 1.0, start the basic temperature control mode, use the heat generated by the equipment to maintain the temperature in the cabinet slightly higher than the dew point temperature, and inhibit the initial condensation; If the condensation risk index value decreases to 0.2 to 0.5, activate the gradient heating and dehumidification mode, and perform zoned heating according to the humidity gradient distribution in the cabinet, while adjusting the heating intensity through closed-loop feedback to avoid local overheating; When the condensation risk index value is less than 0.2 or the humidity mutation rate exceeds 5%RH / min, trigger the third-level prevention and control mode, that is, the heating module is linked with the micro-positive pressure maintenance system and the emergency ventilation device to form a "wet removal-pressure maintenance-temperature stabilization" trinity emergency prevention and control.

5. The active anti-condensation method of electrical cabinet based on the condensation mechanism mathematical model according to claim 4, characterized in that: The gradient heating and dehumidification mode realizes precise temperature control through spatial zoning and power distribution, and the cabinet is divided into three temperature zones of bottom, middle and top, and a high-power heat source is applied to the bottom humidity accumulation area.

6. The active condensation prevention method of electrical cabinet based on the condensation mechanism mathematical model according to claim 1 or 4, characterized in that: The control algorithm in step S4 adopts a mixed architecture of PID and fuzzy control, and the heat source power regulation follows the PID algorithm: (6), wherein, P is the heat source power, T d is the dew point temperature, T is the air temperature in the electrical cabinet, the proportional coefficient k p is the integral coefficient k i is the dynamic optimization according to the temperature zone characteristics, the proportional coefficient k p determines the response strength of the heating power to the temperature deviation, the integral coefficient k i is used to eliminate the steady-state error and suppress long-term temperature fluctuations; The input variables of the fuzzy controller include the condensation risk index value, humidity gradient and air pressure difference, which are fuzzified into "low risk", "medium risk" and "high risk" levels through membership functions, and the output variable is the comprehensive control instruction of fused heating power, fan speed and ventilation opening degree.

7. An active anti-condensation system for electrical cabinets applying the method according to any one of claims 1 to 6, characterized in that It comprises: A data acquisition unit for acquiring station room environmental parameters, cabinet micro-environmental parameters and cable trench state parameters; An actuator including a heating module, a micro-positive pressure maintenance system, an emergency ventilation device and a condenser located in the electrical cabinet, which operates according to the received control instructions to adjust the environmental parameters in the electrical cabinet; An edge computing module arranged in the condenser of the electrical cabinet, used to collect the environmental data of the electrical cabinet, and transmit the preprocessed data to the anti-condensation intelligent gateway; An anti-condensation intelligent gateway corresponding to each substation, which is connected with the condensers of each electrical cabinet and each sensor in the substation through RS485 bus, aggregates the environmental data in the electrical cabinet and the substation, uploads the data to the distributed data center through Ethernet, and forwards the control instructions to the actuator; The distributed data platform receives data uploaded by the anti-condensation intelligent gateway, obtains weather forecast data and historical data of the substation, performs multi-source data fusion, inputs the fused data into a pre-constructed condensation mechanism mathematical model to calculate a condensation risk index, determines a control mode according to the condensation risk index value, generates a control instruction for an execution mechanism, and forwards the control instruction to each execution mechanism through the anti-condensation intelligent gateway.

8. The active de-frosting system for electrical cabinets according to claim 7, characterized in that: The data acquisition unit includes temperature sensors, humidity sensors, air pressure sensors and wind speed sensors, which are distributed around the substation, in the electrical cabinet and in the cable trench.

9. The active de-frosting system for electrical cabinets according to claim 7, characterized in that: The micro-positive pressure maintaining system is a variable frequency fan installed on the electrical cabinet, which is used to adjust the pressure difference between the inside and outside of the electrical cabinet, maintain the micro-positive pressure environment in the electrical cabinet, and automatically increase the pressure to the set threshold after the cabinet door is opened or closed or the emergency ventilation is started, to prevent the invasion of instantaneous moisture.

10. The active de-frosting system for electrical cabinets as claimed in claim 7, wherein: The emergency ventilation linkage device includes a variable frequency fan installed on the electrical cabinet and a ventilation valve located at the ventilation port, the variable frequency fan is used to forcibly remove moisture, and the ventilation valve is used to adjust the opening and closing and opening degree of the ventilation port, to maintain the micro-positive pressure environment in the electrical cabinet.

Images