A switch cabinet microcirculation dehumidification system with double humidity thresholds and a control method thereof

Abstract

The present application relates to the technical field of switch cabinet anti-condensation, and discloses a switch cabinet micro-circulation dehumidification system with double humidity thresholds and a control method. An air inlet assembly sends air through existing fastening holes of a cabinet door, and a distal end exhaust path and a check element form a directional micro-positive pressure air path. A controller generates air supply permission based on humidity thresholds inside and outside the cabinet, humidity content difference and dew point margin, forms two-dimensional heterogeneous excitation with at least two effective air supply quality levels and two shutdown balancing time levels, performs double-parameter joint identification on the distal end participation coefficient and continuous moisture release rate, and reversely calculates subsequent effective air supply quality; the air inlet channel is closed when there is no positive dehumidification feasible solution. The present application can separate the influence of insufficient air flow and continuous moisture source, and inhibit moisture introduction, backflow and ineffective air supply.

Description

Technical Field

[0001] This invention belongs to the field of power equipment moisture protection and environmental control technology, specifically relating to a switchgear micro-circulation dehumidification system and control method with dual humidity thresholds. Background Technology

[0002] Medium-voltage switchgear typically consists of relatively separated spaces such as busbar compartments, circuit breaker compartments, and cable compartments. When the cabinet doors are closed, the natural ventilation is relatively small. Moisture buildup in cable trenches, diurnal temperature variations, and the infiltration of humid air from the switchgear compartment cause water vapor to slowly accumulate inside the cabinet. When the temperature of the insulation components or metal walls is lower than the dew point of the surrounding air, condensation easily forms, reducing the surface insulation margin. Existing solutions include switchgear compartment air conditioning or dehumidifiers, cabinet heaters, semiconductor condensation dehumidifiers, and fan ventilation. However, it is difficult to reliably introduce dry air into the dead zones inside the cabinet; heating mainly changes relative humidity but does not necessarily reduce the total amount of water vapor; condensation dehumidification requires drainage and occupies cabinet space; direct ventilation is often started and stopped based on the humidity inside a single cabinet, without considering whether the outside air has a low moisture content. Relative humidity is affected by temperature; a lower relative humidity outside the cabinet does not necessarily mean a lower moisture content, and introducing warm, humid air may induce condensation on low-temperature surfaces inside the cabinet. Adding additional openings and arranging large air ducts can affect cabinet protection, and continuous air supply can easily cause airflow short circuits and backflow after shutdown. Existing feedback methods often evaluate airflow based on the amount of humidity reduction in a single cycle. Insufficient humidity reduction at the far end could stem from inadequate airflow arrival, or from moisture backflow from cable trenches or material release; these factors lead to different adjustments in the airflow volume. If the degree of airflow participation is not separated from the continuous moisture source through different airflow stimuli, the controller may mask the moisture source or airflow path malfunction by extending the airflow. Therefore, a micro-circulation dehumidification scheme is needed that can verify the actual drying potential, identify far-end transport and continuous moisture release, and deduce an appropriate airflow volume. Summary of the Invention

[0003] In view of this, the present invention provides a micro-circulation dehumidification system and control method for switch cabinets with dual humidity thresholds. Without damaging the openings in the cabinet, it directs outside air with drying potential to the humid area inside the cabinet. By forming a distinguishable humidity response through two-dimensional heterogeneous excitation, it performs dual-parameter joint identification of the airflow arrival degree and the continuous water vapor source, and then performs reverse calculation of the air supply quality to solve the problems of moisture induction by direct ventilation, backflow during shutdown, and confusion between abnormal air path and continuous back dampness.

[0004] To achieve the above objectives, this invention provides a micro-circulation dehumidification system for switchgear with dual humidity thresholds. The system includes a switchgear, an intake micro-circulation component, an exhaust path, an internal humidity sensing node, an external humidity sensing node, a differential pressure sensing node, and a controller. The intake micro-circulation component communicates with the inside of the switchgear via an existing fastening hole in the switchgear door and is equipped with a fan, a filter, an intake check valve, and a guide component pointing towards the far end of the airflow. The exhaust path is located at the far end of the airflow and communicates with the outside of the switchgear via at least one of the following structures: an existing pressure relief channel, an existing exhaust gap, or another existing fastening hole. The controller generates an air supply permit based on the starting humidity inside the switchgear, the permissible humidity outside the switchgear, the difference in humidity between the inside and outside of the switchgear, and the temperature margin of the external air dew point relative to the low-temperature surface inside the switchgear. During the period when the permit is valid, an air supply replacement phase and a shutdown balancing phase are executed alternately, accumulating only the effective air supply quality within a preset micro-positive pressure range, and forming at least four cross-combinations of two-dimensional heterogeneous excitation control wheels with at least two effective air supply quality levels and at least two shutdown balancing duration levels. The controller employs a water vapor mass balance model to jointly identify the remote participation coefficient and continuous moisture release rate using two parameters. Based on the identification results and the target moisture content reduction, it calculates the subsequent effective air supply quality. If no feasible solution for forward dehumidification exists within the allowable air supply quality range, the humidity inside the cabinet reaches the cabinet's stop threshold, or any safety condition in the air supply permit fails, the controller stops the fan and closes the air intake passage with the air intake check valve.

[0005] Furthermore, in the above system, the intake microcirculation component includes a mounting base located on the outside of the cabinet door, a connecting sleeve passing through an existing fastening hole, and a sealing pressure ring located on the inside of the cabinet door. The filter, fan, intake check valve, rectifier section, and guide component are connected sequentially along the intake direction. The exhaust path includes an exhaust connector fitted into another existing fastening hole, an exhaust check valve that opens only to the outside of the cabinet, and a hydrophobic and breathable membrane located on the outside of the exhaust check valve. After the fan stops, the intake check valve retracts as the fan outlet pressure difference disappears, and the exhaust check valve retracts as the pressure difference between the inside and outside of the cabinet decreases; both remain closed after retracting, and the air already supplied into the cabinet continues to diffuse through the internal partition gap in the shutdown equalization section, preventing humid air from the outside of the cabinet from flowing back through the check valves.

[0006] Furthermore, in the above system, the humidity sensing node inside the cabinet adopts a load-bearing hollow bolt body that replaces the original fasteners of the cabinet door. The hollow bolt body has an axial sampling hole communicating with the inside of the cabinet. A porous protective cover is installed at the inner end of the axial sampling hole, and the temperature and humidity probe is located between the porous protective cover and the axial sampling hole. The outer end of the hollow bolt body is equipped with a signal processing cavity connected to the temperature and humidity probe, and a sealing isolation component is installed between the signal processing cavity and the axial sampling hole. The hollow bolt body is also equipped with a temperature contact plate that is in thermal contact with the metal surface of the cabinet door to obtain the temperature of the low-temperature surface inside the cabinet. The controller obtains the humidity content inside the cabinet based on the temperature and relative humidity inside the cabinet, and obtains the humidity content outside the cabinet and the dew point of the air outside the cabinet based on the temperature and relative humidity outside the cabinet. When the difference between the humidity content inside the cabinet and the humidity content outside the cabinet is not less than the preset humidity difference, and the difference between the temperature of the low-temperature surface inside the cabinet and the dew point of the air outside the cabinet is not less than the preset dew point safety margin, the drying potential condition in the air supply permit is determined to be met.

[0007] Furthermore, in the aforementioned system, the two-dimensional heterogeneous excitation includes a cross combination of low effective supply air quality, high effective supply air quality, short shutdown equilibrium duration, and long shutdown equilibrium duration. There are at least two control wheels with the same effective supply air quality but different shutdown equilibrium durations, and at least two control wheels with the same observation duration but different effective supply air quality. Each observation cycle begins at the air supply replacement section and continues until the end of the adjacent shutdown equilibrium section. The measured decrease in moisture content is the difference in moisture content at the far end of the airflow between the observation start and end points. The supply air contribution in the water vapor mass balance model tends to saturate exponentially with the effective supply air quality, while the continuous moisture release contribution increases linearly with the observation duration. The controller configures the weights of each cycle based on the effective pressure ratio and the stability of the moisture content outside the cabinet, minimizing the weighted sum of squared residuals between the measured values ​​and the model values ​​within preset boundaries. The actual drying potential of each cycle is measured separately and substituted into the model, without assuming that the moisture content inside the cabinet remains constant during the identification period.

[0008] Furthermore, in the aforementioned system, the controller substitutes the remote participation coefficient, continuous moisture release rate, the initial remote drying potential for the next round, the effective mass flow rate, and the downtime equalization period into the net moisture content reduction function. Within the allowable supply air mass range where this function remains monotonically increasing, the minimum effective supply air mass required to achieve the target moisture content reduction is calculated. If the upper limit of the range still cannot reach the target but a positive net reduction can still be achieved, the upper limit of the range is used, and insufficient capacity information is output. If no positive net reduction can be achieved throughout the entire range, the supply air permit is revoked, and information on insufficient external drying potential or abnormal continuous moisture release is output. The micro-positive pressure establishment time and the remote participation coefficient together indicate leakage, blockage, non-opening check valve, near-end airflow short circuit, or abnormal resistance in the separation path. When the continuous moisture release rate exceeds the preset upper limit, moisture source inspection information is output.

[0009] Furthermore, the system also includes a power supply component and a communication component. The power supply component has an external DC power supply mode and a battery power supply mode. In battery power supply mode, the controller limits the cumulative effective air supply quality per unit time based on the remaining battery energy. When the remaining energy is lower than a preset energy threshold, the humidity acquisition, backflow prevention shutdown, and anomaly reporting functions are retained. When the humidity sensing node inside or outside the cabinet goes out of bounds, loses connection, or has an abnormal rate of change, the controller revokes the air supply permission and does not update the remote participation coefficient and continuous moisture release rate based on the abnormal cycle.

[0010] This invention also provides a micro-circulation dehumidification control method for switchgear with dual humidity thresholds. This method uses an air intake micro-circulation component that supplies air through existing fastening holes in the switchgear door, allowing the gas inside the cabinet to be discharged via an exhaust path located at the far end of the airflow. During operation, the relative humidity inside and outside the cabinet, the temperature inside and outside the cabinet, the low-temperature surface temperature inside the cabinet, and the differential pressure inside and outside the cabinet are collected. An air supply permit is generated when dual humidity conditions, moisture content direction conditions, and dew point safety conditions are all met. After the permit is established, at least four two-dimensional heterogeneous excitation control wheels are formed with at least two effective air supply quality levels and at least two shutdown equilibrium duration levels. The far-end participation coefficient and continuous moisture release rate are jointly identified using dual parameters, and the subsequent effective air supply quality is calculated inversely based on the target moisture content decrease. The inverse calculation restricts the allowable air supply quality range, pressure range, and model monotonicity interval. If there is no feasible solution for forward dehumidification, the humidity inside the cabinet reaches the cabinet stop threshold, or the safety conditions or data validity conditions in the air supply permit are not met, the air supply is stopped and the air intake channel is closed.

[0011] Furthermore, in the above control method, the cabinet start threshold is higher than the cabinet stop threshold. Air supply permission is revoked when the external relative humidity exceeds the external permissible threshold and remains above the preset revocation confirmation time; air supply permission is only allowed to be regenerated after the external relative humidity decreases to a recovery value below the external permissible threshold and remains below the preset recovery confirmation time. Moisture content is calculated based on relative humidity, temperature, saturated water vapor pressure, and atmospheric pressure. Even if the difference between the internal and external moisture content is positive, air supply is still prohibited if this difference is less than the preset moisture content difference. Internal demand, initial external permission, actual water vapor migration direction, and the risk of low-temperature surface condensation all contribute to the air supply decision.

[0012] Furthermore, in the above control method, for each candidate value of the remote participation coefficient, a candidate value of the continuous moisture release rate constrained by upper and lower limits is calculated based on the observation duration of each round and the model residuals. Then, the parameter set that minimizes the sum of squared weighted residuals is selected. If the number of identified samples is less than 4 rounds, the effective air supply quality or the shutdown balancing time has not reached at least 2 levels respectively, the change in the humidity outside the cabinet exceeds the stability limit, the effective pressure ratio is insufficient, or there is a sudden change in the remote temperature, the previous set of effective parameters is maintained, and the model is not updated with this set of samples.

[0013] Furthermore, in the above control method, the target moisture content decrease is no greater than the difference between the current remote moisture content and the target moisture content at the stop point. The controller constructs the net decrease corresponding to the candidate supply air quality based on the remote participation coefficient and the continuous moisture release rate, and uses a bisection method to find the minimum value that satisfies the target within the monotonic interval. If the target is unattainable but the upper limit of the interval still has a positive dehumidification effect, the controller operates at the upper limit of the interval and reports insufficient capacity; if the upper limit of the interval does not have a positive dehumidification effect, the fan is stopped. When the micro-positive pressure is established slowly and accompanied by a decrease in the remote participation coefficient, the external air path is checked; when the micro-positive pressure is normal but the remote participation coefficient decreases, the internal passage is checked; when the remote participation coefficient is normal but the continuous moisture release rate increases, the moisture source inside the cabinet is checked.

[0014] By adopting the above technical solutions, the present invention achieves the following beneficial effects: the intake microcirculation component is fixed and connected to the air passage by means of the existing fastening holes in the cabinet door, reducing the impact of on-site openings on the cabinet's sealing, corrosion prevention, and internal cleanliness; the humidity thresholds inside and outside the cabinet limit the operating range, the moisture content difference eliminates the misjudgment of relative humidity caused by temperature differences, and the dew point margin prevents air that may condense on low-temperature surfaces from entering; the intake check valve, the remote exhaust path, and the micro-positive pressure control allow dry air to cross the partitioned area inside the cabinet and prevent backflow after shutdown. The controller uses at least two effective air supply quality levels and at least two shutdown equilibrium duration levels to form a two-dimensional heterogeneous excitation, making the transport response that changes nonlinearly with the air supply quality and the moisture release response that changes linearly with the observation duration form distinguishable data characteristics; the measured drying potential of each round is respectively entered into the water vapor mass balance model, and the remote participation coefficient and continuous moisture release rate are jointly identified by two parameters, avoiding the confusion of insufficient airflow arrival and continuous backflow as the same cause. The subsequent effective air supply quality is calculated inversely based on the identified parameters, current drying potential, and target descent rate, adapting to changes in cabinet type, airflow resistance, and moisture source. The inverse calculation interval verifies monotonicity and forward dehumidification capability; if the target cannot be reached, restricted operation or shutdown is implemented to prevent ineffective extension of air supply. Differential pressure settling time, along with two identified parameters, indicates the abnormal direction of the external airflow path, internal cabinet passage, or continuous moisture source. The load-bearing hollow bolt joints accommodate cabinet door fastening, remote sampling, and low-temperature surface temperature measurement. Even in low-energy states, monitoring, check valves, and anomaly reporting are maintained, thus ensuring dehumidification effectiveness, condensation safety, diagnostic interpretability, and energy utilization. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the cabinet installation and directional airflow principle of the micro-circulation dehumidification system for switchgear of the present invention; Figure 2 This is a cross-sectional schematic diagram illustrating the installation principle of the intake microcirculation component and the exhaust component of the present invention using existing fastening holes. Figure 3 This is a schematic diagram of the axial cross-sectional principle of the load-bearing hollow bolt humidity sensing node of the present invention. Figure 4This is a schematic diagram showing the relationship between the air intake, sensing, and exhaust positions in different compartments of the switchgear according to the present invention. Figure 5 (a) is the mass conservation model calculation curve of relative humidity inside the cabinet under stable external conditions, and (b) is the model calculation curve of two-dimensional heterogeneous excitation, dual-parameter joint identification and effective air supply quality inverse calculation. Figure 6 In the middle (a), the control calculation curve of the dual humidity threshold is shown when the humidity outside the cabinet changes, and (b) is the calculation curve of the corresponding air supply permit and normalized fan drive quantity. Detailed Implementation

[0016] The following combination Figures 1 to 6 Specify the feasible structure and control parameters. The listed dimensions, thresholds, and durations can be adjusted equivalently based on the switchgear volume, existing fastening hole specifications, switchgear climate, and allowable energy consumption; the adjusted operating conditions should still meet the following requirements: the humidity of the air outside the cabinet is lower than that inside the cabinet, the low-temperature surface temperature inside the cabinet is higher than the dew point of the air to be introduced, and the airflow reaches the far exhaust side from the intake side.

[0017] like Figure 1 As shown, switchgear 1 can be a metal-clad withdrawable switchgear with a rated voltage of 12 kV. The cabinet contains a circuit breaker compartment 11, a busbar compartment 12, and a cable compartment 13. The cabinet door 14 has multiple existing fastening holes 15 around its perimeter for fixing panels, sealing plates, or accessories. An air intake micro-circulation assembly 2 is installed at the lower part of the cabinet door 14 near the cable compartment 13. An exhaust assembly 3 is installed at the upper part of the cabinet door 14 or utilizes the existing pressure relief channel of switchgear 1 to form an exhaust path. The humidity sensing node 4 inside the cabinet is arranged at the far end of the airflow relative to the air intake position. Dry air first enters the cable compartment 13, then migrates to the circuit breaker compartment 11 and busbar compartment 12 through the existing ventilation holes in the partition, assembly gaps, or air ducts that do not affect electrical isolation, and is finally discharged from the far-end exhaust path. The air intake and exhaust points are far apart, allowing air to cross easily humidified partition spaces and reducing short-circuit airflow that would otherwise return directly to the outside of the cabinet from adjacent gaps. For cabinet types that do not allow cross-compartment communication, one set of components can be configured in each low-pressure or cable separation space that allows independent ventilation, and the controller drives each component in a time-sharing manner.

[0018] Before installation, the switchgear manufacturing drawings, internal electrical clearances, and cabinet type test constraints are checked. Existing fastening holes 15 located on the grounded metal door panel and not serving the critical function of pressure relief and locking are selected. The air intake micro-circulation component 2 and the humidity sensing node 4 inside the cabinet do not extend into the insulation clearance range of live conductors. The internal air guiding components are made of flame-retardant insulating materials, and the fan, filter, and control circuit are all protected from the outside of the door panel. Using existing fastening holes 15 can prevent cutting metal shavings from entering the cabinet and eliminates the need for anti-corrosion and sealing treatment of newly added openings. Existing fastening holes 15 are holes that already exist in the switchgear's factory structure and were originally for bolts or screws to pass through, excluding holes temporarily processed during the installation of this system. When the original fasteners are used for panel clamping, the connecting sleeve 21 or the load-bearing hollow bolt body 41 simultaneously undertakes the original clamping function, and flexible hoses that only pass through the opening without forming a load-bearing clamping function are not used.

[0019] Figure 2 The cross-sectional relationship of the intake and exhaust components is shown. The intake micro-circulation component 2 includes a connecting sleeve 21, a sealing ring 22, a filter element 23, an intake check valve 24, a miniature brushless fan 25, a rectifier section 26, and a guide element 27. The connecting sleeve 21 has a replaceable outer sleeve that is compatible with existing fastening holes of M10 or M12. The mounting seat on the outside of the cabinet door 14 and the sealing ring 22 on the inside of the cabinet door 14 together clamp the door panel, and each clamping surface is equipped with an aging-resistant silicone rubber sealing ring. The effective flow diameter of the connecting sleeve 21 can be 8 mm to 10 mm; when a larger air exchange volume is required, two adjacent existing fastening holes can be used in parallel on the external collector seat to prevent the flow velocity of a single hole from being too high. The filter element 23 is located on the intake side of the fan and is preferably a combination of a replaceable fiber filter layer and a hydrophobic microporous membrane. It first intercepts dust and droplets in the air of the switch chamber and then allows water vapor to pass through with the air. The miniature brushless fan 25 has a rated voltage of 24 volts, a rated power of 0.8 watts, a rated free flow of 2.4 cubic meters per hour, and changes the actual air volume through pulse width modulation (PWM).

[0020] The inlet check valve 24 is located downstream of the fan and uses a duckbill-shaped silicone rubber valve or a thin-plate spring valve, with an opening pressure differential of 6 to 10 Pa. The valve opens when the fan establishes a positive pressure differential, and returns to its original position elastically after the fan stops, preventing the backflow of high-humidity air from outside the cabinet along the inlet channel. The rectifier section 26 consists of short, parallel straight channels used to weaken the swirling flow generated by the fan impeller. The guide component 27 forms a flat diffuser or connects to a flame-retardant flexible air duct inside the cabinet. The jet axis of the diffuser avoids insulating surfaces and exposed terminals and points towards the far end of the airflow. The arrangement of components along the inlet direction balances filtration, shutdown isolation, and diffusion: the filter 23 is located upstream of the fan to protect the impeller; the inlet check valve 24 is located downstream of the fan to isolate the fan cavity; and the rectifier section 26 and the diffuser convert the concentrated airflow within the connecting sleeve into a wider-coverage airflow within the cabinet.

[0021] The exhaust assembly 3 includes an exhaust connector 31, an exhaust check valve 32, and a hydrophobic and breathable membrane 33. The exhaust connector 31 is installed via an existing fastening hole. The exhaust check valve 32 only opens outwards when the pressure inside the cabinet is higher than outside; the opening pressure difference can be 4 to 8 Pa. The hydrophobic and breathable membrane 33 is located on the outside to prevent cleaning water droplets or condensate from entering the cabinet through the exhaust connector. When the fan is running, the controller adjusts the drive amount based on the pressure difference between the inside and outside of the cabinet measured by the differential pressure sensor, maintaining a slight positive pressure of 8 to 25 Pa inside the cabinet. This pressure is sufficient to drive air outwards along the remote exhaust path, but is significantly lower than the pressure that would lift the door panel seal or affect the mechanism's operation. If the cabinet already has a remote pressure relief channel that meets protection requirements, the exhaust connector 31 does not need to be installed separately; instead, a maintainable dustproof component can be added to the outside of this channel while retaining the original pressure relief function. No additional components must obstruct the pressure relief of a fault arc; this is a rigid constraint when selecting the exhaust location.

[0022] Figure 3 The humidity sensing node 4 inside the cabinet shown adopts a load-bearing hollow bolt structure. The hollow bolt body 41 can be made of corrosion-resistant 40Cr alloy steel with an external thread specification of M12 and a strength grade of not less than 8.8. The diameter of the axial sampling hole 42 is 3.2 mm. The nominal tensile stress area of ​​the M12 thread is approximately 84.3 square millimeters. After deducting the cross-sectional area of ​​the 3.2 mm diameter sampling hole of approximately 8.0 square millimeters, approximately 90% of the effective load-bearing area is still retained, which is sufficient to cover the clamping requirements of ordinary cabinet door panels. This value is only used to illustrate the mechanical basis for the hole diameter selection. The actual assembly is still checked according to the tightening torque of the corresponding cabinet type and the manufacturer's requirements. The inner end of the axial sampling hole 42 is equipped with a perforated protective cover 44. The protective cover allows the air inside the cabinet to diffuse to the temperature and humidity probe 43, while preventing wire ends, dust clumps, or maintenance tools from directly touching the probe.

[0023] The temperature and humidity probe 43 is a digital device with a measurement range of 0%RH to 100%RH, a typical error of no more than ±2%RH, and a temperature error of no more than ±0.3 degrees Celsius. The probe's shape is controlled within 2.5 mm square, allowing it to be installed in the inner cavity of the axial sampling hole 42. The signal processing cavity 45 is located on the bolt head outside the cabinet and houses a low-power microcontroller, voltage regulator circuit, and wired interface or low-power wireless transceiver. A sealing isolator separates the signal processing cavity 45 from the sampling section that directly contacts the air inside the cabinet, allowing the sensor leads to pass only through glass-bonded terminals or potted lead holes. The antenna 46 is located on the outside of the grounded door panel to prevent the metal cabinet from completely shielding the wireless signal. The temperature contact piece 47 is pressed against the inner surface of the cabinet door 14 or the lower-temperature metal of the door frame to obtain the low-temperature surface temperature inside the cabinet. The sealing rings 48 are located on both sides of the door panel, serving both moisture sealing and anti-loosening purposes. In environments where wireless communication is restricted, the signal processing cavity 45 is connected to the controller via a shielded twisted-pair cable laid against the door panel, and the cable does not enter the high-voltage compartment.

[0024] Figure 4 The relationship between measuring points in the partitioned spaces is shown. At least one humidity sensing node 4 inside the cabinet is positioned furthest from the air inlet diffuser and most susceptible to moisture from the cable trench. External humidity sensing nodes are located 50 mm to 300 mm upstream of the filter element 23 and are equipped with ventilation and anti-drip covers to avoid direct impact from the exhaust airflow from the exhaust assembly 3. When multiple partitioned spaces need to be assessed, nodes can be arranged in permissible locations in the busbar compartment 12, circuit breaker compartment 11, and cable compartment 13. The controller takes the highest effective relative humidity among the nodes as the humidity inside the cabinet under dual humidity gating, while retaining the independent variation of each node to determine whether the airflow has reached the far end. The moisture content of each node is calculated from the relative humidity and air temperature at the same time stamp for that node; the relative humidity of one node is not combined with the temperature of another node. The highest value can retain local humidity information; using a simple average value may lower the characteristic value of local humid areas and delay activation due to data from dry areas. When only one node is installed, the node is preferentially positioned at historical condensation points, lower corners of the cabinet, or the end of the air path, not directly adjacent to the air inlet.

[0025] The controller reads the relative humidity inside and outside the cabinet, the air temperature inside and outside the cabinet, the low-temperature surface temperature inside the cabinet, and the differential pressure between the inside and outside of the cabinet at a sampling period of 2 seconds. Each relative humidity channel is first filtered by a 5-point causal median filter consisting of the current sample and the previous 4 samples to remove individual communication glitches, and then a first-order recursive smoothing is performed.

[0026] The input after the kth median filter is ; In the formula, This represents the initial relative humidity obtained in the k-th iteration. This represents the median of the 5 points corresponding to the k-th sample, where `median` represents the median of the 5 inputs. This window does not reference future samples; the controller can obtain the result immediately after the current sample is completed.

[0027] Recursive smoothing value ; In the formula, This represents the smoothed relative humidity corresponding to the k-th sample. This indicates the result of the previous smoothing process. The smoothing coefficient is set to 0.2. Both humidity thresholds are compared with the smoothing value, and instantaneous spikes will not directly trigger the fan. All channels inside and outside the cabinet use the same timestamp. If the time difference between the collection of any two data points is greater than 4 seconds, the data in this round will not be included in the humidity comparison to avoid combining two sets of data before and after rapid weather changes into a non-existent air condition.

[0028] Comparing relative humidity alone is insufficient to determine whether the supplied air will remove moisture. For example, the moisture content of air at 24 degrees Celsius and 75% RH is approximately 14.02 g / kg dry air, while the moisture content of air inside the cabinet at 26 degrees Celsius and 65% RH is approximately 13.68 g / kg dry air. If only relative humidity is considered, the external data still falls within the permissible 75% RH limit, but the actual introduced air is more humid. Therefore, the controller calculates the saturated water vapor pressure, actual water vapor partial pressure, and moisture content based on temperature and relative humidity.

[0029] Within the applicable temperature range of -20°C to 50°C, the saturated water vapor pressure is approximately... ; In the formula, This represents the saturated water vapor pressure corresponding to temperature T, in kilopascals. represents air temperature in degrees Celsius; exp represents an exponential function with the natural constant as its base.

[0030] The actual partial pressure of water vapor is ; In the formula, This represents the actual partial pressure of water vapor, expressed in kilopascals (kPa). This represents a percentage value for relative humidity.

[0031] air humidity content is ; In the formula, It indicates the mass of water vapor contained in one kilogram of dry air, with the unit being kilograms of dry air; This represents the atmospheric pressure at the site, in kPa, which can be obtained from a barometric pressure sensor. If no barometric pressure sensor is installed, 101.325 kPa is used at low-altitude switch stations. The moisture content inside and outside the cabinet are recorded as follows: and .

[0032] The dew point of the air outside the cabinet is ; In the formula, The dew point of the air outside the cabinet is expressed in degrees Celsius; ln represents the natural logarithm. It represents the water vapor partial pressure obtained from the outside temperature and relative humidity of the cabinet.

[0033] The difference in moisture content inside and outside the cabinet is defined as ; Dew point safety margin is defined as ; In the formula, This indicates the difference in moisture content between the inside and outside of the cabinet; Indicates the safety margin at the dew point; This represents the temperature of the low-temperature surface inside the cabinet, measured by temperature contact plate 47. A positive moisture content difference indicates that, under the same dry air quality baseline, the air inside the cabinet carries more water vapor than the air to be introduced; a positive dew point safety margin indicates that the air to be introduced needs to continue cooling before reaching saturation on that surface. These two conditions respectively limit the direction of water vapor migration and the risk of condensation on the low-temperature surface, and cannot be replaced by a single relative humidity condition.

[0034] In one specific setup, the cabinet start threshold is set to 65%RH, the cabinet stop threshold to 55%RH, the cabinet external allowable threshold to 75%RH, the cabinet external recovery value to 72%RH, the preset humidity difference to 0.8 g / kg dry air, and the preset dew point safety margin to 2 degrees Celsius. A 10%RH hysteresis is established between the cabinet start and stop thresholds to prevent frequent start-stop cycles when humidity fluctuates near the same critical point. An additional 3%RH hysteresis is established between the cabinet external allowable threshold and the cabinet external recovery value to prevent repeated opening and closing of the check channel due to fluctuations in external data caused by rain showers or air conditioning cycles.

[0035] The cabinet's internal requirements are as follows: ; External humidity conditions for the cabinet are ; Moisture content directional conditions are ; Dew point safety conditions are ; The candidate logical value for air supply permit is ; In the formula, Candidate logical values ​​representing air supply permits; and These represent the smooth relative humidity inside and outside the cabinet, respectively. Indicates the start-up threshold inside the cabinet; Indicates the threshold for permission outside the cabinet; This indicates the difference in preset moisture content; Indicates the preset dew point safety margin; , , and These represent the logical values ​​of the four conditions; the logical symbol ∧ indicates that the conditions connected to it must be true simultaneously. An air supply permit is formally established only after all candidate logical values ​​are true for 30 consecutive seconds. After the permit is established, the internal demand conditions are latched. Operation continues until the internal humidity falls between the start and stop thresholds, at which point the air supply permit is reached. If any of the following conditions persists for 5 seconds: external relative humidity exceeding the limit, insufficient moisture content difference, or insufficient dew point margin, the air supply permit is revoked.

[0036] After the air supply permit is established, the fan alternately executes the air supply replacement phase and the shutdown balancing phase. The initial effective identification window uses a 2×2 two-dimensional heterogeneous excitation: a 60-second replacement phase for low-quality levels and a 180-second replacement phase for high-quality levels; a short balancing phase is 60 seconds, and a long balancing phase is 180 seconds. The four cross combinations are 60-second replacement with 60-second balancing, 180-second replacement with 60-second balancing, 60-second replacement with 180-second balancing, and 180-second replacement with 180-second balancing, corresponding to observation durations of 120 seconds, 240 seconds, 240 seconds, and 360 seconds, respectively. The initial drive of the fan is 70% of the rated drive. During the balancing period, the fan stops, the inlet check valve 24 closes, and the air that has entered the cabinet continues to diffuse through the partition gap. The humidity probe 43 is no longer directly affected by the continuous jet. The controller defines a set of adjacent air supply replacement phases and shutdown balancing phases as the j-th round. If the difference between the high and low effective air supply quality is less than 50% of their average, the drive amount of the high-quality wheel is increased but not exceeding 90%; the process ends immediately when the humidity inside the cabinet reaches the stop threshold in advance, and air supply is not continued to complete the identification. The model is not updated for this incomplete window.

[0037] The effective air supply quality in the j-th round is ; In the formula, This represents the cumulative dry air delivery mass in the j-th round after the preset micro-positive pressure has been established, in kilograms; This indicates the number of discrete sampling intervals included in the air supply replacement section of this cycle; This represents the effective pressure indication value of the kth sampling interval. It is 1 when the differential pressure inside and outside the cabinet is between 8 Pa and 25 Pa, and 0 otherwise. This indicates the density of dry air, expressed in kilograms per cubic meter. This represents the volumetric flow rate of the kth sampling interval obtained based on the fan drive quantity, speed, and calibration pressure versus flow rate curve, in cubic meters per second. This indicates the sampling time interval in seconds. Only the effective pressure range is accumulated, avoiding the misrepresentation of nominal airflow when the fan is powered on but the check valve is not open, the filter is clogged, or there is leakage at the door, as the effective airflow entering the cabinet.

[0038] The initial drying potential at the distal end in the j-th round is: ; The measured decrease in moisture content at the distal end of the j-th airflow is: ; The model value for the decrease in distal moisture content in the j-th round is ; In the formula, This represents the initial drying potential at the distal end of the j-th round, expressed in kilograms of dry air. This indicates the moisture content at the distal end of the displacement section at the beginning of the displacement process. This represents the average moisture content outside the cabinet, obtained by weighting the effective mass flow rate by pressure. This indicates the amount of decrease in moisture content measured at the remote location; and These represent the start time of the current replacement segment and the end time of the adjacent equilibrium segment, respectively. This represents the amount of moisture content reduction at the distal end calculated by the model. The remote participation coefficient is used to characterize the extent to which incoming air crosses the partition space and participates in remote displacement. Indicates the continuous moisture release rate, expressed in kilograms per second; This indicates the observation duration from the start of the permutation segment to the end of the adjacent equilibrium segment, in seconds; The equivalent dry air mass of the remotely separated space is expressed in kilograms; exp represents an exponential function with the natural constant as the base. The exponential supply air term gradually saturates with the increase of effective supply air mass, while the moisture release term increases linearly with the observation duration. By using different effective supply air masses, the two types of effects can be separated in the same set of data.

[0039] The equivalent dry air mass at the far end is initially calculated by multiplying the net air volume of the space separated by the airflow by the dry air density at the corresponding temperature and pressure. The net air volume is determined by the internal dimensions of the cabinet, minus the volume occupied by circuit breakers, busbar supports, cables, and other physical components. When accurate separation is not possible, the attenuation results of a low-concentration safety tracer gas under maintenance conditions can be used for verification. When multiple far-end nodes correspond to different control volumes, each node uses its own equivalent dry air mass. This mass parameter does not drift freely simultaneously with the far-end participation coefficient to prevent volume errors from being absorbed by the participation coefficient and reducing parameter identifiability.

[0040] The identification weight in the j-th round is ; In the formula, This represents the ratio of the number of effective pressure sampling intervals in the j-th round to the total number of air supply sampling intervals; The standard deviation of the moisture content outside the j-th cabinet is expressed in kilograms of dry air. The reference standard deviation for the stability of the moisture content outside the cabinet can be taken as 0.00020 kg / kg dry air; This represents the identification weight for the j-th round. The higher the effective pressure ratio and the more stable the moisture content outside the cabinet, the closer the weight is to 1; rounds with an effective pressure ratio below 0.80 are directly eliminated and retained without being assigned a minimal weight.

[0041] The weighted residual objective function for joint identification is: ; The joint estimate of the remote participation factor and the continuous moisture release rate is: ; In the formula, This represents the weighted residual objective function; The number of valid control rounds in the identification window is not less than 4; The weight of the j-th round is greater than 0 and not greater than 1, and is determined by the effective pressure ratio and the stability of the moisture content outside the cabinet in that round. and These represent the remote participation coefficient and the continuous moisture release rate obtained from the joint estimation, respectively. and These represent the lower and upper limits of the participation coefficient, respectively, and can be 0.15 and 1.20. This indicates the upper limit of the continuous moisture release rate allowed for automatic control, which can be set to 5.0 × 10⁻⁶. -7 Kilograms per second. Rounds with an effective pressure ratio below 0.80, changes in external humidity exceeding 0.20 g / kg dry air within a single round, or changes in remote temperature exceeding 1 degree Celsius will not be entered into the identification window; if any cross combination is missing, the two parameters will not be updated in this window.

[0042] The intermediate residual of the j-th round is ; The observation duration coefficient for the j-th round is ; Given a candidate value for the remote participation coefficient, the candidate value for the continuous moisture release rate is... ; In the formula, This represents the intermediate residual after deducting the air supply contribution corresponding to the candidate participation coefficient; The observation duration coefficient is expressed in seconds per kilogram; `clip` restricts the results to a given upper and lower limit. The controller can scan the remote participation coefficient at intervals of 0.01 between 0.15 and 1.20. For each candidate value, the continuous moisture release rate candidate value is directly calculated using the above formula, and then the set of parameters with the smallest weighted residual is selected. Subsequently, the calculation is repeated at intervals of 0.001 on both sides of the optimal grid point. This implementation only requires one-dimensional scanning and weighted summation, making it suitable for low-power microcontrollers. The change in the participation coefficient between two consecutive identification windows does not exceed 0.03, and the change in the continuous moisture release rate does not exceed 0.2 × 10⁻⁶. -7 After reaching kilograms per second, inverse control is initiated.

[0043] The target moisture content reduction for the next control wheel is ; The net objective function corresponding to the candidate effective air supply quality is: ; The monotonicity condition for the interval of inverse quality calculation is: ; In the formula, This indicates the target decrease in moisture content for the next control cycle; This indicates the maximum allowable target descent in a single round, which can be taken as 0.25 g / kg of dry air; This indicates the moisture content at the distal end at the inverse time. This indicates the moisture content corresponding to the stop threshold inside the cabinet at the current remote temperature; Represent the net objective function; This indicates the expected initial drying potential at the far end for the next control wheel; Indicates the candidate's effective air supply quality; This represents the average dry air mass flow rate of the most recent effective replacement section; This indicates the downtime for equalization. The monotonic condition ensures that the marginal increment of the supply air contribution is still greater than the marginal increment of continuous moisture release. The controller selects an interval satisfying the monotonic condition from the allowable range of 0.0236 kg to 0.0708 kg; if the lower limit of the interval has already reached the target, the lower limit is used; if the function has opposite signs at both ends of the interval, the minimum root is found using the bisection method until the mass interval width is less than 0.0002 kg.

[0044] If the net moisture content decreases at the upper limit of the interval is positive but still below the target, the upper limit of the interval is used to complete the limited replacement and output information indicating insufficient dehumidification capacity. If the maximum allowable decrease in net moisture content within the interval is not greater than 0, increasing the fan operating time will only be offset by continuous moisture release. The controller will then revoke the air supply permission, close the air intake channel, and record insufficient external drying potential or abnormal continuous moisture source. This determination constrains the fan operation to the achievable net moisture removal amount, and does not handle backflow that cannot be eliminated by ventilation by unboundedly extending the replacement section.

[0045] During fan operation, the differential pressure sensor participates in the inner loop control. The controller sets the target micro-positive pressure to 15 Pa, slightly adjusts the pulse width modulation duty cycle according to the deviation between the actual differential pressure and the target value, and limits the duty cycle to 30% to 90%. If the differential pressure is still below 5 Pa 10 seconds after the fan starts, it records information indicating leakage near the door gap or insufficient fan flow. If the differential pressure exceeds 30 Pa, the drive volume is immediately reduced; if it exceeds 40 Pa for 2 seconds, the fan is stopped and exhaust path blockage is recorded. The differential pressure is not used to determine air humidity, but to verify whether the permitted air supply has established the expected air path. The humidity outer loop determines the permitted air supply and replacement duration, while the differential pressure inner loop checks the air path status and adjusts the fan drive volume.

[0046] After the humidity inside the cabinet drops to 55%RH and remains there for 60 seconds, the controller ends the current dehumidification cycle and stops the fan. The inlet check valve 24 retracts as the fan outlet pressure difference disappears, and the exhaust check valve 32 closes after the pressure difference between the inside and outside of the cabinet drops back to the retraction pressure difference. After stopping, monitoring continues in low-power mode, and the system will not restart due to a slight increase in humidity if the humidity inside the cabinet does not return to 65%RH. If the relative humidity outside the cabinet is higher than 75%RH for 5 consecutive seconds, the air supply permission is revoked even if the humidity inside the cabinet has not yet reached 55%RH. A new cycle begins only after the relative humidity outside the cabinet drops below 72%RH, the humidity difference and dew point margin are satisfied again, and this is continuously confirmed for 60 seconds. If the humidity inside the cabinet still reaches 65%RH, the cycle begins. This confirmation process prevents unstable air supply triggered by a short-term drop in humidity during high-humidity weather. If the humidity inside the cabinet is between 55%RH and 65%RH and the previous cycle was paused due to the invalidation of the external permission, the system remains locked, waiting for the humidity inside the cabinet to reach the start-up threshold again, without using the permission status of the previous cycle.

[0047] Figure 5 A calculation example based on the conservation of air and water vapor mass is given. The curves in the figure represent calculation results obtained from the model and are not considered as actual field measurement data. The model treats the air inside the cabinet as a fully mixed control volume.

[0048] The change in water vapor mass inside the cabinet meets the requirements ; In the formula, This indicates the mass of dry air inside the cabinet, expressed in kilograms. and These represent the moisture content inside and outside the cabinet, respectively. Indicates time, in seconds; This represents the effective coefficient of the intake air for mixing within the cabinet, with a value ranging from 0 to 1; This indicates the dry air mass flow rate during fan operation, expressed in kilograms per second. This indicates the fan status; the value is 1 for the air supply replacement section and 0 for the shutdown balancing section. This represents the mass flow rate of water vapor generated by moisture release and uncontrolled infiltration from the materials inside the cabinet, expressed in kilograms per second. The left side of the equation represents the rate of change of water vapor storage inside the cabinet, the first term on the right side represents the water vapor carried away or introduced by controlled ventilation, and the second term represents the continuous water vapor source inside the cabinet.

[0049] The calculations used an effective air volume of 0.62 cubic meters inside the cabinet, a dry air density of 1.18 kg / m³, a volumetric flow rate calibrated at a 30 Pa operating point of 1.2 cubic meters / hour, a far-end participation factor of 0.62, and a continuous moisture release rate of 1.0 × 10⁻⁶. -7The humidity is measured in kilograms per second; the outside air temperature is 24 degrees Celsius and 52% RH, while the initial air temperature inside the cabinet is 26 degrees Celsius and 72% RH. Calculated using the aforementioned formula, the initial moisture content outside the cabinet is approximately 9.65 g / kg dry air, and the initial moisture content inside the cabinet is approximately 15.19 g / kg dry air, resulting in a moisture content difference of approximately 5.54 g / kg dry air. The dew point outside the cabinet is approximately 13.54 degrees Celsius, and with the low-temperature surface temperature inside the cabinet taken as 22 degrees Celsius, the dew point safety margin is approximately 8.46 degrees Celsius. Figure 5 (a) Using the same value as the overall mixing efficiency coefficient and water vapor source parameter, and alternating between 120 seconds of air supply and 60 seconds of equalization, the relative humidity inside the cabinet drops to 55%RH at approximately 100.6 minutes, at which point air supply stops. Subsequently, a continuous water vapor source causes the relative humidity to rise back to approximately 55.74%RH at 120 minutes, which does not yet meet the restart condition of 65%RH. The periodic slope change in the curve corresponds to the alternation between the air supply replacement phase and the shutdown equalization phase.

[0050] Figure 5 (b) Under a unified calculation benchmark of 0.7316 kg of equivalent dry air at the far end and 5.54 g / kg dry air initial drying potential, the supply air quality dimension and equilibrium duration dimension of two-dimensional heterogeneous excitation are shown. The low-quality replacement segment is 60 seconds, corresponding to an effective supply air quality of 0.0236 kg; the high-quality replacement segment is 180 seconds, corresponding to an effective supply air quality of 0.0708 kg. The two quality levels are cross-combined with the 60-second and 180-second shutdown equilibrium segments, respectively. Substituting the low-quality short equilibrium, high-quality short equilibrium, low-quality long equilibrium, and high-quality long equilibrium into the model in sequence, the decrease in moisture content at the far end is 0.0933 g / kg dry air, 0.2898 g / kg dry air, 0.0769 g / kg dry air, and 0.2734 g / kg dry air, respectively. At the same quality level, extending the equilibrium observation period increases the continuous moisture release. At the same observation duration of 240 seconds, the difference in moisture release between the high and low quality combinations is caused by the air transport term. The two types of changes have different directions, providing independent response information for the two parameters to be estimated. A uniform initial drying potential is used in the figure for intuitive comparison of excitation sensitivity. When the controller executes continuous control cycles, it still uses the measured initial drying potential round by round, without setting the initial states of the four cycles to be the same.

[0051] Will Figure 5 (b) Four model samples were used as calibration data without measurement disturbance, with a weight of 1 for each sample. The participation coefficient at the far end ranged from 0.15 to 1.20, and the continuous moisture release rate ranged from 0 to 5.0 × 10⁻⁶. -7 Joint identification was performed within the boundary of kg / s. First, a participation coefficient scanning interval of 0.01 was used, then refined to 0.001, yielding an estimated far-end participation coefficient of 0.620 and a sustained moisture release rate of 1.00 × 10⁻⁶. -7Kilograms per second, consistent with the sample generation parameters. If the next round uses a 60-second shutdown equalization period and the target moisture content decrease is taken as 0.250 g / kg dry air, substituting the estimated parameters and performing a binary inverse calculation within the range of 0.0236 kg to 0.0708 kg, the minimum effective supply air mass is approximately 0.0611 kg; based on an effective dry air mass flow rate of 3.933 × 10⁻⁶ kg / s. -4 Converted to kg / s, the replacement time is approximately 155.3 seconds. The allowable net decrease corresponding to the upper mass limit is approximately 0.2898 g / kg dry air, thus the target has a feasible solution. When the target is increased to 0.350 g / kg dry air, the upper limit of the interval can still produce a positive decrease but cannot reach the target. The controller adopts the upper limit and outputs a message indicating insufficient dehumidification capacity, without extending the air supply beyond the mass boundary. All the above values ​​are calculated by the listed model to verify the recalculation of the identification and reverse calculation process.

[0052] Figure 6 The curve illustrates the control logic's response to changes in external humidity. It is calculated using the aforementioned mass conservation model and threshold values. Air supply permission is established after all four candidate conditions are met consecutively for 30 seconds in the initial state. If the external relative humidity rises from 55%RH, and the moisture content difference or dew point safety margin remains invalid for 5 seconds, the air supply permission is revoked, and the fan drive is reduced to 0. If the external relative humidity exceeds 75%RH, the external threshold conditions also remain invalid. Subsequently, if the external relative humidity drops below 72%RH, the moisture content difference and dew point safety margin recover and are continuously confirmed for 60 seconds, and the internal humidity still meets the activation conditions, the air supply permission is re-established. Figure 6 (a) Provide the relative humidity inside and outside the cabinet and three threshold values ​​within the same relative humidity coordinate system. Figure 6 (b) Plot the air supply permit with a value of 0 or 1 and the normalized drive quantity of the fan with a value of 0 or 0.70 respectively; within the permit window, air supply is still intermittently supplied in 120-second replacement segment and 60-second equalization segment.

[0053] When there are multiple humidity sensing nodes inside the cabinet, the controller compares the response order of each remote node. If the nodes near the air inlet show a significant decrease, while the remote nodes in the busbar compartment show a decrease of less than 0.3%RH for three consecutive rounds, and the slight positive pressure inside the cabinet is normal, the controller records information indicating a near-end airflow short circuit or obstruction of the partition ventilation path. During maintenance, first check and seal any unexpected near-end leaks, then adjust the air outlet direction of the guide vane 27; cabinet types that allow independent ventilation in different zones can also supply air to the corresponding partitioned spaces separately. When all nodes decrease slowly and the differential pressure is low, prioritize checking the filter 23, the air inlet check valve 24, and the door seal. Spatial response, differential pressure, and speed data are used together to distinguish between airflow short circuits, path blockages, and insufficient overall flow.

[0054] The controller performs feasible fault diagnosis on sensor data. If the relative humidity exceeds 0%RH to 100%RH, the temperature exceeds -40°C to 85°C, three consecutive data verification failures occur, or the relative humidity changes by more than 20%RH within one minute without other nodes showing the same trend, the corresponding node is marked as invalid. Airflow must not be supplied when an external node is invalid, as the system cannot confirm the intended air intake. If a single internal node is invalid while other remote nodes remain valid, monitoring continues with the highest value from the remaining nodes, and a degraded status is reported. If all internal nodes are invalid, the fan stops, the check valve closes, and an alarm is triggered. Data frames may include node identifier, sampling sequence number, relative humidity, temperature, battery level, and cyclic redundancy check (CR) code. The controller only accepts data with incrementing sequence numbers and correct verification to prevent duplicate frames from being treated as continuous confirmations.

[0055] The power supply unit offers two options: external 24V DC power and lithium iron phosphate battery power. When powered externally, a higher duty cycle is permitted within permissible limits. When powered by the battery, the maximum cumulative airflow time per hour is given based on the remaining energy. For example, a 12.8V, 10Ah battery has a nominal energy of 128Wh. The controller and sensor's average power consumption during sleep mode is controlled at 0.03W, and the fan's operating power consumption is 0.8W. Based on an average fan duty cycle of 10%, the system's average power consumption is estimated to be the sum of 0.03W and 0.08W, approximately 0.11W. The 30-day energy requirement is approximately 79.2Wh. The difference between the nominal energy and this requirement is used to cover conversion losses, low-temperature capacity degradation, and communication reporting. This calculation is for battery selection and does not guarantee continuous dehumidification for 30 days under any given environment. If persistent heavy humidity inside the cabinet causes the fan demand to exceed the battery budget, the system will report external power demand or moisture source maintenance information, while retaining backflow prevention, monitoring, and fault reporting functions.

[0056] When the remaining battery energy is below 20%, the controller reduces the maximum single-cycle replacement time to 90 seconds and extends the equalization period. When it falls below 10%, regular air supply stops. Short-term emergency replacement is only performed when the relative humidity inside the cabinet exceeds 80%RH and the external drying potential is met. Humidity data is collected every 5 minutes, and fault reporting is maintained. After external power is restored, the external air condition is verified first, and then the control cycle is re-entered based on the highest humidity inside the cabinet. The fan cannot be started directly upon power-on. The communication component can use an RS-485 bus or a low-power wireless link to send the current threshold status, moisture content difference, dew point margin, estimated remote participation coefficient, estimated continuous moisture release rate, calculated effective air supply quality, actual effective air supply quality, differential pressure, and fault codes to the station. The station only adjusts the authorized threshold range; the safety upper limit and check-back fault interlock are retained in the local controller.

[0057] Before putting the system into operation, perform path verification. Keep the cabinet door closed and run at 50% drive for 30 seconds when the outside air supply meets the allowable airflow. Check if the exhaust paper indicator or differential pressure sensor shows outward airflow. Then stop the fan and confirm that the inlet check valve 24 and exhaust check valve 32 reset within 10 seconds, and the pressure difference between the inside and outside of the cabinet returns to near zero. Then, use a calibrated humidity source to perform two-point verifications on the external and internal nodes of the cabinet. If the node deviation exceeds 3%RH, the system should not be put into automatic mode.

[0058] The theoretical number of air exchanges can be estimated as follows: ; In the formula, This indicates the theoretical number of air exchanges in a single replacement segment; This indicates volumetric flow rate, expressed in cubic meters per second. Indicates the duration of the substitution segment, in seconds; This indicates the effective air volume inside the cabinet, in cubic meters. The theoretical air change rate is only used to set the quality level of the initial two-dimensional heterogeneous excitation. Baffles, components, and leaks cause the actual remote participation level to deviate from the assumption of full mixing. Subsequent control uses the remote participation coefficient obtained from joint identification and the continuous moisture release rate to back-calculate the air supply quality, and does not replace remote feedback with the theoretical air change rate.

[0059] After the system stops, a humidity recovery observation is performed. If the relative humidity inside the cabinet reaches 65%RH again within 2 hours, the controller associates and saves this recovery amount with the cumulative air volume of the previous cycle. If the humidity recovery accelerates for three consecutive cycles while the humidity outside the cabinet does not increase, the source of moisture may be continuous upward flow from the cable trench, sealing failure, moisture release from the moisture-absorbing material inside the cabinet, or jamming of the exhaust check valve. In this state, the controller keeps the air intake channel closed and reports maintenance information to avoid missing the continuous moisture source by only causing a short-term humidity drop after increasing the air volume. After maintenance is completed, the path confirmation and sensor verification are re-executed, the fault latch is cleared, and automatic operation is resumed. The humidity recovery record is used to supplement the distinction between changes in ambient humidity and continuous moisture release inside the cabinet.

[0060] The mechanical design of the intake micro-circulation component 2 also considers the thickness difference of the door panel. The clamping section of the connecting sleeve 21 can be set with an effective adjustment range of 12 mm to 28 mm. The outer mounting base is equipped with an anti-rotation plane, and the inner sealing pressure ring 22 adopts a double nut or a clamping structure with locking teeth. During installation, first use the original fasteners to confirm the function of the hole positions, and then replace them one by one. Do not remove multiple fasteners on the same edge of the door panel at the same time. After the connecting sleeve is inserted into the existing fastening hole, adjust the pressure ring according to the thickness of the door panel so that the compression of the sealing ring is maintained at 20% to 30% of its free thickness; insufficient compression will cause bypass leakage, and excessive compression will cause premature aging of the sealing ring. After installation, check for leakage with a low-pressure air source of no more than 50 Pa when the fan is stopped. Do not use high-pressure air to directly impact the components inside the cabinet. The external mounting base and the grounded door panel should maintain a reliable metal connection or be equipped with an equipotential bonding plate to avoid static electricity accumulation on the floating shell.

[0061] The flow area of ​​filter element 23 should be at least six times the total flow area of ​​connecting sleeve 21, ensuring that the initial pressure drop of the filter layer is less than the combined pressure drop of the check valve and the guide section. The fiber pre-filter layer can be made of flame-retardant material with a thickness of 3 mm to 5 mm and a nominal filter particle size of 10 micrometers. The average pore size of the outer hydrophobic membrane can be 0.2 micrometers to 1.0 micrometers. The hydrophobic membrane is used to prevent droplets and allow water vapor to pass through with the air; it does not act as a moisture-absorbing material. The dehumidification capacity comes from mass exchange when the humidity of the air outside the cabinet is lower than that of the air inside the cabinet. Dust accumulation on the filter layer will increase the pressure drop and reduce the airflow. Therefore, the controller records the differential pressure and remote response under the same fan drive. When the differential pressure increases monthly and the response decreases, a filter maintenance reminder is generated. Maintenance personnel can pull out the filter box from the outside of the cabinet door without opening the high-pressure compartment.

[0062] The micro brushless fan 25 is preferably a centrifugal micro-fan rather than a conventional axial flow fan that only claims a large flow rate under zero static pressure. The existing air path, consisting of a fastening hole, check valve, and hydrophobic membrane, has resistance of tens of Pascals. When selecting a fan, the complete pressure-flow curve should be verified to ensure a flow rate of at least 1.2 cubic meters per hour under 30 Pa static pressure. The fan rotor is equipped with a Hall effect commutation signal or speed feedback. If the speed is continuously below 60% of the target speed for 5 seconds after the controller issues a drive, a stall is detected, and forced restart by increasing the duty cycle is no longer allowed. A 1-2 mm thick silicone rubber vibration damping ring is installed between the fan and the mounting base to reduce vibration propagation along the door panel to the instruments and secondary terminals. A buffer chamber is retained between the fan outlet and the inlet check valve 24. The cross-sectional area of ​​the buffer chamber gradually shrinks to the inner diameter of the connecting sleeve to avoid additional eddy current losses caused by sudden narrowing.

[0063] The end position of the air guide 27 is determined by the available space inside the cabinet. For cabinets with a large cavity at the bottom of the cable compartment, the air guide forms a flat nozzle that is upward and deflected 15 to 30 degrees towards the rear of the cabinet, allowing the airflow to rise along the outer surface of the cable sheath. For secondary instrument compartments, the nozzle is directed towards the side wall without exposed contacts, allowing the airflow to diffuse along the wall. The minimum bending radius of the air guide duct is not less than three times its inner diameter, and the duct wall is made of insulating material that meets the corresponding flame retardant rating. The fixing clamps and grounding metal maintain sufficient creepage and electrical clearance. The air guide duct must not cross movable contacts, interlocking mechanisms, or pressure relief flaps, nor should the nozzle be directed directly at the dust deposition surface to prevent strong winds from stirring up the accumulated dust. The purpose of the intake jet is only to establish a slow displacement, not to perform localized high-speed purging of the insulating components.

[0064] The effective flow area of ​​the exhaust path at the far end should not be less than the total flow area of ​​the intake connection sleeve to avoid excessive pressure drop at the exhaust end. The intake check valve 24 is opened by the pressure difference between the fan outlet and the inside of the cabinet; after a slight positive pressure is established inside the cabinet, the exhaust check valve 32 with a lower pressure difference opens outwards. The opening pressure difference of the exhaust check valve 32 can be lower than that of the intake check valve 24 to reduce exhaust resistance during stable air supply, but this difference does not mean that the exhaust check valve opens before the intake check valve. The two check valves can use different elastic coefficients, and the valve plate material should be able to withstand repeated bending at temperatures from -20 degrees Celsius to 50 degrees Celsius without sticking. The opening of the exhaust assembly cover faces downwards to prevent top condensate from falling directly onto the membrane surface. When using the existing exhaust gap of the cabinet, the installer observes the main exhaust position through a smoke generator or low-pressure tracer airflow, and places the humidity sensing node outside the cabinet on the upwind side of the main exhaust position to prevent the exhaust humid air from being re-introduced into the cabinet.

[0065] The probe position of humidity sensing node 4 inside the cabinet needs to balance response speed and condensation protection. The sensitive surface of the temperature and humidity probe 43 maintains a 1-3 mm gap from the inner surface of the porous protective cover 44, allowing air exchange while preventing water droplets from directly bridging the electrodes. The axial sampling hole 42 has a slight upward slope from the inner end of the cabinet to the outer end, or an annular liquid collection groove at the inner end of the cabinet, causing occasional droplets to fall back towards the inner side of the cabinet and not enter the signal processing cavity 45 along the sampling hole. The probe operates with intermittent power supply, powered on for 200-500 milliseconds before sampling and entering sleep mode after reading to reduce the risk of lower relative humidity due to chip self-heating. If the probe detects near saturation and the surface temperature is close to the dew point, the controller does not activate the probe's internal heating function to artificially change the reading, but instead maintains the original environmental measurement and reports the risk of condensation.

[0066] The temperature contact plate 47 and the temperature and humidity probe 43 measure two different types of temperatures. The probe temperature represents the sampled air temperature, used for moisture content calculation; the contact plate temperature represents the door panel or selected metal surface temperature, used for dew point safety assessment. Combining the two into a single temperature would mask cold bridging: in winter, the door frame may be several degrees Celsius lower than the air inside the cabinet; the air temperature should still be used for moisture content calculation, while the lower surface temperature must be used for condensation assessment. The contact plate is constructed of a thin copper sheet and an insulating backing, with the temperature sensing element close to the metal contact end, and the outer side is isolated from the ambient air by a low thermal conductivity potting material. During initial installation, the contact plate reading should be compared with the attached reference thermometer; the dew point interlock should only be activated after the steady-state deviation is no greater than 0.5 degrees Celsius.

[0067] External humidity sensing nodes should not be installed directly below the air conditioner vent, inside the cold air jet in the door gap, or near the exhaust assembly 3. The preferred location is no more than 300 mm from the air intake filter 23 and a position that represents the actual intake air. The lower end of the protective cover should be open, and an insect screen should be installed on the side. If there are obvious layers in the switch room, and the fan intake is located at the bottom of the cabinet door, the external node should also be arranged at a similar height. Low temperature humidity data from near the indoor ceiling cannot be used as a substitute. When centrally controlling multiple switch cabinets, at least one local external node should be reserved for each cabinet, with the station-end environmental sensor used for rationality verification. If the difference in moisture content between the local node and the station-end node consistently exceeds 2 g / kg of dry air, it indicates the presence of local air conditioning jet, sensor drift, or exhaust backflow, and the data source needs to be investigated first.

[0068] Sensor factory calibration can be performed at two stable humidity points: 33%RH and 75%RH, with the temperature maintained at 25 degrees Celsius while waiting for the probe to stabilize. Two calibration coefficients—bias and slope—are saved for each node. During runtime, the raw digital values ​​are calibrated first before entering the median and recursive filtering. When a standard humidity generator is unavailable on-site, a traceable portable thermometer and hygrometer can be used for comparison within the same ventilation hood, but only repeatable biases are corrected; the slope is not refitted based on short-term environmental fluctuations. Every 24 hours, the controller selects a period when the fan is off and the temperature changes slowly inside and outside the cabinet to compare multiple nodes within the cabinet. If one node consistently deviates from other nodes with a different trend, it is marked as suspected drift. Drift markings do not automatically rewrite the calibration coefficients to preserve the measurement results of true local humidity.

[0069] The controller hardware can employ a low-power microcontroller with an independent watchdog timer. Humidity acquisition, fan drive, and communication are handled by isolated task scheduling. The fan power switch uses a high-side switch with overcurrent protection, and the overcurrent threshold is set to 1.5 to 2 times the rated operating current. After three consecutive overcurrent readings, the fan is locked, awaiting manual confirmation. The external 24V power input is equipped with reverse connection protection, surge suppression, and common-mode filtering. The communication port uses an isolated transceiver or a protection network that meets the field grounding requirements. Each time the program resets, it first configures the fan drive output to be off, then reads the sensor and check status to prevent the fan from momentarily activating due to floating microcontroller pins. Watchdog reset, undervoltage reset, and parameter verification failure are all written as independent events to non-volatile memory.

[0070] Threshold parameters are divided into field-adjustable parameters and inviolable safety boundaries. The cabinet start-up threshold can be set within the range of 60%RH to 75%RH, and the cabinet stop threshold must be at least 5%RH lower than the start-up threshold. The cabinet exit threshold can be set within the range of 65%RH to 80%RH based on local climate, but dryness potential and dew point safety conditions are always considered. A larger preset moisture content difference means more water vapor can be removed per unit airflow, suitable for battery mode; however, an excessively large value may result in prolonged periods without exit permission during the rainy season. The preset dew point safety margin can be between 2 and 5 degrees Celsius; a higher value should be used when there is heavy contamination or significant uncertainty in surface temperature measurement. After new parameters are written to the station, the controller first checks the above interrelationships and then stores them in a double-copied parameter area with a checksum; if any relationship is not satisfied, the parameter will be rejected.

[0071] The control cycle begins in standby mode. In standby mode, the fan is off, both check valves retract, and the controller samples at a low frequency. Once the humidity inside the cabinet reaches the activation threshold, the sampling frequency increases and the system enters a permission confirmation state. First, the external permission threshold is confirmed, then the moisture content difference and dew point safety margin are calculated, and finally, the sensor validity, battery energy, and exhaust path fault latch are checked. The system only enters the replacement state if all conditions are met consecutively. During the replacement state, the differential pressure is established first, and only after reaching 8 Pa and entering the preset micro-positive pressure range is the effective air supply quality accumulated; the time when the fan is running but the differential pressure has not entered this range is not counted as effective air supply. After replacement, the system enters the equilibrium state. In the equilibrium state, the final value of the remote observation is obtained, and the data for this round is written into the two-dimensional excitation window; when the window is complete and the data is valid, the remote participation coefficient and continuous moisture release rate are updated. After the parameters reach the stability condition, the effective air supply quality for the next round is calculated. If any safety condition fails, the system enters the lockout state. In the lockout state, the fan is first shut down, then the check valves are waited for to reset, and finally the trigger reason is recorded. The action sequence ensures that the air path returns to the safe position before the control judgment.

[0072] Short-term humidity exceedances outside the cabinet and prolonged high humidity are recorded using different methods, with air supply stopped during the exceedance period. Exceedances of less than 10 minutes are recorded as environmental pauses, with normal monitoring maintained. Exceedances exceeding 6 hours and continued increases in cabinet humidity are recorded as a lack of dehumidification conditions, requesting the station to activate the dehumidifier or air conditioner in the switchgear room. After the station equipment reduces the humidity in the switchgear room to the permissible range, this system still performs local verification and does not accept forced air supply commands that bypass local safety conditions. For large-space equipment, the air condition in the switchgear room is first changed, and then the cabinet door micro-circulation component delivers the satisfied air to the far-end area inside the cabinet; the two levels of equipment respectively handle room dehumidification and directional air exchange within the cabinet.

[0073] When the relative humidity inside the cabinet exceeds 80%RH and the outside air lacks the potential for dryness, the system activates a high humidity alarm and keeps the intake check valve 24 closed. When the difference between the low-temperature surface temperature inside the cabinet and the dew point of the air inside the cabinet is less than 2 degrees Celsius, the controller raises the alarm level and shortens the sampling cycle. If the switchgear has an existing heater and has obtained independent safety permission, the station can activate the heater to increase the surface temperature; heating only changes the temperature and relative humidity and does not directly reduce the amount of water vapor in the air. After the humidity of the outside air decreases and meets the permission conditions again, the micro-circulation component will then expel the water vapor from the cabinet, thereby preventing the relative humidity from rising again after heating stops and the temperature drops.

[0074] During the initial commissioning phase, a complete two-dimensional anomaly identification window is executed first. The difference between low and high effective air supply quality levels should not be less than 50% of their average value, and the difference between short and long shutdown balancing times should not be less than the short balancing time. At least one combination with the same observation time but different effective air supply quality should be retained. The controller checks whether the two columns of sensitivity data corresponding to the air supply quality and the observation time are approximately proportional. When the absolute value of the correlation coefficient of the two columns of data reaches 0.98 or the joint estimation result falls on the parameter boundary, this window is only used for safety control and is not written into the long-term parameter area. The spacing between the two quality levels or the two balancing time levels is increased in the next window. Only after the remote participation coefficient and continuous moisture release rate of two consecutive effective windows meet the stability limits will the air supply quality be reversed. Commissioning calibration does not automatically change the cabinet start threshold, the cabinet permissible threshold, and the dew point safety margin. Control wheels that handle station-end air conditioning switching, door opening, sudden changes in remote temperature, sensor disconnection, or insufficient pressure effective ratio will not enter the identification window.

[0075] When multiple parallel switchgear cabinets share the same dry air in the switch compartment, the station can stagger the allocation of air supply windows. Each controller reports the highest humidity, moisture content difference, dew point margin, and estimated replacement time within the cabinet to the station. The station prioritizes the operation of cabinets with high humidity and high drying potential to avoid rapid changes in the local air conditions in the switch compartment caused by the simultaneous operation of a large number of micro-fans. Staggered scheduling only determines the permitted time window; the local controller still checks the dual humidity gating and safety conditions sequentially. If the window arrives but local conditions are not met, the cabinet will not be activated. If a cabinet reaches a high risk of condensation during the waiting period, its scheduling priority can be increased, but the external moisture content and dew point limits must not be bypassed. This scheduling method retains the local feedback adjustment of each cabinet and reduces the disturbance of centralized operation to the local dry air conditions.

[0076] When the communication component uses RS-485, the address is determined by both the cabinet number and the node number, the polling period can be 10 seconds, and the frame tail uses a 16-bit cyclic redundancy check. When using a low-power wireless link, the four humidity sensing nodes inside the cabinet only send measurement frames and do not accept wireless commands to directly drive the fan; the fan drive is generated locally by the controller outside the cabinet door based on the verified data. Key parameters are written with version numbers and authentication flags, and after writing, they are read back for verification and the operator's identifier is recorded. When communication is interrupted, the fan does not remain in its last on state; if the controller does not obtain valid data from outside the cabinet for more than two normal polling cycles, the air supply permission is revoked. At this time, the air condition outside the cabinet cannot be confirmed, and closing the air intake channel can prevent air with unknown moisture content from being introduced into the cabinet.

[0077] Non-volatile event logs include at least the trigger time, internal and external temperature and humidity, internal low-temperature surface temperature, moisture content difference, dew point margin, two-dimensional excitation level, effective air supply quality, shutdown balancing time, measured moisture content decrease, estimated remote participation coefficient, estimated continuous moisture release rate, inverse quality, micro-positive pressure establishment time, and reason for shutdown. Records use a circular queue; when full, the oldest normal cycle record is overwritten. However, high humidity alarms, sensor malfunctions, stalled operation, and exhaust blockage records are protected and can only be overwritten after export or confirmation. Each parameter modification and calibration is also recorded, ensuring that subsequent humidity curves correspond to the thresholds and identification parameters used at the time. Remote display can plot curves, but control decisions always use the local original time series and calibration coefficients to avoid interpolation or sparse sampling at the display end altering the decision-making criteria.

[0078] On-site verification can be conducted on an empty cabinet of the same type without high voltage applied, or on a cabinet that has been taken out of service for maintenance. The test environment temperature is set to 25°C ± 2°C, and the relative humidity outside the cabinet is stabilized at 50%, 70%, and 80%RH, respectively. The relative humidity inside the cabinet is raised to approximately 70%RH using a controlled steam source. The 50%RH condition is used to verify normal dehumidification and remote response; the 70%RH condition is used to verify that air supply is still determined by moisture content even when the relative humidity is close to the permissible limit; and the 80%RH condition is used to verify external threshold lockout. For each condition, all measuring points, air volume, differential pressure, and fan status are recorded. Sensor calibration is checked before and after the test. Only after actual testing and data acquisition can the curve be called a test curve; this document Figure 5 and Figure 6 The model is explicitly designed to be recalculated, thus avoiding the inclusion of expected indicators as results from completed experiments.

[0079] Airflow verification can be performed by connecting a low-range laminar flow meter to the outside of the connecting sleeve, without using a large-section handheld anemometer for rough conversion at the small orifice. Differential pressure verification involves reading the fan at different duty cycles and establishing a table corresponding to the drive quantity, differential pressure, and flow rate of the cabinet. Remote ventilation effect can be assessed by releasing a small amount of safety tracer gas and detecting concentration changes at the exhaust end, or by using a calibrated humidity step for identification; when operating equipment is involved, gases that affect insulation or corrode materials must not be released. If the measured effective mixing coefficient is significantly lower than the design value, first check the flow direction, exhaust location, and measuring point arrangement, then update the permissible model parameters with measured data, while maintaining the mass conservation relationship. Model parameters are derived from geometric volume, measured airflow, and response identification, and their sources are recorded in the commissioning file.

[0080] Figure 5 The fully mixed model used provides an overall trend, but actual switchgear may exhibit multi-zone lag. When higher model resolution is needed, the busbar compartment, circuit breaker compartment, and cable compartment can be considered as separate control units, each with independent dry air mass and moisture source. Adjacent control units are coupled via exchanged mass flow rates. The intake control unit receives the moisture content outside the cabinet, and the exhaust control unit discharges it outside the cabinet. The openings in the partitions determine the zone exchange. Each control unit satisfies the condition that the incoming moisture mass minus the outgoing moisture mass equals the change in storage. Multi-zone parameters can be derived from... Figure 4 The phase difference of each node's response in the replacement segment and the equalization segment is identified; when the delay of the far node increases, the interval exchange weakens, and the controller extends the equalization segment accordingly.

[0081] The system's performance under low-temperature conditions also needs to consider battery and valve plate performance. When the ambient temperature is below -10 degrees Celsius, the available battery capacity is reduced according to the manufacturer's curve, and the controller correspondingly reduces the allowable cumulative air supply time. The check valve uses silicone rubber that remains elastic at low temperatures, and its opening and reseating are checked at -20 degrees Celsius before commissioning. Low temperatures do not mean there is no risk of condensation; warm, humid air entering the freezer is actually more dangerous. Therefore, the dew point safety margin cannot be eliminated due to a drop in temperature. If the surface temperature sensor inside the freezer is lower than the dew point outside the freezer, air supply should be prohibited even if the relative humidity outside the freezer is very low. The situation should be reassessed after the air condition in the switchgear room or the temperature of the freezer changes. When the temperature approaches 50 degrees Celsius, the continuous operation of the fan should be limited to avoid cumulative motor temperature rise.

[0082] Maintenance cycles are not solely based on calendar settings. Filter maintenance is based on differential pressure increase and effective response decay under the same drive quantity; check valve maintenance is based on differential pressure return to zero time and humidity anomalies after shutdown; sensor maintenance is based on multi-node deviation and calibration expiration time. After maintenance personnel replace filters, the controller clears the filter life count but retains historical differential pressure. After replacing humidity nodes, the new node identifier and calibration coefficient must be entered; the old node offset cannot be reused. If there are water stains on the exhaust membrane surface, the source of condensation or cleaning should be identified first, and then treated in a way that does not damage the membrane pores. After any maintenance is completed, a low-pressure path verification should be performed to confirm that the airflow direction is from the inlet to the exhaust end, preventing the check valve from being installed backwards and causing the passage to remain closed while the fan is running.

[0083] In one optional structure, the intake micro-circulation component 2 does not use a separate housing. Instead, it is integrated with the external humidity sensing node, controller, power supply, and communication components into a strip module arranged along the edge of the cabinet door. The strip module is clamped and fixed by two existing fastening holes, one of which serves as the air intake channel, and the other only serves as a mechanical anti-rotation device. Inside the module, the sensor is placed upstream of the filter, and the fan and check valve are placed downstream, ensuring that the actual intake air is measured before being delivered. Another optional structure uses an external fan connected to multiple cabinets via a short flexible hose. Each cabinet inlet has an independent electromagnetic check valve and a local humidity node; the controller only opens branches that meet local permits, and the pressure of the common fan is adjusted by the number of opened branches. Regardless of whether an independent or common fan type is used, the external permit of one cabinet cannot be directly copied to other cabinets with different temperature environments.

[0084] In another optional control method, the cabinet activation threshold can be slowly adjusted within permissible limits based on the degree of contamination and surface temperature. When the surface temperature is close to the cabinet dew point and condensation alarms have historically occurred at that location, the cabinet activation threshold can be lowered, but the minimum value is still constrained by sensor accuracy and the dry air supply capacity of the switch chamber. When the surface temperature is consistently much higher than the dew point and battery power is insufficient, the activation threshold can be moderately raised. The correction cycle is no less than 24 hours, with each change not exceeding 1% RH, and authorized by the station. When the humidity inside the cabinet is too high, the permissible threshold outside the cabinet is not relaxed to avoid introducing air with higher moisture content; in emergency situations, alarms, room dehumidification, or maintenance of the moisture source are used, while still maintaining the moisture content direction and dew point safety limits.

[0085] Each round of control records includes existing fastening port air path, dual humidity thresholds inside and outside the cabinet, moisture content difference, dew point safety margin, effective air supply quality under micro-positive pressure, shutdown balancing time, initial drying potential at the remote end, measured decrease at the remote end, and micro-positive pressure establishment time. Dual humidity thresholds determine the dehumidification requirements inside the cabinet and the range of the external environment. Moisture content difference and dew point safety margin further limit the direction of water vapor migration and the risk of condensation on low-temperature surfaces. A shutdown balancing phase is arranged after the air supply replacement phase to ensure that remote measurements avoid the influence of continuous jets. The response of the same quality level at different balancing times is used to constrain the continuous moisture release rate, and the response of different quality levels at the same observation time is used to constrain the remote participation coefficient. The micro-positive pressure establishment time supplements the identification of external air path anomalies, and the post-shutdown moisture return record verifies the continuous moisture source. All the above data are correlated with a unified timestamp within the same control round, and the controller updates two parameters accordingly, calculates the subsequent effective air supply quality, or generates maintenance prompts.

[0086] The above structure is not limited to the KYN28 cabinet type. For ring main units, low-voltage switchgear, or relay protection cabinets with different fastening hole diameters, the connecting sleeve 21 and sealing ring 22 can be replaced while retaining the filtration, check valve, airflow diversion, and dual humidity control relationships. When the outside air is supplied by a centralized dehumidification device, the outside humidity sensing node is still located at the actual intake of this component, and environmental data from the other end of the room is not used as a substitute. The outside permissible threshold, moisture content difference, and dew point safety margin can be tightened according to the insulation material, pollution level, and site climate, but the inside stop threshold must be lower than the inside start threshold, and the outside recovery value must be lower than the outside permissible threshold. Any threshold adjustment retains the moisture content direction and dew point safety verification. Different cabinet types and power supply methods use the same air supply permission and remote response judgment relationship.

Claims

1. A micro-circulation dehumidification system for switchgear with dual humidity thresholds, comprising a switchgear, an intake micro-circulation component, a remote exhaust path, an in-cabinet humidity sensing node, an external humidity sensing node, a differential pressure sensing node, and a controller, characterized in that: The intake micro-circulation component communicates with the interior of the cabinet via existing fastening holes in the cabinet door, and includes a fan, filter, intake check valve, and air guide towards the far end of the airflow. The humidity sensing node inside the cabinet is located at the far end of the airflow, and the humidity sensing node outside the cabinet is located on the fan intake side. When the humidity inside the cabinet reaches the cabinet activation threshold and the humidity outside the cabinet is not higher than the cabinet permissible threshold, the controller generates an air supply permit based on the difference in humidity inside and outside the cabinet and the temperature margin of the dew point of the outside air relative to the low-temperature surface inside the cabinet. When the air supply permit is established, the air supply replacement phase and the shutdown balancing phase are executed alternately, accumulating only the effective air supply quality within the preset micro-positive pressure range, and forming at least four cross-combinations of two-dimensional heterogeneous excitation control with at least two effective air supply quality levels and at least two shutdown balancing duration levels. The controller employs a set of control wheels, where at least two wheels have the same effective air supply quality but different shutdown balancing times, and at least two wheels have the same observation time but different effective air supply qualities. Based on the effective air supply quality of each control wheel, shutdown balancing time, the difference between the initial moisture content at the remote end and the moisture content outside the cabinet, the equivalent dry air quality at the remote end, and the measured decrease in moisture content at the remote end, the controller performs dual-parameter joint identification of the remote participation coefficient and continuous moisture release rate within preset parameter boundaries. It then calculates the subsequent effective air supply quality based on the identified values ​​and the target decrease in moisture content. If no feasible solution for forward dehumidification exists within the allowable air supply quality range, the humidity inside the cabinet reaches a threshold lower than the cabinet's start-up threshold, or any safety condition in the air supply permission fails, the controller stops the fan and closes the air intake channel.

2. The micro-circulation dehumidification system for switchgear according to claim 1, characterized in that: The intake microcirculation component includes a mounting base located on the outside of the cabinet door, a connecting sleeve passing through an existing fastening hole, and a sealing pressure ring located on the inside of the cabinet door. The filter, fan, intake check valve, rectifier section, and guide are connected sequentially along the intake direction. The exhaust path includes an exhaust connector fitted into another existing fastening hole, an exhaust check valve that opens only to the outside of the cabinet, and a hydrophobic and breathable membrane located on the outside of the exhaust check valve. After the fan stops, the intake check valve returns to its seat as the pressure difference at the fan outlet disappears, and the exhaust check valve returns to its seat as the pressure difference between the inside and outside of the cabinet decreases. Both remain closed after returning to their seats, and the air that has been introduced into the cabinet continues to diffuse through the partition gap inside the cabinet in the shutdown equalization section.

3. The micro-circulation dehumidification system for switchgear according to claim 1, characterized in that: The humidity sensing node inside the cabinet includes a load-bearing hollow bolt body that replaces the original fasteners of the cabinet door. The hollow bolt body has an axial sampling hole that connects to the inside of the cabinet. A porous protective cover is installed at the inner end of the axial sampling hole. The temperature and humidity probe is located between the porous protective cover and the axial sampling hole. The outer end of the hollow bolt body has a signal processing cavity that connects to the temperature and humidity probe. A sealing isolation component is installed between the signal processing cavity and the axial sampling hole. The hollow bolt body also has a temperature contact plate that is in thermal contact with the metal surface of the cabinet door to obtain the temperature of the low-temperature surface inside the cabinet.

4. The micro-circulation dehumidification system for switchgear according to claim 1, characterized in that: The controller obtains the moisture content inside the cabinet based on the temperature and relative humidity inside the cabinet, and obtains the moisture content outside the cabinet and the dew point of the air outside the cabinet based on the temperature and relative humidity outside the cabinet. When the difference between the moisture content inside the cabinet and the moisture content outside the cabinet is not less than the preset moisture content difference and the difference between the low temperature surface temperature inside the cabinet and the dew point of the air outside the cabinet is not less than the preset dew point safety margin, the dry potential condition in the air supply permit is determined to be met.

5. The micro-circulation dehumidification system for switchgear according to claim 1, characterized in that: For any two-dimensional heterogeneous excitation control wheel, the model value of the decrease in remote moisture content is equal to the difference between the initial remote moisture content and the moisture content outside the cabinet multiplied by a factor obtained by subtracting the exponential term, and then subtracting the product of the continuous moisture release rate and the observation duration divided by the equivalent dry air mass at the remote end. The exponent of the exponential term is the negative value of the product of the remote participation coefficient and the effective supply air mass divided by the equivalent dry air mass at the remote end. The controller determines the weight of each wheel based on the effective pressure ratio and the stability of the moisture content outside the cabinet. Within the parameter boundaries, the controller minimizes the sum of squared weighted residuals between the measured values ​​and the model values ​​of at least four cross-combined control wheels to obtain the remote participation coefficient and the continuous moisture release rate.

6. The micro-circulation dehumidification system for switchgear according to claim 1, characterized in that: It also includes power supply components and communication components. The power supply components have external DC power supply mode and battery power supply mode. In battery power supply mode, the controller limits the cumulative running time of the fan per unit time based on the remaining battery energy, and retains the functions of humidity acquisition, backflow prevention shutdown and abnormal reporting when the remaining energy is lower than the preset energy threshold. The controller revokes the air supply permission and generates a fault record when the humidity sensing node inside the cabinet or the humidity sensing node outside the cabinet goes out of bounds, loses connection or has an abnormal rate of change.

7. A micro-circulation dehumidification control method for switchgear with dual humidity thresholds, wherein the air intake micro-circulation component supplies air into the switchgear through existing fastening holes in the cabinet door, and the gas inside the cabinet is discharged through the exhaust path at the far end of the airflow, characterized in that... This includes: collecting relative humidity inside and outside the cabinet, temperature inside and outside the cabinet, low-temperature surface temperature inside the cabinet, and differential pressure inside and outside the cabinet; generating air supply permission based on humidity thresholds inside and outside the cabinet, moisture content difference inside and outside the cabinet, and temperature margin of outside air dew point relative to low-temperature surface inside the cabinet; alternately executing air supply replacement phase and shutdown balancing phase, accumulating only effective air supply quality within a preset micro-positive pressure range, forming at least 4 cross-combined two-dimensional heterogeneous excitation control wheels with at least 2 effective air supply quality levels and at least 2 shutdown balancing duration levels, wherein at least 2 wheels have the same effective air supply quality but different shutdown balancing durations, and at least 2 wheels have the same observation duration and effective air supply. Different quality; obtain the measured decrease in moisture content at the remote end of each control wheel, and based on the water vapor mass balance model with the effective air supply quality, observation time, the difference between the initial moisture content at the remote end and the moisture content outside the cabinet, and the equivalent dry air quality at the remote end as inputs, perform dual-parameter joint identification of the remote end participation coefficient and continuous moisture release rate within the preset parameter boundary; calculate the subsequent effective air supply quality based on the identified value and the target decrease in moisture content; stop air supply and close the air intake channel when there is no feasible solution for positive dehumidification within the allowable air supply quality range, the humidity inside the cabinet reaches the cabinet stop threshold below the cabinet start threshold, any safety condition in the air supply permission fails, or the collected value is abnormal.

8. The micro-circulation dehumidification control method for switchgear according to claim 7, characterized in that: The cabinet start threshold is higher than the cabinet stop threshold; the air supply permit is revoked when the relative humidity outside the cabinet is higher than the cabinet permit threshold and continues for a preset cancellation confirmation time. The air supply permit is only allowed to be regenerated after the relative humidity outside the cabinet drops to a recovery value lower than the cabinet permit threshold and continues for a preset recovery confirmation time. The moisture content is calculated based on relative humidity, temperature, saturated water vapor pressure and atmospheric pressure. Air supply is still prohibited when the difference between the moisture content inside the cabinet and the moisture content outside the cabinet is positive but less than the preset moisture content difference.

9. The micro-circulation dehumidification control method for switchgear according to claim 7, characterized in that: A net moisture content reduction function for candidate effective supply air quality is constructed. Within the allowable supply air quality range where this function remains monotonically increasing, a bisection method is used to find the minimum effective supply air quality that makes the net moisture content reduction reach the target moisture content reduction. When the net moisture content reduction corresponding to the upper limit of the range is positive but lower than the target moisture content reduction, the upper limit of the range is used and insufficient capacity information is output. When the maximum net moisture content reduction within the range is not greater than 0, the supply air permit is revoked and information of insufficient external drying potential or continuous moisture release abnormality is output.

10. The micro-circulation dehumidification control method for switchgear according to claim 7, characterized in that: When the micro-positive pressure build-up time exceeds the preset time and the far-end participation coefficient decreases, output gas path inspection information such as leakage, filter blockage, or check valve not opening; when the micro-positive pressure build-up time does not exceed the preset time and the far-end participation coefficient decreases, output information such as near-end airflow short circuit or abnormal resistance of the separation path. When the remote participation coefficient is within the allowable range but the continuous moisture release rate exceeds the upper limit of the moisture release rate, output moisture source inspection information for cable trench backflow, cabinet sealing failure, or moisture release from materials inside the cabinet.

Images