Energy-saving device closing and opening control method based on humidity prediction

Abstract

This invention discloses a method for controlling the closing and opening of an energy-saving device based on humidity prediction. The method includes: real-time monitoring of ambient humidity, recording the first maximum humidity value during the period from closing to opening each day, and the second maximum humidity value during the period from opening to closing the next day. During opening control, if the first maximum humidity value is greater than a first threshold, or if the predicted humidity value for the next day before closing, based on the historical second maximum humidity value, is greater than the second threshold, then opening is prohibited and a direct-through mode is entered; otherwise, normal opening occurs. During closing control, if the device entered a direct-through mode the previous day, the closing state is maintained; if the device opened normally the previous day, the real-time sampled humidity value is obtained. When the value exceeds the first threshold, closing is prohibited and an alarm is triggered; otherwise, normal closing occurs. This invention solves the safety hazards associated with closing and opening operations in high-humidity environments and addresses the complexity and difficulty in deployment of existing prediction methods.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power plant technology, and specifically to a method for controlling the opening and closing of an energy-saving device based on humidity prediction. Background Technology

[0002] In photovoltaic power plants, energy-saving devices collect information such as system voltage and current to determine the current system operating status, and then control the opening and closing of high-voltage switches. This enables the transformer and its downstream loads to be switched off at night and put into operation during the day, thereby achieving energy conservation. These energy-saving devices are switched on and off daily to achieve normal power generation during the day and energy saving at night.

[0003] However, operating high-voltage switches in high humidity or even condensation conditions constitutes a high-risk operation and a typical hazardous condition in the power industry. Specifically, when the ambient humidity exceeds 80% RH, it is close to the condensation threshold, and a water film easily forms on the equipment surface, leading to a significant decrease in insulation strength of approximately 50% to 90%. Operating under such conditions can directly cause serious safety accidents such as bushing breakdown and internal arcing short circuits due to overvoltage. Therefore, for photovoltaic power plant energy-saving devices, how to achieve daily switching control while ensuring safety is a pressing technical problem that needs to be solved.

[0004] In existing technologies, such as patent CN2026101630519, a method, device, equipment, medium, and product for handling faults in energy storage systems are disclosed. These methods employ complex models such as Long Short-Term Memory (LSTM) networks for predictive control, requiring extensive historical data for model training. Furthermore, different models need to be adapted for different sites, resulting in high implementation costs and long development cycles. This makes them unsuitable for humidity prediction scenarios in energy-saving devices with relatively simple structures and high requirements for real-time performance and convenience. Therefore, a simple and effective humidity prediction algorithm and a closing / opening control strategy are essential. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method for controlling the closing and opening of an energy-saving device based on humidity prediction, so as to solve the problems of safety hazards in the closing and opening operation in high humidity environment, and the complexity and difficulty in deployment of existing prediction methods.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows.

[0007] A method for controlling the opening and closing of an energy-saving device based on humidity prediction includes the following steps: Humidity monitoring and recording: Real-time monitoring of ambient humidity, and recording the first maximum humidity value during the period from closing the circuit to opening the circuit each day, and the second maximum humidity value during the period from opening the circuit to the next closing operation each day; The circuit breaker tripping control process involves performing the following checks before the scheduled tripping operation each day: Determine whether the maximum humidity value of the day is greater than the first humidity threshold; Based on the second-highest historical humidity value, predict the humidity value before closing the gate the next day. Determine whether the predicted humidity value is greater than the second humidity threshold; If the first maximum humidity value is greater than the first humidity threshold, or the predicted humidity value is greater than the second humidity threshold, then the daily circuit breaker tripping operation is prohibited, and the energy-saving device is controlled to enter the direct-flow mode; otherwise, the circuit breaker tripping operation is performed. Closing control process: When the closing operation is planned to be executed: If the energy-saving device entered the direct-flow mode due to humidity conditions the previous day, it will remain in the closed state and the direct-flow mode during the closing operation on the current day. If a normal tripping operation was performed the previous day, the real-time sampled humidity value at the current moment is obtained, and it is determined whether the real-time sampled humidity value is greater than the first humidity threshold. If it is, the closing operation is prohibited and a fault alarm is issued; otherwise, the closing operation is performed.

[0008] Preferably, the step of predicting the humidity forecast value before closing the circuit breaker the next day based on the second highest historical humidity data specifically involves: S1. Predict the base humidity for the current day based on the historical morning humidity of the previous 7 consecutive days. S2. The baseline humidity for the day is adjusted based on the temperature sampled that day for humidity compensation. S3. The prediction weighting coefficient is automatically adjusted daily based on actual humidity sampling.

[0009] Preferably, the tripping control process further includes: if the first humidity maximum value is greater than the third humidity threshold and less than or equal to the first humidity threshold, then an alarm signal is output.

[0010] Preferably, the condition for exiting the direct-through mode is: using hysteresis control, the maximum value of the first humidity recorded for M consecutive days is less than or equal to the third humidity threshold.

[0011] Preferably, the first humidity threshold is 85%RH, the second humidity threshold is 80%RH, and the third humidity threshold is 75%RH.

[0012] Preferably, M=3.

[0013] Preferably, the direct-through mode is as follows: the high-voltage switch in the energy-saving device remains closed and does not perform time-based automatic tripping operation.

[0014] Preferably, the real-time monitoring of ambient humidity includes: periodically acquiring raw humidity data at a preset first sampling frequency; recording the first maximum humidity value and recording the second maximum humidity value specifically involves: based on the raw data acquired at the first sampling frequency, further calculating the first representative humidity value at the second sampling frequency, and respectively selecting the maximum value among all the first representative humidity values ​​in the two time periods of "from closing the circuit to before the opening operation" and "from opening the circuit to before the next closing operation" each day, as the corresponding first maximum humidity value and second maximum humidity value; The real-time sampled humidity value is calculated by taking multiple raw humidity data obtained at the first sampling frequency within a preset time window before the closing operation, and obtaining a second representative humidity value within the time window, which is used as the real-time sampled humidity value.

[0015] Preferably, the first humidity representative value and the second humidity representative value are the average values ​​of multiple raw humidity data within the corresponding time period.

[0016] Due to the adoption of the above technical solutions, the technical progress achieved by this invention is as follows.

[0017] This invention achieves an effective balance between safety and energy efficiency: by incorporating humidity prediction for the next day's closing time into the circuit breaker decision-making process, and employing a differentiated threshold strategy (comparing actual humidity to 85% and predicted humidity to 80%), the nighttime circuit breaker energy-saving function of the energy-saving device is maintained to the maximum extent while ensuring safety. When high humidity is predicted for the next day, circuit breaker operation is prohibited in advance to prevent the equipment from being in an unsafe state for an extended period due to the inability to safely close the circuit breaker the following day.

[0018] The predictive model of this invention is simple and practical: it uses a weighted prediction method to predict the ambient humidity before the next day's closing. The model is simple, requires little input data (only the historical maximum humidity values ​​of the past 7 days), requires no complex training, and its parameters are adjustable, making it easy to deploy and debug in the field. Compared with predictive control methods that use complex models such as LSTM, it has stronger practical value and wider applicability.

[0019] This invention features a comprehensive hierarchical protection system: by setting a first humidity threshold (85%RH), a second humidity threshold (80%RH), and a third humidity threshold (75%RH), a three-level humidity protection system is formed. The first humidity threshold is used for the highest risk judgment (prohibiting circuit breaker opening / closing), the second humidity threshold is used for predictive prohibition of circuit breaker opening based on prediction, and the third humidity threshold is used for alarm and recovery judgment after exiting the direct-connection mode, achieving comprehensive protection from early warning and prediction to prohibition.

[0020] This invention offers high control stability: it employs a hysteresis control method to exit the direct-through mode, only recovering after the humidity has been below the threshold for several consecutive days. This effectively prevents frequent switching between direct-through mode and normal tripping mode near the humidity threshold, thus improving the stability of system operation.

[0021] This invention features high data acquisition and processing reliability: through a hierarchical data processing and decision-making mechanism, it significantly improves data reliability and decision robustness. In the monitoring and recording stage, the raw data sampled every second is calculated into a minute-average "first representative humidity value," from which the daily maximum value is then selected, effectively filtering out transient interference and providing a stable foundation for subsequent prediction and trend judgment. In the closing decision stage, multiple raw data points within a longer window (e.g., 1 hour) before closing are used to calculate a "second representative humidity value" as the basis for judgment. This more comprehensively reflects the current environmental state, effectively avoiding misjudgments caused by short-term data fluctuations, thereby greatly improving the reliability of safe opening and closing decisions. Attached Figure Description

[0022] Figure 1 This is a flowchart of the tripping control process of the present invention; Figure 2 This is a flowchart of the closing control process of the present invention; Figure 3 This is a schematic diagram of the closing and opening control module of the present invention. Detailed Implementation

[0023] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0024] A method for controlling the opening and closing of energy-saving devices based on humidity prediction, which can be applied to energy-saving devices in photovoltaic power plants.

[0025] (1) System Architecture like Figure 3 As shown, the energy-saving device mainly includes: a high-voltage switch, a humidity sensor, and a data processing and control module.

[0026] The humidity sensor monitors ambient humidity in real time and transmits the humidity data to the data processing and control module. The data processing and control module receives the humidity data, performs humidity recording, humidity prediction, threshold comparison, and closing / opening decisions, and sends control commands to the high-voltage switch. The high-voltage switch then performs closing or opening operations according to the control commands.

[0027] (2) Humidity data acquisition and processing The data processing and control module periodically acquires raw humidity data at a preset first sampling frequency. In this embodiment, the first sampling frequency is once per second, that is, one raw humidity data point is acquired per second.

[0028] Based on this, the module further calculates the first representative humidity value at the second sampling frequency. In this embodiment, the second sampling frequency is once per minute, that is, the average value of the 60 raw humidity data points transmitted back each minute is taken as the first representative humidity value within one minute. This minutely average value is used to filter out transient fluctuation interference from the humidity sensor and improve data reliability.

[0029] The module filters the maximum value among all representative humidity values ​​within two specific time periods each day, using these as the corresponding first and second maximum humidity values. Specifically, this includes: The maximum first humidity value (Hamax) is the highest of all representative first humidity values ​​recorded during the period from closing the circuit breaker to performing the tripping operation each day. This value reflects the humidity conditions during the daytime operation and is used to determine the safety of the tripping operation for that day.

[0030] The second maximum humidity value, Hbmax, is the maximum value among all representative humidity values ​​recorded during the period from the time the circuit breaker is opened each day until the next closing operation. This value reflects the humidity conditions from night to the following morning and serves as the basis for humidity prediction before closing the circuit breaker the next day.

[0031] Meanwhile, the module records continuous N The second-highest humidity level in history for today. h i ,in i =1,2,..., N ,and i =1 indicates that the data is from the day closest to the prediction date and is used as input for humidity prediction.

[0032] The aforementioned first humidity representative value is used for daily humidity recording and maximum value filtering. However, when making the closing decision, a longer time window is used to calculate the second humidity representative value to improve the stability of the decision. For details, please refer to Part (5).

[0033] (3) Humidity prediction method This invention employs a weighted prediction method to calculate the predicted humidity value before the next day's power-on. The main steps include: S1. Predicting the base humidity for the current day based on the historical morning humidity of the preceding 7 consecutive days; S2. Correcting the base humidity for the current day using temperature sampling as humidity compensation; and S3. Automatically adjusting the prediction weighting coefficients daily based on actual humidity sampling. The specific method is as follows: S1. Predict the base humidity for the current day based on the historical morning humidity of the previous 7 consecutive days. H pred_base .

[0034] Through formula Calculate humidity forecast values H pred_base ,in wi For corresponding h i The weight coefficients, and satisfying w 1 > w 2 >...> w N .

[0035] Specifically, obtain the second-highest historical humidity value over seven consecutive days. h i ( i =1,2,…,7), where i=1 represents the data from the day closest to the prediction date. Humidity data closer to the current date is assigned a larger weight. w i The weighting coefficients satisfy w 1 > w 2 >...> w 7 .

[0036] The formula for outputting the predicted data is: H pred_base = h 1 × w 1 + h 2 × w 2 + h 3 × w 3 + h 4 × w 4 + h 5 × w 5 + h 6 × w 6 + h 7 × w 7 .

[0037] In this embodiment, the weighting coefficients are taken from the following empirical values: w 1 =0.3 (the closest day to the current day). w 2 =0.20, w 3 =0.15,w 4 =0.12, w 5 =0.10, w 6 =0.08, w 7 =0.05 (the farthest day from the current day); and w 1 + w 2 + w 3 + w 4 + w 5 + w 6 + w 7 =1.

[0038] It should be noted that the weighting coefficients mentioned above are empirical values ​​and can be configured and adjusted according to the environmental characteristics of different sites. For example, in areas with drastic humidity changes, the weighting coefficient for the most recent day can be appropriately increased; in areas with relatively stable humidity changes, the weighting coefficients for each day can be appropriately balanced to make the forecast data more accurate.

[0039] S2. The baseline humidity for the day is adjusted based on the temperature sampled that day.

[0040] Following the principle that higher temperatures result in lower humidity, specifically, humidity decreases when the temperature is above 25℃ and increases when the temperature is below 25℃. The temperature compensation coefficient is selected as 0.015 (a commonly used industrial compensation coefficient). K t Calculate according to the following formula: K t =1-0.015× delta_T ; delta_T = T_now -25.

[0041] in: delta_T This is the temperature compensation value. T_now This is the current temperature value.

[0042] Final predicted humidity H pred for: H pred = H pred_base × K t .

[0043] S3. The prediction weighting coefficient is automatically adjusted daily based on actual humidity sampling.

[0044] The prediction error is calculated based on the actual temperature of the day and the predicted humidity obtained in step S2: error = H real - H pred Adjust the weighting coefficients based on the prediction error: w 1_new = w 1 +α× error ×0.35 w 2_new = w 2 +α× error ×0.25 w 3_new = w 3 +α× error ×0.15 w 4_new = w 4 +α× error ×0.10 w 5_new = w 5 +α× error ×0.08 w 6_new = w 6 +α× error ×0.05 w 7_new = w 7 +α× error ×0.02 Wherein, α: weight learning rate; takes a value of 0~1, the smaller the value, the more stable it is, and in this invention, the value is 0.05.

[0045] Then, normalization is performed to ensure that the sum of the new weight coefficients is 1.

[0046] sum_w = w 1_new +...+w 7_new w i = w i_new / sum_w in i =1,2,...,7.

[0047] (4) Circuit breaker tripping control process The circuit breaker tripping control process is executed before the daily scheduled tripping operations. For example... Figure 1 As shown, the specific steps are as follows: Step 1: Determine whether the maximum humidity value of the day is greater than the first humidity threshold.

[0048] The data processing and control module first determines whether the maximum humidity value Hamax recorded for the day is greater than the first humidity threshold hlimita. In this embodiment, hlimita is set to 85%RH.

[0049] If Hamax > hlimita, it indicates that the ambient humidity was too high during the day's operation. Performing a tripping operation under these conditions poses safety risks such as insulation breakdown and arcing short circuits. Therefore, tripping operations are prohibited for the day, and the energy-saving device should be switched to direct-flow mode.

[0050] Step 2: Predict the humidity value before closing the gate the next day.

[0051] If Hamax ≤ hlimita, then based on the second highest historical humidity value, the predicted humidity value before the next day's shutdown is calculated using the aforementioned weighted prediction method. H pred .

[0052] Step 3: Determine whether the predicted humidity value is greater than the second humidity threshold.

[0053] The calculated humidity prediction value H pred It is compared with the second humidity threshold hlimitb. In this embodiment, hlimitb is set to 80%RH.

[0054] like H pred If >hlimitb indicates that the predicted ambient humidity is too high before the circuit breaker closes the next morning, and if a tripping operation is performed on that day, the circuit breaker may not be able to close safely the next day, resulting in the equipment being in an unsafe state for an extended period. Therefore, tripping operations on that day are also prohibited, and the energy-saving device is controlled to enter direct-flow mode.

[0055] Step 4: Normal tripping.

[0056] Only if Hamax ≤ hlimita and H predOnly when ≤ hlimitb, the normal opening operation is performed to cut off the subsequent load and achieve energy saving.

[0057] Step 5: Hierarchical alarm mechanism.

[0058] In addition, the opening control process also includes a hierarchical alarm mechanism. If the maximum humidity value Hamax on the current day is greater than the third humidity threshold hlimitc and less than or equal to the first humidity threshold hlimita, that is, hlimitc < Hamax ≤ hlimita, the data processing and control module outputs an alarm signal to remind the operation and maintenance personnel to handle it actively. In this embodiment, the value of hlimitc is 75%RH.

[0059] After receiving the alarm signal, the operation and maintenance personnel can choose to directly control the device to enter the direct-through mode according to the actual situation, or wait for the humidity to further increase to reach the first humidity threshold and then the system automatically enters the direct-through mode.

[0060] (5) Closing control process.

[0061] The closing control process is executed when the planned closing operation is to be performed. As Figure 2 shown, the specific steps are as follows: Step 1: Determine whether the direct-through mode was entered the previous day.

[0062] The data processing and control module first determines whether the energy-saving device entered the direct-through mode due to humidity conditions the previous day.

[0063] Step 2: Processing when the direct-through mode was entered the previous day.

[0064] If the direct-through mode was entered the previous day, the direct-through mode and the closed state are maintained at the closing operation time of this day without switching. The energy-saving device continues to maintain the closed state until the conditions for exiting the direct-through mode are met. This can avoid the risk of forcibly performing the closing operation when the humidity conditions have not improved.

[0065] Step 3: Processing of normal opening the previous day.

[0066] If the normal opening operation was performed the previous day, the real-time sampled humidity value at the current moment is further obtained. In this embodiment, the real-time sampled humidity value is calculated from multiple humidity raw data obtained at the first sampling frequency (once per second) within a preset time window before the closing operation is executed, and the second humidity representative value within this time window is used as the real-time sampled humidity value. Specifically, the preset time window is 1 hour, that is, the average value of 3600 humidity raw data collected within 1 hour before closing is taken as the real-time sampled humidity value h average .

[0067] Step 4: Determine whether the real-time sampled humidity value is greater than the first humidity threshold.

[0068] The acquired real-time sampling humidity value h average Compare with the first humidity threshold hlimita (85%RH).

[0069] If h average If the value >hlimita indicates that the current ambient humidity is too high, there is a safety risk in performing the closing operation. In this case, closing the circuit is prohibited, a fault alarm is issued, and the system awaits intervention from inspection personnel.

[0070] If h average If ≤hlimita, it means that the current ambient humidity is within a safe range. Perform normal closing operation, connect the downstream load, and start power generation for the day.

[0071] (6) Exit mechanism of the direct access mode The direct-connect mode is a safety protection mode that the energy-saving device enters when the humidity conditions do not meet the tripping requirements. In this mode, the high-voltage switch in the energy-saving device remains closed and does not perform time-based automatic tripping operations, sacrificing energy-saving functions in exchange for equipment safety.

[0072] To avoid frequent switching between direct-flow mode and normal tripping mode near the humidity threshold, this invention employs hysteresis control to exit direct-flow mode. Specifically, the condition for exiting direct-flow mode is that the maximum first humidity value recorded for M consecutive days is less than or equal to the third humidity threshold hlimitc (75%RH). In this embodiment, M is set to 3.

[0073] That is, the system exits the direct-flow mode and resumes normal trip control only when the maximum humidity value recorded for three consecutive days is less than or equal to 75%RH. This hysteresis control method effectively prevents the system from oscillating near the humidity threshold and improves control stability.

[0074] (7) Threshold system description This invention sets three humidity thresholds to form a complete protection system:

[0075] The three thresholds decrease sequentially, corresponding to different risk levels and control strategies, thus achieving comprehensive humidity protection from early warning and prediction to prohibition.

[0076] (8) Parameter adjustment and adaptability It should be noted that the weighting coefficients and humidity thresholds in the above process are empirical values ​​and can be configured and adjusted according to the environmental characteristics and equipment requirements of different sites. For example: In areas with drastic humidity changes, the weighting coefficient for the most recent day can be appropriately increased to make the forecast more sensitive. In scenarios where the equipment has poor insulation performance, the humidity thresholds can be appropriately reduced to increase safety redundancy. In power plants with strong operation and maintenance capabilities, alarm thresholds can be adjusted appropriately to balance safety and energy conservation.

[0077] This adjustable parameter design enables the invention to adapt to diverse application scenarios in different regions, seasons, and with different equipment characteristics, thus having wide applicability.

[0078] When using this invention, it can be summarized as follows: First, the ambient humidity is monitored in real time by a humidity sensor at a first sampling frequency (once per second). The data processing and control module calculates the first representative humidity value (minute average) at a second sampling frequency (once per minute), filters out transient fluctuation interference, and selects the maximum values ​​in two time periods each day, recording them as the first maximum humidity value (reflecting the operating humidity of the day) and the second maximum humidity value (reflecting the nighttime humidity).

[0079] In the tripping decision-making stage, this invention employs a dual judgment mechanism: Firstly, it determines whether the maximum daily humidity exceeds the first humidity threshold (85%RH). If it does, it indicates that the day is already in a high-humidity state, posing a high risk of tripping, and tripping should be prohibited. Secondly, based on historical data of the second maximum humidity, a weighted prediction method is used to calculate the predicted humidity value before closing the circuit the next day, and this value is compared with the second humidity threshold (80%RH). If the predicted value exceeds the threshold, it indicates a high risk of closing the circuit the next morning, and tripping should be prohibited in advance to avoid unsafe closing the circuit the following day. If either condition in the dual judgment is met, the circuit enters a direct-connect mode (keeping the circuit closed, without energy saving); only when neither condition is met is normal tripping performed to achieve energy saving.

[0080] In the closing decision-making process, the present invention handles different situations based on the previous day's status: if the previous day's humidity caused the system to enter the direct mode, the system will continue to be closed and wait for the humidity to improve; if the previous day's system was normally opened, the real-time sampled humidity value before closing (e.g., the average humidity value within 1 hour before closing) will be obtained and compared with the first humidity threshold (85%RH). If the value exceeds the threshold, closing will be prohibited and an alarm will be triggered; otherwise, closing will proceed normally.

[0081] The exit from the direct mode uses hysteresis control, and it will only resume when the maximum first humidity value is lower than the third humidity threshold (75%RH) for three consecutive days to prevent frequent mode switching.

[0082] Through the above mechanism, this invention maximizes the automatic switching control of photovoltaic power plant energy-saving devices while ensuring safety, and resolves the contradiction between operational risks and energy-saving requirements in high-humidity environments.

Claims

1. A method for controlling the opening and closing of an energy-saving device based on humidity prediction, characterized in that: Includes the following steps: Humidity monitoring and recording: Real-time monitoring of ambient humidity, and recording the first maximum humidity value during the period from closing the circuit to opening the circuit each day, and the second maximum humidity value during the period from opening the circuit to the next closing operation each day; The circuit breaker tripping control process involves performing the following checks before the scheduled tripping operation each day: Determine whether the maximum humidity value of the day is greater than the first humidity threshold; Based on the second-highest historical humidity value, predict the humidity value before closing the gate the next day. Determine whether the predicted humidity value is greater than the second humidity threshold; If the first maximum humidity value is greater than the first humidity threshold, or the humidity prediction value is greater than the second humidity threshold, then the daily circuit breaker operation is prohibited, and the energy-saving device is controlled to enter the direct-flow mode. Otherwise, perform a trip operation; Closing control process: When the closing operation is planned to be executed: If the energy-saving device entered the direct-flow mode due to humidity conditions the previous day, it will remain in the closed state and the direct-flow mode during the closing operation on the current day. If a normal tripping operation was performed the previous day, the real-time sampled humidity value at the current moment is obtained, and it is determined whether the real-time sampled humidity value is greater than the first humidity threshold. If it is, the closing operation is prohibited and a fault alarm is issued; otherwise, the closing operation is performed.

2. The method for controlling the opening and closing of an energy-saving device based on humidity prediction according to claim 1, characterized in that: The prediction of the humidity value before closing the circuit breaker the next day based on the second highest historical humidity value data is as follows: S1. Predict the base humidity for the current day based on the historical morning humidity of the previous 7 consecutive days. S2. The baseline humidity for the day is adjusted based on the temperature sampled that day for humidity compensation. S3. The prediction weighting coefficient is automatically adjusted daily based on actual humidity sampling.

3. The method for controlling the opening and closing of an energy-saving device based on humidity prediction according to claim 1, characterized in that: The tripping control process further includes: if the first maximum humidity value is greater than the third humidity threshold and less than or equal to the first humidity threshold, then an alarm signal is output.

4. The method for controlling the opening and closing of an energy-saving device based on humidity prediction according to claim 3, characterized in that: The condition for exiting the direct-access mode is: using hysteresis control, the maximum value of the first humidity recorded for M consecutive days is less than or equal to the third humidity threshold.

5. The method for controlling the opening and closing of an energy-saving device based on humidity prediction according to claim 4, characterized in that: The first humidity threshold is 85%RH, the second humidity threshold is 80%RH, and the third humidity threshold is 75%RH.

6. The method for controlling the opening and closing of an energy-saving device based on humidity prediction according to claim 5, characterized in that: The value of M is 3.

7. The method for controlling the opening and closing of an energy-saving device based on humidity prediction according to claim 1, characterized in that: The direct-through mode is as follows: the high-voltage switch in the energy-saving device remains closed and does not perform time-based automatic tripping operation.

8. The method for controlling the opening and closing of an energy-saving device based on humidity prediction according to claim 1, characterized in that: The real-time monitoring of ambient humidity includes: periodically acquiring raw humidity data at a preset first sampling frequency; specifically, recording the first maximum humidity value and recording the second maximum humidity value involves: based on the raw data acquired at the first sampling frequency, further calculating the first representative humidity value at the second sampling frequency, and selecting the maximum value among all the first representative humidity values ​​in the two time periods of "from closing the circuit to before the opening operation" and "from opening the circuit to before the next closing operation" each day, as the corresponding first maximum humidity value and second maximum humidity value; The real-time sampled humidity value is calculated by taking multiple raw humidity data obtained at the first sampling frequency within a preset time window before the closing operation, and obtaining a second representative humidity value within the time window, which is used as the real-time sampled humidity value.

9. The method for controlling the opening and closing of an energy-saving device based on humidity prediction according to claim 8, characterized in that: The first and second humidity representative values ​​are the average values ​​of multiple raw humidity data within the corresponding time period.

Images