Control method for hollow louver glass energy-saving air conditioner

Abstract

The application discloses a hollow louver glass energy-saving air conditioner control method, and belongs to the field of building environment and equipment energy-saving control. In order to solve the problem that the shading, lighting and air conditioner load of the hollow louver Low-E glass lack collaborative control, the application identifies the window thermal parameters by the Low-E film surface orientation and the louver angle, decomposes the solar heat gain residual error and the temperature difference heat transfer residual error, predicts the window net heat flow and the indoor environment trajectory under the candidate control combination, and jointly determines the louver angle, the air supply and the set temperature, so that the window load reduction and the air conditioner residual load are coordinately matched.

Description

Technical Field

[0001] This invention relates to the field of energy-saving control of building environment and equipment, and in particular to a method for controlling energy-saving air conditioning with insulated louvered glass. Background Technology

[0002] Insulating glass with adjustable louvers incorporates adjustable louvers within the glass cavity, allowing for the adjustment of solar radiation and visible light entering the room by changing the louver angle. A Low-E coating improves the window's thermal insulation performance by reducing radiative heat transfer across the glass surface. Existing buildings typically employ separate louver controllers and air conditioning controllers. The louver controller primarily adjusts the angle based on solar radiation or illuminance, while the air conditioning controller adjusts the airflow or sets the temperature based on room temperature deviations.

[0003] Since the orientation of the Low-E membrane surface, the angle of the louvers, and the heat storage capacity of the hollow cavity all affect the solar heat gain and temperature difference heat transfer of the window, it is difficult to cover different solar radiation and temperature difference conditions with fixed window parameters. The existing control method also lacks a mechanism to separate the heat flow channel of the window online by utilizing the temperature change of the zone and the measured heat exchange of the air conditioner. It is easy to continue to update parameters when the excitation is insufficient or the sensor is abnormal, resulting in the accumulation of prediction errors and fluctuations in air conditioning load.

[0004] In addition, louver angle, air volume and set temperature are usually adjusted independently according to their respective local targets, without comparing candidate combinations in the same prediction time domain under constraints of illuminance, glare, thermal comfort and equipment capacity. Therefore, problems such as excessive shading, insufficient indoor illuminance, or mismatch between the load reduction of louvers and the remaining load undertaken by the air conditioner may occur.

[0005] Therefore, there is a need for an energy-saving air conditioning control method for hollow louvered glass that can overcome the shortcomings of the existing technology. Summary of the Invention

[0006] One objective of this invention is to propose an energy-saving air conditioning control method for hollow louvered glass. Addressing the problems in existing technologies where there is a lack of coordination between louver angle, Low-E film thermal performance, air supply volume, and set temperature, and the difficulty in reliably identifying window parameters under different sunlight, temperature differences, and air conditioning operating conditions, this invention proposes a method for evaluating the effectiveness and excitation of multi-source operating data. It utilizes thermal zone energy balance to form the window's total heat flux residual and decomposes solar heat gain and temperature difference heat transfer channels. The invention updates window thermal parameters and uncertainties according to the film orientation and louver angle intervals, and then evaluates candidate combinations of louver angle, air supply volume, and set temperature to jointly determine the control target. This invention achieves the technical effect of balancing lighting, thermal comfort, and equipment feasibility, while coordinating window load reduction with the remaining air conditioning load.

[0007] This invention provides a method for controlling an energy-saving air conditioner with hollow louvered glass, comprising:

[0008] S1. Collect the operating data of building thermal zones, glass windows with hollow louvers and Low-E film layers, and air conditioning equipment, as well as the Low-E film orientation markings. Perform time alignment, data validity processing, and valid excitation judgment on the operating data to obtain a valid observation dataset and excitation status.

[0009] S2. Based on the temperature change and measured heat exchange of the air conditioner in the effective observation dataset, construct the total heat flux residual of the window. According to the excitation relationship between the solar radiation change and the indoor and outdoor temperature difference change, decompose it into solar heat gain residual and temperature difference heat transfer residual.

[0010] S3. According to the Low-E membrane surface orientation mark and louver angle interval index window thermal parameter unit, only when the excitation state meets the identifiable condition of the corresponding residual channel, update the effective solar heat gain coefficient, effective heat transfer coefficient, cavity heat storage state parameters and uncertainty using two types of residuals; when the update condition is not met, maintain the corresponding parameters and adjust their uncertainty to obtain the window thermal state parameter set.

[0011] S4. Based on the set of thermal state parameters of the window, future solar radiation and future outdoor temperature, evaluate the candidate combinations of louver angle, air conditioning air volume and air conditioning set temperature, and generate the window net heat flow prediction, room temperature trajectory, indoor illuminance trajectory and air conditioning load trajectory with uncertainty under each candidate combination.

[0012] S5. Using the predicted net heat flow of the window and its uncertainty and the three types of trajectories as input, determine the control feasible region under the constraints of illuminance, glare, thermal comfort and air conditioning equipment capacity, and jointly determine and execute the target louver angle, the target value of air conditioning air supply volume and the target value of air conditioning set temperature. Return the execution feedback of the next control cycle to steps S1 and S2, and correct the window thermal state parameter set according to the update conditions of step S3.

[0013] Optionally, S1 includes:

[0014] The operating data includes at least one of the following: indoor temperature, outdoor temperature, solar radiation, sun position, surface temperature of both sides of the glass and cavity temperature, window temperature, current angle of the louvers, indoor illuminance, supply air temperature, return air temperature, supply air volume, air conditioning power, and air conditioning set temperature.

[0015] The running data is resampled and timestamp corrected based on the data acquisition time and preset control cycle;

[0016] Perform drift checks based on the range boundaries and rate of change per unit time boundaries of the corresponding measurement channels in the sensor calibration table;

[0017] Information matrices are constructed for the solar radiation variation sequence and the indoor-outdoor temperature difference variation sequence, respectively, and the excitation state is determined by whether the determinant value of each information matrix reaches the preset threshold of the corresponding channel.

[0018] Optionally, S2 includes:

[0019] Taking the heat flowing into the hot zone as positive, the change in energy storage in the hot zone is determined based on the equivalent heat capacity of the air and the building envelope in the hot zone and the temperature change in the zone. The measured heat exchange of the air conditioner with the flow direction symbol is determined based on the air supply volume, the specific heat capacity of the air, and the difference between the supply air temperature and the return air temperature. The heat flow at the non-window boundary is determined based on the heat transfer parameters of the non-window building envelope and the infiltration ventilation volume.

[0020] The residual heat flux of the window is obtained by subtracting the internal heat gain, the measured heat exchange of the air conditioner, and the heat flux at the non-window boundary from the change in the energy storage of the hot zone.

[0021] Using the solar radiation variation sequence as the first regression basis, the indoor and outdoor temperature difference variation sequence after deducting its projection on the first regression basis is used as the second regression basis. The projection components of the total heat flux residual of the window on the first and second regression bases are successively determined as the solar heat gain residual and the temperature difference heat transfer residual.

[0022] Optionally, S3 includes:

[0023] The orientation category of the Low-E film surface located on the outer surface of the outdoor glass, the cavity side surface of the outdoor glass, the cavity side surface of the indoor glass, or the inner surface of the indoor glass is combined with the louver angle range divided according to the adjustable louver angle range to form an index.

[0024] Each index corresponds to a window thermal parameter unit that stores the effective solar heat gain coefficient, effective heat transfer coefficient, cavity heat storage time constant, cavity heat storage state quantity, and the covariance of each parameter.

[0025] The preset physical allowable range is determined by the intersection of the parameter range given by the applicable architectural glass standard and the parameter range given by the corresponding glass product parameter, and the inspection parameter range of the corresponding glass product is used when the intersection is empty;

[0026] Furthermore, the effective solar heat gain coefficient is updated for the solar heat gain residual channel, the effective heat transfer coefficient is updated for the temperature difference heat transfer residual channel, and the cavity heat storage time constant and cavity heat storage state quantity are updated according to the time-series decay components of the two residual channels.

[0027] When the determinant value of the corresponding information matrix reaches a preset threshold and the residual is within the confidence interval determined based on the historical effective residual distribution, a recursive update is performed, the recursive update value is projected to the preset physical allowable interval, and the corresponding covariance is reduced.

[0028] When the determinant value does not reach the preset threshold or the residual exceeds the confidence interval, the corresponding parameter value remains unchanged, and the corresponding covariance is increased according to the number of consecutive freeze cycles.

[0029] Optionally, S4 includes:

[0030] The future sun position is calculated based on the building's latitude and longitude, window orientation, and time of prediction within the time domain. The incident direction of sunlight relative to the window is determined based on the future sun position. For each candidate louver angle, the direct light blocking ratio and the predicted indoor illuminance value are calculated.

[0031] Input the louver angle range to which the candidate louver angle belongs, the Low-E membrane orientation identifier, future solar radiation, and future outdoor temperature into the corresponding window thermal parameter unit to calculate the expected value of the net heat flux of the window in the prediction time domain and the confidence boundary.

[0032] The expected net heat flux of the window and the confidence boundary are used as the boundary heat flux input of the hot zone prediction model. Combined with the candidate air supply volume and candidate set temperature of the air conditioner, the room temperature trajectory, indoor illuminance trajectory and air conditioner load trajectory corresponding to each candidate combination are obtained.

[0033] Optionally, S5 includes:

[0034] The feasible boundaries of the thermal comfort constraint and the air conditioning equipment capacity constraint are tightened based on the uncertainty of the net heat flow prediction of the window.

[0035] The net heat flow of the window corresponding to the initial louver angle of the control cycle is used as the benchmark value. The difference between the benchmark value and the net heat flow of the window corresponding to the candidate angle is used as the window load reduction amount. The hot zone prediction model forms the hot zone prediction load based on the future environmental input, internal heat gain prediction and the benchmark value. The portion of the hot zone prediction load after deducting the window load reduction amount is used as the remaining load borne by the air conditioner.

[0036] Using the weighted sum of air conditioning power consumption, louver operation frequency, set temperature change, and illuminance deviation in the predicted time domain as the control objective, within the feasible domain jointly defined by illuminance, glare, thermal comfort, air conditioning equipment capacity, louver holding time, and control rate constraints, the system solves for the target louver angle, air conditioning air volume, and air conditioning set temperature that satisfy the remaining load on the air conditioning load trajectory. The constraint boundaries and weights are given in advance by the building operation parameter table.

[0037] Furthermore, a candidate solution for the control objective is determined by coordinating the fast and slow load share layers. The slow layer determines the window load reduction share borne by the louvers according to the expected net heat flux of the window in the predicted time domain, the indoor illuminance requirement, and the louver action constraints, and determines the target angle and holding range of the louvers.

[0038] Within the holding interval, the fast layer adjusts the target value of the air supply volume and the target value of the air conditioning set temperature based on the deviation of the actual zone temperature from the room temperature prediction value of the corresponding control cycle of the selected candidate combination, and the deviation of the actual air conditioning power from the predicted air conditioning power of the candidate combination in the same control cycle. The fast layer adjustment shall not change the illuminance constraint and glare constraint satisfaction status determined by the slow layer.

[0039] Furthermore, four control states are set: normal, frozen, downgraded, and restored;

[0040] When the corresponding residual channel experiences an excitation anomaly where the determinant value of the information matrix does not reach the threshold, or when a residual anomaly occurs where the residual exceeds the confidence interval determined based on the historical effective residual distribution, it enters a frozen state, stops updating the corresponding window thermal parameters, and tightens the control constraints using the confidence boundary predicted by the window net heat flow.

[0041] When the sensor drift check fails, the joint control problem has no feasible solution, or the number of consecutive freeze cycles reaches the preset cycle threshold, the system enters a degraded state. The net heat flow of the window is predicted according to the conservative effective solar heat gain coefficient and effective heat transfer coefficient in the product parameters, and the louver angle is kept at the nearest feasible angle that meets the preset lower limit of illuminance. The remaining heat zone load is borne by the air conditioning equipment. When any of the aforementioned degraded conditions are met, the state transition of the degraded state takes precedence over the freeze state and the recovery state.

[0042] When the drift check passes, the determinant value of the corresponding information matrix continuously reaches the threshold, and the number of cycles in which the residual is continuously within the confidence interval reaches the recovery threshold, the recovery state is entered. The constraints are released and tightened periodically according to the preset rate of change, and the parameter updates are restored. Then, the state returns to normal.

[0043] The preset cycle threshold, recovery threshold, and preset change rate are given in advance in the building operation parameter table based on the control cycle length and the window thermal parameters to estimate the window length.

[0044] Furthermore, in each control cycle, the following information is saved: Low-E membrane orientation identifier, louver angle interval index, window thermal state parameter set and its version, effective observation dataset, excitation state, residual decomposition results, window net heat flow prediction, joint control target value, control state, and execution feedback.

[0045] The target louver angle, air conditioning air volume target value, and air conditioning set temperature target value are input into the hot zone prediction model and the air conditioning equipment prediction model to obtain the predicted zone temperature change, predicted air conditioning heat exchange and predicted air conditioning power for the next control cycle.

[0046] The window total heat flow verification residual is formed based on the difference between the actual temperature change and the predicted temperature change, and the difference between the measured value obtained according to the method for determining the measured heat exchange of the air conditioner and the predicted heat exchange of the air conditioner. The window total heat flow verification residual is decomposed into solar heat gain channel verification result and temperature difference heat transfer channel verification result using the first regression basis and the second regression basis. The difference between the actual air conditioner power and the predicted air conditioner power is determined as the equipment verification result.

[0047] The freezing condition for this item is determined when only one channel's verification result continuously exceeds the verification interval determined based on the corresponding historical valid verification result, and the device verification result is continuously within the corresponding verification interval. The degrading condition for this item is determined when the sensor drift check fails, the joint control problem has no feasible solution, the verification results of two channels continuously exceed the corresponding verification interval, the device verification result and any channel verification result continuously exceed the corresponding verification interval, or the number of consecutive freezing cycles reaches the preset cycle threshold. The recovery condition for this item is determined when the number of cycles in which the verification results of two channels and the device verification result are continuously within the corresponding verification interval reaches the verification recovery threshold. The verification recovery threshold is given in advance in the building operation parameter table according to the control cycle length and the closed-loop verification window length.

[0048] The aforementioned triggering conditions and the current state conditions are jointly determined for the four control states. If any degradation condition is met, the state enters the degradation state; if there is no degradation condition and any freezing condition is met, the state enters the freezing state; and if there are no degradation or freezing conditions and both the aforementioned recovery condition and the current recovery condition are met, the state enters the recovery state. In other cases, the current control state is maintained.

[0049] The beneficial effects of this invention are:

[0050] 1. By using the Low-E membrane orientation and louver angle range as the index for the window thermal parameter unit, and performing identifiable gating updates on the solar heat gain residual and the thermal difference heat transfer residual respectively, the prediction deviation of the window net heat flow caused by the mutual confusion of different heat transfer mechanisms can be reduced.

[0051] 2. By freezing parameters, increasing uncertainty, or entering degraded control when there is insufficient excitation, abnormal residuals, or abnormal equipment calibration, and by coordinating freezing, degraded control, and recovery with a unified state priority, it is possible to avoid continuous parameter drift caused by abnormal data and maintain the determinism of control state transitions.

[0052] 3. By evaluating candidate combinations of louver angle, air conditioning air volume and set temperature within the same prediction time domain, and solving them jointly under constraints of illuminance, glare, thermal comfort and equipment capacity, excessive shading and insufficient lighting can be reduced, so that the share of window load reduction borne by louvers and the remaining load borne by air conditioning can be matched. Attached Figure Description

[0053] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0054] Figure 1 This is a flowchart of the energy-saving air conditioning control method for hollow louvered glass in this invention.

[0055] Figure 2 This is a flowchart of the S2 method of the present invention, which decomposes the total heat flux residual of the window based on orthogonal projection. Detailed Implementation

[0056] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0057] refer to Figures 1-2 Energy-saving air conditioning control methods for hollow louvered glass include:

[0058] S1. Collect the operating data of building thermal zones, glass windows with hollow louvers and Low-E film layers, and air conditioning equipment, as well as the Low-E film orientation markings. Perform time alignment, data validity processing, and valid excitation judgment on the operating data to obtain a valid observation dataset and excitation status.

[0059] S2. Based on the temperature change and measured heat exchange of the air conditioner in the effective observation dataset, construct the total heat flux residual of the window. According to the excitation relationship between the solar radiation change and the indoor and outdoor temperature difference change, decompose it into solar heat gain residual and temperature difference heat transfer residual.

[0060] S3. According to the Low-E membrane surface orientation mark and louver angle interval index window thermal parameter unit, only when the excitation state meets the identifiable condition of the corresponding residual channel, update the effective solar heat gain coefficient, effective heat transfer coefficient, cavity heat storage state parameters and uncertainty using two types of residuals; when the update condition is not met, maintain the corresponding parameters and adjust their uncertainty to obtain the window thermal state parameter set.

[0061] S4. Based on the set of thermal state parameters of the window, future solar radiation and future outdoor temperature, evaluate the candidate combinations of louver angle, air conditioning air volume and air conditioning set temperature, and generate the window net heat flow prediction, room temperature trajectory, indoor illuminance trajectory and air conditioning load trajectory with uncertainty under each candidate combination.

[0062] S5. Using the predicted net heat flow of the window and its uncertainty and the three types of trajectories as input, determine the control feasible region under the constraints of illuminance, glare, thermal comfort and air conditioning equipment capacity, and jointly determine and execute the target louver angle, the target value of air conditioning air supply volume and the target value of air conditioning set temperature. Return the execution feedback of the next control cycle to steps S1 and S2, and correct the window thermal state parameter set according to the update conditions of step S3.

[0063] In this specific embodiment, S1 includes:

[0064] According to the preset control cycle Simultaneous data collection was conducted on building thermal zones, glass windows with double-glazed louvers and Low-E film coatings, and air conditioning equipment. The collected operational data included indoor temperature. Outdoor temperature Solar radiation Solar altitude angle Sun azimuth , Temperature of the outer surface of the glass , Inner surface temperature of glass Hollow cavity temperature Current angle of the louvers , indoor illumination air supply temperature Return air temperature Air volume Air conditioner power and the air conditioner set temperature And read the Low-E membrane orientation indicator. ,in This indicates that the Low-E film is located on the outer surface of the outdoor glass. This indicates that the Low-E film is located on the cavity-side surface of the outdoor glass. This indicates that the Low-E film is located on the cavity-side surface of the interior glass. This indicates that the Low-E film is located on the inner surface of the interior glass.

[0065] To control the endpoints of the cycle As a unified time reference, among which For the first Timestamp of each control cycle This is the initial timestamp. To control the cycle number, the average value of the cycle is used as the observation value for temperature, solar radiation, indoor illuminance, air supply volume, and air conditioning power. The current louver angle and the air conditioning set temperature are used as the effective value at the end of the cycle. The solar altitude angle and solar azimuth angle are used as the midpoint value of the cycle calculated from the building's latitude and longitude, date and time, and window orientation. When the original sampling time of the same measurement channel is... The deviation does not exceed Resampling is performed in real time. When a single cycle is missing a measurement, linear interpolation between two adjacent valid cycles is used to obtain a replacement value. When the number of consecutive missing cycles is greater than 1, the corresponding time period is removed from the valid observation dataset. Data validity processing is performed on each measurement channel according to the sensor calibration table. The sensor calibration table stores the lower limit, upper limit, and upper limit of the rate of change per unit time for each measurement channel. When the measured value exceeds the corresponding range boundary, or the change in the measured value between two adjacent control cycles is divided by... When the rate of change exceeds the corresponding unit time boundary, the measurement channel is marked as a drift anomaly in the corresponding period. Drift anomaly data does not participate in subsequent residual construction and parameter updates. The synchronous data retained after time alignment and drift check constitutes a valid observation dataset.

[0066] The most recent within the valid observation dataset The data from each control cycle are used to determine the effective stimulus and generate solar radiation variation sequences. and indoor-outdoor temperature difference variation sequence and in accordance with , Construct an information matrix, where This is the information matrix for solar heat gain channels. This is the information matrix for the temperature difference heat transfer channel. To estimate the control cycle number within the window, To estimate the number of control cycles contained in the window, This is the regression vector for changes in solar radiation. The regression vector represents the change in indoor and outdoor temperature difference. For the first The change in solar radiation in each control cycle relative to the previous control cycle. For the first The change in indoor and outdoor temperature difference relative to the previous control period, [ ] Indicates transpose. This represents the summation over each control cycle within the estimation window, when... The solar heat gain channel is marked as having effective excitation when it is active, and otherwise marked as having insufficient excitation when it is active. The temperature difference heat transfer channel is marked as having effective excitation when it is active, and otherwise marked as having insufficient excitation. Represents the determinant of a matrix. Pre-set thresholds for solar heat gain channels and take Preset a threshold for the temperature difference heat transfer channel and take The excitation state is formed by the combined effect of effective or insufficient excitation of the solar heat gain channel and the temperature difference heat transfer channel.

[0067] In this specific embodiment, S2 includes:

[0068] Using an effective observation dataset, the energy balance of the thermal zone is established with the heat flowing into the thermal zone as the positive value, and this is done in each control cycle. The total heat flux residual of the form is calculated, and the change in energy storage in the hot zone is determined by the equivalent heat capacity of the air in the hot zone and the building envelope. Determined by changes in indoor temperature, Given and taken from the building operation parameter table This value includes the heat capacity of the hot zone air and the equivalent heat capacity of the indoor building envelope;

[0069] Actual heat exchange of air conditioner is measured according to air volume. air density Specific heat capacity of air air supply temperature With return air temperature The difference is determined. Pick Pick ,when Furthermore, when the air conditioner is in cooling mode, the measured heat exchange rate is negative, indicating that the air conditioner is removing heat from the hot area. Furthermore, when the air conditioner is in heating mode, the measured heat exchange rate is positive, indicating that the air conditioner is inputting heat into the hot area.

[0070] Non-window boundary heat flux includes heat flux transferred through the non-window envelope and heat flux transferred through infiltration ventilation. The heat transfer parameters of the non-window envelope... Given and taken from the building operation parameter table osmotic ventilation Given and taken from the building operation parameter table Internal heat gain The operating plans for personnel, lighting, and indoor equipment are determined and stored in the building operation parameter table according to the control cycle; after completing the calculation of the above quantities, according to:

[0071] ,

[0072] ;

[0073] The total heat flux residual of the form is obtained, where For the first Changes in thermal energy storage over a control cycle The equivalent heat capacity of the hot zone, For the first Indoor temperature for each control cycle, For the first Indoor temperature for each control cycle, To preset the control cycle, For the first Measured heat exchange of the air conditioner during each control cycle air density, The specific heat capacity of air, For the first Air volume per control cycle. For the first The supply air temperature for each control cycle. For the first Return air temperature for each control cycle For the first Non-window boundary heat flow in each control cycle These are the heat transfer parameters for non-window-shaped building envelopes. For the first Outdoor temperature for each control cycle, This refers to the osmotic ventilation rate. For the first Internal heat gain per control cycle For the first The total heat flux residual of the window in each control cycle;

[0074] For the recent The residual sequence is composed of the total heat flux residual of the window in each control cycle, and the solar radiation variation sequence in step S1 is used. As the first regression base, and using the indoor and outdoor temperature difference change sequence... Subtracting its projection onto the first regression basis is used as the second regression basis. The first regression basis represents the solar heat gain channel caused by changes in solar radiation in the heat flow of the window, and the second regression basis represents the temperature difference heat transfer channel caused by changes in indoor and outdoor temperature difference after excluding the solar radiation related components.

[0075] In the specific decomposition, the projection component of the residual sequence in the first regression basis direction is first determined as the solar heat gain residual. Then, the projection component of the remaining residual after deducting the solar heat gain residual in the second regression basis direction is determined as the temperature difference heat transfer residual. The unexplained part after projection is saved as a random disturbance term but is not used for parameter recursion update in step S3.

[0076] Therefore, we obtain the first The solar heat gain residual and temperature difference heat transfer residual corresponding to each control cycle are written together with the total heat flux residual of the window, the first regression basis, the second regression basis and the random disturbance term into the residual decomposition result, so that step S3 can perform sub-channel parameter update according to the Low-E membrane surface orientation label and the louver angle interval.

[0077] In this specific embodiment, S3 includes:

[0078] Low-E membrane orientation marking Current angle of the louvers Information matrix of solar heat gain channels Information matrix of temperature difference heat transfer channels and solar heat gain residual and temperature difference heat transfer residual , for the The thermal parameter unit of the window in each control cycle is updated in separate channels;

[0079] The adjustable angle range of the louvers is set to It is divided into five louver angle ranges, namely , and ,when When the value falls within the corresponding interval, determine the louver angle interval index. ,in This refers to the louver angle interval number;

[0080] Low-E film surface orientation marking Following the definition of step S1, This indicates that the Low-E film is located on the outer surface of the outdoor glass. This indicates that the Low-E film is located on the cavity-side surface of the outdoor glass. This indicates that the Low-E film is located on the cavity-side surface of the interior glass. This indicates that the Low-E film is located on the inner surface of the interior glass.

[0081] Will and Combined to form window thermal parameter unit index Each window's thermal parameter unit stores the effective solar heat gain coefficient. Effective heat transfer coefficient Cavity heat storage time constant , cavity heat storage state quantity and the parameter covariance matrix ,in This is a dimensionless parameter used to characterize the proportion of effective solar heat gain entering the hot zone under the corresponding membrane surface orientation and louver angle range. The unit is This is used to characterize the effective temperature difference heat transfer capacity of the window under the corresponding membrane surface orientation and louver angle range. The unit is seconds (s), used to characterize the decay time of heat absorption and release to the hot zone in the hollow cavity and louvers. The unit is W, used to characterize the first... The hysteretic contribution of cavity heat storage to the net heat flow of the window during each control cycle Used to characterize the uncertainty of the above parameter estimates; during initialization, Take the nominal value of the solar heat gain coefficient for the same film surface orientation and louver angle range from the corresponding glass product inspection report. Take the nominal value of the heat transfer coefficient from the corresponding glass product inspection report. Pick Pick Take the diagonal matrix Its four diagonal elements correspond to and The initial variance;

[0082] The preset physical allowable range is determined by the intersection of the parameter range given by the applicable architectural glass standard and the parameter range given by the corresponding glass product parameter. In this embodiment, the standard range of the effective solar heat gain coefficient is... The product parameter range is The standard range of the effective heat transfer coefficient is: The product parameter range is The standard range of the cavity heat storage time constant is: The product parameter range is ,in This corresponds to the nominal value of the solar heat gain coefficient in the glass product inspection report. To correspond to the nominal value of the heat transfer coefficient in the glass product inspection report, For the nominal value of the cavity heat storage time constant given in the corresponding glass product inspection report or building operation parameter table, when any intersection is empty, the inspection parameter range of the corresponding glass product is used as the preset physical allowable range of the parameter.

[0083] Parameter updates employ a channel-gated recursive method, with the solar heat gain channel used solely for updates. The temperature difference heat transfer channel is only used for replacement. The time-decrease components of the two residual channels are used together to update and In the Within each control cycle, according to Perform recursion, where The parameter vector before the update. For the updated parameter vector, The components are, in order, the current index. Below and , This indicates that the recursive update value will be projected to a preset physical allowable range. , For the current window's thermal parameter unit and The set of physically allowed intervals The recursive gain matrix is ​​determined by the building operation controller according to the recursive least squares rule. and calculate, The residual vector is used to determine the corresponding component when the solar heat gain channel is effective. When the temperature difference heat transfer channel is effective, the corresponding component is taken. Take when both channels are valid For the observation matrix, its solar heat gain channel elements are determined by the first... Solar radiation in a control cycle With window area Its composition, its temperature difference heat transfer channel elements are determined by the window area With indoor and outdoor temperature difference constitute, Given and taken from the building operation parameter table The parameter covariance matrix before the update. The updated parameter covariance matrix, This is the covariance adjustment coefficient. It is the identity matrix. This refers to the updated cavity heat storage state. This is the cavity heat storage state quantity updated in the previous control cycle. It is a natural constant. To preset the control cycle, The updated cavity heat storage time constant. For the first Solar heat gain residual for each control cycle For the first The residual temperature transfer over each control cycle, [ ] Indicates transpose;

[0084] The identifiable conditions for solar heat gain channels are: and The identifiable condition for a temperature difference heat transfer path located within the confidence interval of the solar heat gain residual is: and Located within the confidence interval of the thermal transfer residual, the confidence intervals of the solar heat gain residual and the thermal transfer residual are respectively derived from the nearest... The mean of the historical effective residuals plus or minus 3 standard deviations is used to determine the value.

[0085] When the corresponding residual channel meets the identifiable condition, perform a recursive update for that channel and retrieve the result. This reduces the covariance of the corresponding parameter.

[0086] When the determinant value of the corresponding information matrix does not reach the preset threshold, or the corresponding residual exceeds its confidence interval, the corresponding parameter value of the channel is kept unchanged, the channel is marked as frozen, and the value is taken as follows: Add the corresponding parameter covariance, where This represents the number of consecutive freeze cycles for the corresponding residual channel;

[0087] After completing the channel-by-channel recursive update, physical allowable interval projection, cavity heat storage state update, and covariance adjustment, the current index will be... corresponding The number of consecutive freeze cycles and parameter version number are written into the thermal status parameter set of the form. Other thermal parameter units of the form that are not hit by the current Low-E membrane orientation identifier and louver angle range index retain the parameter values ​​and status records of the previous control cycle.

[0088] In this specific embodiment, S4 includes:

[0089] With the first Starting with the set of window thermal state parameters updated in each control cycle, the prediction time domain is set. The control cycle has one prediction step size, which is consistent with the control cycle. And construct candidate combinations ,in Pick and , Pick , and , Pick and , The candidate combination number is used, and each candidate combination is assigned a louver angle range to which its louver angle belongs. Similar to the Low-E membrane orientation marking in step S3 Enter the corresponding window thermal parameter unit together;

[0090] The future position of the sun will be calculated using a standard solar position algorithm, which uses building latitude as the basis for the calculation. ,longitude Window orientation The system takes the date and time of each moment within the prediction time domain as input, calculates the solar hour angle and declination angle from the date and time, and outputs the future solar altitude angle and solar azimuth angle by combining the building latitude, longitude, and window orientation. The incident direction is then determined by the angle between the sunlight and the window normal, along with the direct sunlight shading ratio. According to the width of the louvers louver spacing and louver thickness The geometric projection relationship is calculated by taking a value of 1 when the light is completely blocked by the veils and 0 when the light passes through the gaps in the veils. For other cases, the result is obtained by linear interpolation of the projection overlap length and truncated to the interval [0, 1]. The indoor illuminance trajectory is calculated using the transmission illuminance model, based on the outdoor horizontal illuminance and the visible light transmittance of the window. The indoor illuminance prediction values ​​for the corresponding candidate combinations are obtained by superimposing the direct and scattered components with indoor surface reflectance of 0.50, wall reflectance of 0.60, and floor reflectance of 0.20. And it forms an indoor illuminance trajectory by rolling within the prediction time domain;

[0091] In the prediction of net heat flow in the form, according to:

[0092] ;

[0093] in For the first Control cycle, candidate combination The expected net heat flux of the window is as follows. and These are its 95% confidence lower bound and confidence upper bound, respectively. For the effective solar heat gain coefficient, Indicator for Low-E membrane orientation. Index for louver angle range, The area of ​​the window. This is a predicted value for future solar radiation. This is a forecast of future outdoor temperatures. For the first Measured indoor temperature values ​​for each control cycle For the effective heat transfer coefficient, This refers to the heat storage state of the cavity. For the proportion of direct sunlight blocking, For the parameter covariance matrix The standard deviation of the net heat flux of the window obtained through first-order linearized propagation is 1.96, which is two-sided. Normal quantile coefficients;

[0094] The thermal zone prediction model adopts a first-order lumped-parameter discrete model, with room temperature as the state variable. State updates are performed using the forward Euler method, with hot capacity taken. The overall heat transfer coefficient of the hot zone is taken as The internal heat gain is input according to the time period plan in the building operation parameter table, with 1.2kW for working hours and 0.4kW for non-working hours. The air conditioning equipment prediction model adopts a bilinear performance surface, based on candidate air supply volumes. and candidate set temperature The corresponding air conditioning heat exchange and power are obtained by looking up the table and interpolating. The performance surface consists of three sets of air volume nodes and three sets of set temperature nodes. The node data comes from the actual measurement records of the equipment under rated operating conditions. The expected value of the net heat flux of the window and its confidence boundary are used as the boundary heat flux input of the hot zone prediction model. Together with the candidate air volume and candidate set temperature, the room temperature trajectory, indoor illuminance trajectory and air conditioning load trajectory of each candidate combination in the prediction time domain are obtained by rolling the solution.

[0095] In this specific embodiment, S5 includes:

[0096] Receive each candidate combination The corresponding expected net heat flux of the window , Lower bound of net heat flow confidence for the form , upper limit of window net heat flow confidence Room temperature trajectory Indoor illuminance trajectory and air conditioning load trajectory ,in This refers to the prediction step number within the prediction time domain. To predict the length of the time domain;

[0097] The feasible control domain is first screened based on constraints of illuminance, glare, thermal comfort, and air conditioning equipment capacity. The illuminance constraint is set as follows: 300 lx is the lower limit of illuminance for maintaining basic visual operations in the office hot zone, and 1500 lx is the upper limit of illuminance to avoid excessive brightness and reflection interference; glare constraint adopts the daylight glare probability index. The brightness of the window surface, the direction of the line of sight, and the indoor illuminance are interpolated according to the solar glare probability calculation table stored in the building operation parameter table. The brightness of the window surface is calculated from the future position of the sun, the candidate angle of the louvers, and the proportion of direct light blocking.

[0098] Thermal comfort constraints use room temperature boundary And based on the uncertainty of the net heat flow prediction of the window. Tighten ,in Bilateral Normal quantile coefficients The overall heat transfer coefficient for the hot zone is used; the capacity constraint of the air conditioning equipment is adopted. And tightened to the uncertainty of the net heat flow prediction of the window. ,in The maximum available heat exchange capacity of the air conditioning equipment is given in the building operation parameter table; the louver holding time constraint is set to maintain the same louver target angle for at least 3 control cycles after execution, and the control rate constraint is set to ensure that the louver angle change between two adjacent control cycles does not exceed the maximum allowable heat exchange capacity. The change in air supply volume shall not exceed The change in the air conditioner's set temperature should not exceed ;

[0099] After completing the constraint screening, the initial louver angle is controlled by the cycle. The corresponding net heat flow of the window is used as the baseline value. The difference between the net heat flow of the window corresponding to the baseline value and the candidate louver angle is used as the window load reduction amount. The hot zone prediction model forms the hot zone prediction load based on future environmental input, internal heat gain prediction and baseline value. The portion of the hot zone prediction load after deducting the window load reduction amount is determined as the remaining load borne by the air conditioner.

[0100] Joint control objectives according to , Solve for, where Candidate combinations The value of the control objective function, To meet the controllable feasible domain constraints of illuminance, glare, thermal comfort, air conditioning equipment capacity, louver holding time, and control rate, Weighting of air conditioner power consumption. The air conditioning power is obtained from the air conditioning equipment prediction model in step S4. To preset the control cycle, For illuminance deviation weight, For indoor target illuminance, Set temperature change weights for the air conditioner. Set the temperature for the candidate air conditioner. For the first The current air conditioning set temperature for each control cycle. Weights for the louver motion. For candidate louver angles, For the first The current louver angle for each control cycle For prediction steps Candidate Combinations The corresponding form load reduction amount, This represents the expected net heat flux of the window corresponding to the initial louver angle. Candidate combinations The corresponding expected net heat flux of the window, The remaining load to be borne by the air conditioner This refers to the predicted load for the hot zone generated by the hot zone prediction model under the initial louver angle reference value. Represents absolute value. This represents the summation of prediction steps over the prediction time domain;

[0101] The controller employs a fast-slow load share layer to coordinate the execution of the above solution results. The slow layer runs once every 3 control cycles. Based on the expected net heat flux of the window in the predicted time domain, indoor illuminance requirements, glare constraints, and louver motion constraints, it determines the window load reduction share borne by the louvers from the control feasible domain and outputs the target louver angle. and its holding interval;

[0102] The fast layer runs once every control cycle and remains within the holding interval defined by the slow layer. Unchanged, and adjusted according to the actual indoor temperature. The room temperature prediction relative to the selected candidate combination in the same control period The deviation, and the actual air conditioning power Compared to predicted air conditioning power The deviation from the target value of air supply volume and the target temperature set by the air conditioner Closed-loop correction is performed, adjusting only within the range where the illuminance and glare constraints already met in the slow layer remain satisfied, and each adjustment does not exceed the control rate constraint. The control state is set to four states: normal, frozen, degraded, and recovery. In the normal state, sub-channel parameter updates and joint control solutions are performed. In the frozen state, updates to the corresponding window's thermal parameters are stopped, and the control constraints are tightened using the confidence boundary of the window's net heat flow prediction. In the degraded state, the window's net heat flow is predicted using the conservative effective solar heat gain coefficient and effective heat transfer coefficient from the product parameters, and the louver angle is maintained at the closest feasible angle that meets the preset lower limit of illuminance. The remaining heat zone load is borne by the air conditioning equipment. In the recovery state, the control cycle is followed. The proportion of periodic unconstraint tightening and parameter updates is restored;

[0103] When the determinant value of the information matrix of the corresponding residual channel in step S3 does not reach the preset threshold, or the corresponding residual exceeds the confidence interval determined based on the historical effective residual distribution, the system enters the frozen state and records the number of consecutive frozen periods.

[0104] When the sensor drift check fails, the joint control problem has no feasible solution, or the number of consecutive freeze cycles reaches the preset cycle threshold of 6, the state enters the degraded state. When any degraded condition is met, the state transition of the degraded state takes precedence over the freeze state and the recovery state.

[0105] When the sensor drift check passes, the determinant value of the corresponding information matrix continuously reaches the preset threshold, the number of cycles in which the corresponding residual is continuously within the confidence interval reaches the recovery threshold of 4, and the closed-loop verification meets the recovery conditions, the system enters the recovery state. After the constraint tightening is released and the parameter update is restored, the system returns to the normal state.

[0106] The Low-E membrane orientation information used is saved for each control cycle. 1. Index of louver angle range The document includes: window thermal state parameter set and its version, effective observation dataset, excitation state, residual decomposition results, window net heat flux prediction, joint control target value, control state, and execution feedback. The execution feedback includes the actual executed louver target angle. Air conditioning air volume target value Air conditioner set temperature target value The actual indoor temperature, actual supply air temperature, actual return air temperature, actual supply air volume, and actual air conditioning power are measured in one control cycle.

[0107] During closed-loop verification, the already executed... and Input the hot zone prediction model and air conditioning equipment prediction model from step S4 to obtain the predicted zone temperature change, predicted air conditioning heat exchange, and predicted air conditioning power for the next control cycle. The difference between the actual zone temperature change and the predicted zone temperature change, and the difference between the measured air conditioning heat exchange and the predicted air conditioning heat exchange obtained according to the method for determining the measured air conditioning heat exchange in step S2, are used to form the window total heat flow verification residual. Then, the first and second regression bases in step S2 are used to decompose the window total heat flow verification residual into the solar heat gain channel verification result and the temperature difference heat transfer channel verification result. At the same time, the difference between the actual air conditioning power and the predicted air conditioning power is determined as the equipment verification result.

[0108] When only one channel's verification result exceeds the verification interval determined based on the corresponding historical valid verification result for three consecutive control cycles, and the device's verification result is continuously within the corresponding verification interval, this is determined as the freezing condition.

[0109] When the sensor drift check fails, the joint control problem has no feasible solution, the verification results of both channels exceed the corresponding verification range for three consecutive control cycles, the verification results of the device and any channel exceed the corresponding verification range for three consecutive control cycles, or the number of consecutive freeze cycles reaches the preset cycle threshold of 6, this is determined as a downgrade condition.

[0110] When the number of consecutive cycles in which the verification results of the two channels and the device verification results are within the corresponding verification interval reaches the verification recovery threshold of 4, it is determined as the recovery condition for this item.

[0111] The aforementioned triggering conditions and the current state conditions for the four control states are jointly determined. When any degradation condition is met, the state is entered into degradation. When there is no degradation condition and any freezing condition is met, the state is entered into freezing. The state is entered into recovery only when there is no degradation condition or freezing condition and both the aforementioned recovery condition and the current recovery condition are met. In other cases, the current control state is maintained, and the execution feedback of the next control cycle is returned to steps S1 and S2. At the same time, the window thermal state parameter set is corrected according to the update conditions of step S3.

[0112] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0113] This invention obtains the basis for verifying the total heat flux of the window through thermal zone energy balance and measured heat exchange from air conditioning. It then utilizes a regression basis corresponding to changes in solar radiation and indoor / outdoor temperature difference to distribute the window's heat flux information to the solar heat gain channel and the temperature difference heat transfer channel, enabling the window's thermal parameters corresponding to different membrane orientations and louver angle ranges to be updated according to the actual excitation state. The resulting expected value and confidence boundary of the window's net heat flux can be used simultaneously for thermal zone prediction and control constraint tightening.

[0114] Compared to the independent technical structures of louver control and air conditioning control, this invention connects window thermal state estimation, candidate combination prediction, load share fast and slow layer coordination, and freeze, degrade, and recovery state machines into a closed-loop control structure. It also uses the zone temperature, air conditioning heat exchange, and power feedback of the next cycle to verify the prediction results, thereby coordinating window load and air conditioning load while meeting the boundaries of lighting, comfort, and equipment operation.

Claims

1. A method for controlling energy-saving air conditioning with hollow louvered glass, characterized in that, include: S1. Collect operational data and Low-E membrane orientation indicators of building thermal zones, glass windows with hollow louvers and Low-E membranes, and air conditioning equipment. Perform time alignment, data validity processing, and valid excitation judgment on the operational data to obtain a valid observation dataset and excitation status. S2. Based on the temperature change and measured heat exchange of the air conditioner in the effective observation data, construct the total heat flux residual of the window. According to the excitation relationship between the solar radiation change and the indoor and outdoor temperature difference change, decompose it into solar heat gain residual and temperature difference heat transfer residual. S3. According to the Low-E membrane orientation mark and the louver angle interval index window thermal parameter unit, only when the excitation state meets the identifiable condition of the corresponding residual channel, use the two types of residuals to update the effective solar heat gain coefficient, effective heat transfer coefficient, cavity heat storage state parameters and uncertainty. When the update conditions are not met, the corresponding parameters are maintained and their uncertainty is adjusted to obtain the set of thermal state parameters of the form. S4. Based on the window thermal state parameter set, future solar radiation and future outdoor temperature, evaluate candidate combinations of louver angle, air conditioning supply volume and air conditioning set temperature, and generate window net heat flow prediction, room temperature trajectory, indoor illuminance trajectory and air conditioning load trajectory with uncertainty under each candidate combination. S5. Using the window net heat flow prediction and its uncertainty and three types of trajectories as input, determine the control feasible region under the constraints of illuminance, glare, thermal comfort and air conditioning equipment capacity, jointly determine and execute the louver target angle, air conditioning air supply volume target value and air conditioning set temperature target value, return the execution feedback of the next control cycle to steps S1 and S2, and correct the window thermal state parameter set according to the update conditions of step S3.

2. The energy-saving air conditioning control method for hollow louvered glass according to claim 1, characterized in that, S1 includes: The operating data includes at least one of the following: indoor temperature, outdoor temperature, solar radiation, sun position, surface temperature of both sides of the glass and cavity temperature, window temperature, current angle of the louvers, indoor illuminance, supply air temperature, return air temperature, supply air volume, air conditioning power, and air conditioning set temperature. The running data is resampled and timestamp corrected based on the data acquisition time and preset control cycle; Perform drift checks based on the range boundaries and rate of change per unit time boundaries of the corresponding measurement channels in the sensor calibration table; Information matrices are constructed for the solar radiation variation sequence and the indoor-outdoor temperature difference variation sequence, respectively, and the excitation state is determined by whether the determinant value of each information matrix reaches the preset threshold of the corresponding channel.

3. The energy-saving air conditioning control method for hollow louvered glass according to claim 2, characterized in that, S2 include: Taking the heat flowing into the hot zone as positive, the change in energy storage in the hot zone is determined based on the equivalent heat capacity of the air and the building envelope in the hot zone and the temperature change in the zone. The measured heat exchange of the air conditioner with the flow direction symbol is determined based on the air supply volume, the specific heat capacity of the air, and the difference between the supply air temperature and the return air temperature. The heat flow at the non-window boundary is determined based on the heat transfer parameters of the non-window building envelope and the infiltration ventilation volume. The residual heat flux of the window is obtained by subtracting the internal heat gain, the measured heat exchange of the air conditioner, and the heat flux at the non-window boundary from the change in the energy storage of the hot zone. Using the solar radiation variation sequence as the first regression basis, the indoor and outdoor temperature difference variation sequence after deducting its projection on the first regression basis is used as the second regression basis. The projection components of the total heat flux residual of the window on the first and second regression bases are successively determined as the solar heat gain residual and the temperature difference heat transfer residual.

4. The energy-saving air conditioning control method for hollow louvered glass according to claim 3, characterized in that, S3 includes: The orientation category of the Low-E film surface located on the outer surface of the outdoor glass, the cavity side surface of the outdoor glass, the cavity side surface of the indoor glass, or the inner surface of the indoor glass is combined with the louver angle range divided according to the adjustable louver angle range to form an index. Each index corresponds to a window thermal parameter unit that stores the effective solar heat gain coefficient, effective heat transfer coefficient, cavity heat storage time constant, cavity heat storage state quantity, and the covariance of each parameter. The preset physical allowable range is determined by the intersection of the parameter range given by the applicable architectural glass standard and the parameter range given by the corresponding glass product parameter. When the intersection is empty, the inspection parameter range of the corresponding glass product is used.

5. The energy-saving air conditioning control method for hollow louvered glass according to claim 4, characterized in that, In step S3, the effective solar heat gain coefficient is updated for the solar heat gain residual channel, the effective heat transfer coefficient is updated for the temperature difference heat transfer residual channel, and the cavity heat storage time constant and cavity heat storage state quantity are updated according to the time-series decay components of the two residual channels. When the determinant value of the corresponding information matrix reaches a preset threshold and the residual is within the confidence interval determined based on the historical effective residual distribution, a recursive update is performed, the recursive update value is projected to the preset physical allowable interval, and the corresponding covariance is reduced. When the determinant value does not reach the preset threshold or the residual exceeds the confidence interval, the corresponding parameter value remains unchanged, and the corresponding covariance is increased according to the number of consecutive freeze cycles.

6. The energy-saving air conditioning control method for hollow louvered glass according to claim 5, characterized in that, S4 include: The future sun position is calculated based on the building's latitude and longitude, window orientation, and time of prediction within the time domain. The incident direction of sunlight relative to the window is determined based on the future sun position. For each candidate louver angle, the direct light blocking ratio and the predicted indoor illuminance value are calculated. Input the louver angle range to which the candidate louver angle belongs, the Low-E membrane orientation identifier, future solar radiation, and future outdoor temperature into the corresponding window thermal parameter unit to calculate the expected value of the net heat flux of the window in the prediction time domain and the confidence boundary. The expected net heat flux of the window and the confidence boundary are used as the boundary heat flux input of the hot zone prediction model. Combined with the candidate air supply volume and candidate set temperature of the air conditioner, the room temperature trajectory, indoor illuminance trajectory and air conditioner load trajectory corresponding to each candidate combination are obtained.

7. The energy-saving air conditioning control method for hollow louvered glass according to claim 6, characterized in that, S5 include: The feasible boundaries of the thermal comfort constraint and the air conditioning equipment capacity constraint are tightened based on the uncertainty of the net heat flow prediction of the window. The net heat flow of the window corresponding to the initial louver angle of the control cycle is used as the benchmark value. The difference between the benchmark value and the net heat flow of the window corresponding to the candidate angle is used as the window load reduction amount. The hot zone prediction model forms the hot zone prediction load based on the future environmental input, internal heat gain prediction and the benchmark value. The portion of the hot zone prediction load after deducting the window load reduction amount is used as the remaining load borne by the air conditioner. Using the weighted sum of air conditioning power consumption, louver operation frequency, set temperature change, and illuminance deviation in the predicted time domain as the control objective, within the feasible domain jointly defined by illuminance, glare, thermal comfort, air conditioning equipment capacity, louver holding time, and control rate constraints, the system solves for the target louver angle, air conditioning air volume, and air conditioning set temperature values ​​that satisfy the remaining load on the air conditioning load trajectory. The boundaries and weights of each constraint are given in advance by the building operation parameter table.

8. The energy-saving air conditioning control method for hollow louvered glass according to claim 7, characterized in that, Candidate solutions for the control objective are determined by coordinating the fast and slow load share layers. The slow layer determines the window load reduction share borne by the louvers according to the expected net heat flux of the window in the predicted time domain, the indoor illuminance requirement, and the louver action constraints, and determines the target angle and holding range of the louvers. Within the holding interval, the fast layer adjusts the target value of the air supply volume and the target value of the air conditioning set temperature based on the deviation of the actual zone temperature from the room temperature prediction value of the corresponding control cycle of the selected candidate combination, and the deviation of the actual air conditioning power from the predicted air conditioning power of the candidate combination in the same control cycle. The fast layer adjustment shall not change the illuminance constraint and glare constraint satisfaction status determined by the slow layer.

9. The energy-saving air conditioning control method for hollow louvered glass according to claim 8, characterized in that, Set four control states: Normal, Freeze, Degraded, and Restore; The excitation anomaly where the determinant value of the information matrix of the corresponding residual channel does not reach the threshold, or the residual anomaly where the residual exceeds the confidence interval determined based on the historical effective residual distribution, is determined as a freezing condition. When the freezing condition is met, the system enters a frozen state, stops updating the corresponding window thermal parameters, and tightens the control constraints using the confidence boundary of the window net heat flow prediction. The following conditions are defined as sensor drift check failure, joint control problem with no feasible solution, or the number of consecutive freeze cycles reaching a preset cycle threshold: when the degradation condition is met, the system enters the degradation state. The net heat flow of the window is predicted according to the conservative effective solar heat gain coefficient and effective heat transfer coefficient in the product parameters, and the louver angle is kept at the nearest feasible angle that meets the preset lower limit of illuminance. The remaining heat zone load is borne by the air conditioning equipment. When any of the degradation conditions are met, the state transition of the degradation state takes precedence over the freeze state and the recovery state. The recovery condition is defined as the number of cycles in which the drift check passes, the determinant value of the corresponding information matrix continuously reaches the threshold, and the residual is continuously within the confidence interval. When the recovery condition is met, the system enters the recovery state, releases and tightens constraints and restores parameter updates cycle by cycle according to the preset rate of change, and then returns to the normal state. The preset cycle threshold, recovery threshold, and preset change rate are given in advance in the building operation parameter table based on the control cycle length and the window thermal parameters to estimate the window length.

10. The energy-saving air conditioning control method for hollow louvered glass according to claim 9, characterized in that, In each control cycle, the following information is saved: Low-E membrane orientation identifier, louver angle interval index, window thermal state parameter set and its version, effective observation dataset, excitation state, residual decomposition results, window net heat flow prediction, joint control target value, control state, and execution feedback. The target louver angle, air conditioning air volume target value, and air conditioning set temperature target value are input into the hot zone prediction model and the air conditioning equipment prediction model to obtain the predicted zone temperature change, predicted air conditioning heat exchange and predicted air conditioning power for the next control cycle. The window total heat flow verification residual is formed based on the difference between the actual temperature change and the predicted temperature change, and the difference between the measured value obtained according to the method for determining the measured heat exchange of the air conditioner and the predicted heat exchange of the air conditioner. The window total heat flow verification residual is decomposed into solar heat gain channel verification result and temperature difference heat transfer channel verification result using the first regression basis and the second regression basis. The difference between the actual air conditioner power and the predicted air conditioner power is determined as the equipment verification result. The freezing condition for this item is determined when only one channel's verification result continuously exceeds the verification interval determined based on the corresponding historical valid verification result, and the device verification result is continuously within the corresponding verification interval. The degrading condition for this item is determined when the sensor drift check fails, the joint control problem has no feasible solution, the verification results of two channels continuously exceed the corresponding verification interval, the device verification result and any channel verification result continuously exceed the corresponding verification interval, or the number of consecutive freezing cycles reaches the preset cycle threshold. The recovery condition for this item is determined when the number of cycles in which the verification results of two channels and the device verification result are continuously within the corresponding verification interval reaches the verification recovery threshold. The verification recovery threshold is given in advance in the building operation parameter table according to the control cycle length and the closed-loop verification window length. The system enters a downgrade state when the aforementioned downgrade condition or this downgrade condition is met. If neither the aforementioned downgrade condition nor this downgrade condition is met, the system enters a freeze state when the aforementioned freeze condition or this freeze condition is met. The system enters a recovery state only when neither the aforementioned downgrade condition, this downgrade condition, the aforementioned freeze condition, nor this freeze condition is met, and both the aforementioned recovery condition and this recovery condition are met. In all other cases, the current control state is maintained.

Images