Development of dynamic shading control for daylight measures in green buildings towards overall energy efficiency of lighting and air-conditioning systems

The introduction of daylight can improve buildings’ energy efficiency and bring benefit to occupant satisfaction. However, the introduction of daylight may accompany with excessive heat. Properly counterbalancing the energy consumption of air conditioning and lighting systems owing to the entry of daylight is a critical control target of dynamic shading adjustment in cooling season. Most dynamic shading control strategies in use only consider one single system. Additionally, for advanced control mode like performance-based control, the predictive model usually only examines the instantaneous effect of energy performance to determine the shading adjustment state, unable to quantify the overall influence of shading adjustment state on building energy consumption. In order to address this issue, special consideration is given to calculating the cumulative contribution of heat gains to cooling load in this study. An overall energy-efficient shading control metric is proposed and used as basis to develop optimized dynamic shading control strategy. An application example demonstrates that the SGR-Optimal control strategy can further save energy by 21.8% ~ 38.8% when compared to the Rule-based control strategy, thus allowing a better exploration of the energy efficiency potential of daylight measure.


Introduction
Introducing natural light into buildings is an effective way to improve occupants' productivity [1,2], health state [3][4][5] and comfort levels including thermal comfort [6,7] and visual comfort [8][9][10].As one most basic natural element and renewable resource, daylight can serve as a free light source to supplement indoor illumination thus owing great potential in elevating energy efficiency for the design and development of zero carbon buildings.Early studies have already indicated that direct lighting with natural light can enhance the efficiency by 25 times compared with utilizing solar heat to generate power for indoor lighting [11].And more than 40% of overall energy consumption of indoor lighting and air-conditioning can be saved by simply linking lighting control with the obtained natural light from side windows [12].
The outcome of daylight measures is significantly relied on the coordination between the building and its surrounding luminous and thermal environments.With dynamic characteristics of outdoor environment, the implemented control strategy of dynamic envelope system has become the key to goal reaching in energy efficiency.As one of the most recognized daylight measures, dynamic shading systems can better adapt to outdoor environment by altering the transparent portions of building envelopes.Dynamic shading's control modes can be generally categorized as manual control and automatic control [13].Compared with manual control mode, automatic control mode has the following benefits: (1) A higher level of daylight illumination can be achieved while preventing overheating and glare [14][15][16], (2) The shading condition is automatically changed according to the link between the set-point and measured values of indoor temperature and illuminance [17], (3) Beam solar radiation is blocked through automate incident angle calculation and slat angle adjustment, (4) The energy consumption of lighting and cooling can both be cut simultaneously [14,[17][18][19].
Block beam solar control (BSC), Rule-based Control (RBC) and Performance-based control (PBC) are the three most conventional control systems used within automatic shading control mode.To prevent glare and block direct sunlight from entering the room, shading devices in BSC mode adjust in accordance with the position of the sun [20][21][22][23].For example, to maintain a cut-off angle, the slat angle of blind varies from time to time so as to make it always perpendicular to the beam solar radiation [20,22,23].RBC mode uses indoor environment variables as control object, including transmitted or incident direct radiation, incident total irradiance, glare or illuminance index, room air temperature, etc.The shading device will be activated and adjusted when the value of control object exceeds its threshold.In contrast to RBC and BSC, PBC frequently establishes a direct links between shading control and building energy performance.For various scenarios and performance expectation, PBC adjusts shading device based on performance model's prediction.According to performance indicators, activation and the degree of shading adjustment are determined in order to meet specific constraints (usually in terms of occupants' comfort) while minimizing control objectives (usually in terms of energy consumption).As a result, PBC mode can fully explore the potential in energy efficiency of daylight measures.
Excessive heat may accompany with the introduction of natural light, especially in cooling season.It's important to counterbalance the energy consumption of lighting and air conditioning systems in building envelopes' design and operation, especially the control of shading devices.As the amount of natural light and heat that enters a space may vary among different shading control state, one shading control state may bring more lighting saving while with more cooling demands in cooling season compared with other control states.As lighting and air conditioning systems are dominant users in buildings' energy consumption, better balancing the amount of introduced natural light and heat can have a significant impact on buildings' energy performance.The optimized control of dynamic shading adjustment has to properly handle the conflicts between the energy consumptions of lighting and air conditioning systems due to daylight introduction.From the energy efficiency point of view, the assessment of the overall energy performance of lighting and air conditioning systems is vital for daylight measures.However, many dynamic shading control strategies simply pay attention to the energy efficiency of one particular system.It is also worth noting that, varying with the physical properties of building materials, heat gain affects indoor cooling load gradually rather than instantaneously.
As the booming of intelligent control products recently, building systems such as shading devices are more and more integrated with the IoT and AI technologies.However, AI aided control algorithms often require complex simulation of coupling software or large amount of data for model training [10,24,25], which inevitably brings large burden of real-time calculation.And the involved data-driven model also can raise certain doubts about the rationality of the operation sometimes.An easy-to-use dynamic shading control strategy which are derived from physical models is still needed to maximize the daylight measures' potential for energy efficiency during operation stage.Additionally, most of currently used control strategies just investigate the instantaneous impact on energy performance.For shading control, it is important to account the cumulative contribution of heat gain to cooling load so as to identify the feasible action.
In this study, the impact of shading adjustment state on air conditioning and lighting energy consumptions was examined to fill in these gaps.Concepts of effective shading adjustment state and Save-Gain-Ratio (SGR) were defined as indicators to rate the cumulative and overall impact of real-time shading control status on building energy consumption.Based on SGR maximized, a workflow of PBC for dynamic shading control was established.The cumulative contribution of heat gains to cooling load has been considered during the calculation process.The effectiveness of the proposed control strategy was demonstrated through a case study.

Methodology
The flexibility of dynamic shading device allows the users to adjust shading states according to the energy efficiency of various shading states.Through the shadeglazing system, daylight can be introduced for lighting purposes, supplementing or even replacing the indoor electrical lighting daylight.There are three aspects that variations in shading states can affect the energy performance of building: i.The lighting system's energy consumption ii.The energy consumption of air-conditioning system due to lighting system's heat dissipation iii.The energy consumption of air-conditioning system due to energy flow through shading-glazing system, which can be written as: where Q is the instantaneous energy flow, U is overall coefficient of heat transfer, A is window area, T out is outdoor air temperature, T in is indoor air temperature, Q b,t,i and Q d,t,i are direct and diffuse solar heat gain, respectively.Each point in time, presuming that the air-conditioning system can keep the room temperature at the set-point level, the outdoor-indoor temperature difference is consequently identical in all shading adjustment states.Due to the little influence of the shading adjustment states on the U value of the shade-glazing system [26], the third term on the right-hand side of Eq. ( 1) is considered to be constant at the same time point for all shading adjustment states.Thus, the variation in Q that corresponding to different adjustment states is mainly caused by the solar heat gain that the shading-glazing system introduces.Among the three components mentioned above, the lighting system's heat dissipation and the solar heat gain introduced by the shade-glazing system are the contributors of indoor heat gains.There is a time lag before the (1) internal heat gain converting to the cooling load because the heat storage effect of building envelopes.In order to evaluate the overall energy performance of shading adjustment state, it is necessary to account for the cumulative contribution of these internal heat gains to the energy consumption of air-conditioning system.

Definition of effective shading state
The fully activated state of shading is taken as the base state at time point t.In the base case, all window area is shaded and no natural light is introduced.Among all the alternative shading states, base case owns the maximum lighting input power (P L,max ) and the minimum solar heat gain. For the base state at time t, ΔP L,t,i is the difference of input lighting power between the maximal value and the shading state i at time t, ΔCCL L,t,i is the difference of accumulative cooling load caused by lighting heat dissipation between the base state and the state i at time t.
The base case is used as a benchmark to derive the concept of effective shading state.For an effective shading state, the energy saved by introducing natural light for indoor lighting (E save,t,i ) must be larger than the energy gained by the accompanied solar heat gain (E gain,t,i ).More simply, the overall energy consumption of air-conditioning and lighting systems that correspond to an effective shading state should be lower than that of the base state.If ΔE Total,t,i < 0, shading adjustment state i can be assessed as an effective shading state at time point t.
To facilitate the assessment of effective shading state, a ratio of E save,t,i to E gain,t,i under the shade state i is defined as follows: At time t, SGR t,i of all optional shading adjustment states are screened and compared.When SGR t,i > 1, the shading state i can be seen as an effective shading state at time t.By selecting the effective shading state with maximum SGR t,i , the overall energy perfor- mance of daylight measures can be maximized through optimization.

Workflow of shading state optimization
The workflow of shading state optimization follows a simple two-steps process: Step 1 Collecting weather prediction data at time t and calculating SGR t,i ( i = 1, . . ., n, base state is excluded) of n optional shading states.
Step 2 Selecting the maximal SGR t,i at time t (SGR t,max ).If SGR t,max < 1, there is no effective shading state for the shading state at time t, and the base state is taken as the best possible solution of shading state at time t.Then shading is activated and set to fully cover the window area.If SGR t,max ≥ 1, the optimal shading state at time t is recognized to be the shading state that corresponds to the largest SGR value.When SGR t,max = 1, even though the shading state that corresponds to the maximum SGR value shows no further benefit of energy consumption than the base state, it continues to be recognized as the feasible shading adjustment state at time t since it is capable of providing access to the outside view.
It is worth noting that E save,t,i has an upper limit.When the level of daylight illuminance on the working plane reaches or beyond the target value E set (e.g.500 lx), the (9 lighting system would be completely switched off.In this situation, the input power of lighting system is zero, and the energy saved by daylight supplemented lighting reaches its maximum value.The upper limit of E save,t,i is equal to the maximum lighting input power, plus the cumulative contribution of reduced heat dissipation to air conditioning system.Sections 2.3 and 2.4 present the details in thermal and daylight modules as required in the workflow.

Daylight module calculation
Daylight module calculation are conducted in four parts so as to get the difference between the maximal value and the shading state i of input lighting power (ΔP L,t,i ),

Incident illuminance
For time t, the total incident illuminance (E t ) on the vertical building façade is determined by adding the sky diffuse illuminance (E ds,t ), direct illuminance (E b,t ) and ground reflected diffuse illuminance (E dg,t ).The illuminance parameters at time t can be derived from the predicted values of the outdoor irradiance parameters according to the Perez luminous models [27].
where τ vis,b (t,i) and τ vis,d (t,i) are the visible transmittance of the shading-glazing system for direct and diffuse illuminance, respectively.Their values vary with the incident angle as well as the shading state and are obtained using the ASHWAT model [28].The model regards the shading layer as an equivalent layer and calculates layer by layer to obtain the overall optical properties of the shadingglazing system.

Illuminance at daylighting reference point
A simplified method, namely Luminous Exitance Method, is used to calculate the distribution of indoor daylight illuminance [29].The prediction method implemented the radiosity theory and calculated the illuminance of a surface (E) with matrix of the initial luminous exitance (M o ) and the illuminance transfer factor ( T ′ ) as in Eq. ( 13): (10) The amount of diffuse and direct daylight received by each surface is calculated respectively.The final luminous value of each surface is calculated after adding the results.Daylight illuminance at the daylight reference point (E A,t,i ) is derived with: where C A-p represents the configuration factor between the daylight reference point and the interior surface p and is calculated with the method in Murdoch's study [30].

Lighting Input Power
If artificial lighting system permits a complete replacement, the lighting input power of shading state i at time t (P L,t,i ) is as follows: Using base state as the benchmark, the saved lighting input power △P L,t,i can be calculated as:

Thermal module calculation
Three steps are used to compute the energy consumption of air conditioning system during cooling season due to the cumulative contribution of heat gain.

Incident solar radiation
Similar as illuminance calculation, the total incident solar radiation on the vertical building façade at time t is calculated by adding direct radiation, sky diffuse and ground reflected diffuse radiation according to the Perez model [31].

Heat Gain
This part consists of the heat dissipation of lighting system and the solar heat gain.Since lighting input power will eventually turn to heat, at time t, when shading is set to state i, the amount of heat gained from indoor lighting system's heat dissipation (Q L,t,i ) equals to the value of P L,t,i .
For shading state i at time t, total solar heat gain (Q f,t,i ) is calculated with the simplified approach proposed by Wright et al. [28].In Wright et al. 's method, simplified (13) Eset , E A,t,i < E set P L,max , E A,t,i ≥ E set coefficients IAC were proposed using ASHWAT model to facilitate calculation [28,32,33].The SHGC of the glazing center with and without a shading device is estimated using simulations, and the IAC coefficients for direct and diffuse radiation were generated respectively to represent how shade reduces heat gain.Approximate values of direct solar heat gain (Q b,t,i ) and diffuse solar heat gain (Q d,t,i ) are then resulted.

Cumulative contribution of heat gains to the energy consumption of cooling system
Between indoor heat gain and cooling load, there is a time lag.Based on the idea of the Radiant Time Factors (RTF) and Radiant Time Series method (RTS method), the cumulative contribution of heat gains to indoor cooling load is taken into considerations [34].
Within RTS method, all types of heat gains are divided into convective and radiant portions.The cooling load at a given hour contains the sum of the hour's convective portion and the radiant portion from time-delayed heat gains of current hour and previous twenty-three hours.While convective portion converts to indoor cooling load instantaneously, radiant portion transforms to indoor cooling load for the current and following 23 h and can be counted with RTF which follows Eq. ( 17).
Room size and the building material's physical properties affect the assignment of RTF.Solar RTF and Non-solar RTF are the two different categories of radiant time series.Solar RTF is suitable for direct solar radiation entering the room and assumes that all radiation falls on the floor.Non-solar RTF is suitable for diffuse solar radiation entering the room and assumes that radiation is evenly distributed within the space among interior surfaces.
The lighting heat dissipation and solar heat gain are split into a convective portion and a radiant portion, according to the corresponding radiant ratio (F r,L , F r,SHG ) and convective ratio (1-F r,L , 1-F r,SHG ), to calculate cumulative cooling load through RTS method.The convective portion converts to instantaneous cooling load directly.For the radiant portion, a cumulative ratio CR t is defined as Eq. ( 18) to determine the overall contribution of indoor heat gains to cooling load.If cooling system operation begins at hour t on of the day and stops at the end of hour t off , radiant heat gain at hour t ( t on ≦ t ≦ t off ) cumulatively contribute to cooling load with a CR t as shown in Fig. 1. (17) When the shading is changed to state i at time t, for direct solar heat gain: If the incident direct radiation is still in direct form after transmitting: For some certain shading types, incident direct radiation converts to diffuse form after transmitting.In that case, CR t that generated from the Non-solar RTF is feasible as: For diffuse solar heat gain: For heat dissipation from indoor lighting system: (18) where CCL SHGb,t,i and CCL SHGd,t,i are the accumulative cooling load caused by direct and diffuse solar heat gain under state i at time t, CR non-solar,t and CR solar,t are the values of CR t that generated from the Non-solar RTF and the solar RTF.Detailed nomenclature can be found in the 6..
Figure 2 shows the workflow of SGR-Optimal.A case study is conducted in the following session to investigate the outcome of proposed dynamic shading control strategy in energy efficiency.

Application example
A model of office building is established in EnergyPlus, and both conventional RBC strategies and SGR-Optimal control strategy are applied.Building energy consumption values of under various shading control strategies are compared.

Model setup
In EnergyPlus, a medium-sized office with a long strip shape has been established as shown in Fig. 3.The wall with glazing is assigned as an exterior wall, while the other three walls are assigned as interior walls.The exterior wall's window-to-wall ratio (WWR) is 0.5.Simulation simulations with external walls oriented to the east, west and south are employed by rotating the building model (as illustrated in Fig. 4).
The window has an external roller shade equipped.Tables 1, 2 and 3 contain information about building materials and shading equipment.Due to the limitations in EnergyPlus function, external shade can only set as fully open or fully closed.In simulation configuration Fig. 1 Cumulative ratio of the radiant portion heat gains to cooling load of EnergyPlus, exterior window is divided vertically into four pieces and the external shade of each window component can be individually regulated.Four shade pieces can be combined in various ways to achieve shading adjustment states with pull-down degree of 0%, 25%, 50%, 75%, and 100%.
A daylight reference point is positioned as depicted in Fig. 3. 500 lx is the illuminance target at the daylight reference point.Indoor artificial lighting system is completely turned off when the daylight illuminance value at the daylight reference point reaches or exceeds 500 lx.Otherwise a continuous linear dimming control related to daylight will be applied to meet the goal value at the reference point.The typical meteorological year of Hong Kong is applied as the weather input for the simulation.
To mitigate the impact of HVAC system selections within EnergyPlus, the cooling load is calculated using ZoneHVAC: IdealLoadsAirSystem module.As the building is situated in the cooling-dominant subtropical city of Hong Kong, only the cooling mode of air conditioning is taken into consideration.The temperature for indoor cooling is set at 25 °C.During operation period, the air conditioning system will begin cooling when indoor air temperature is higher than the set-point.Table 4 contains a list of the building model's operational parameters.
Run-period of simulation starts from January 1st to December 31st.Simulation cases with various shading control strategies include:

i. Static case group
Static-I: Throughout the year, the shade is always retracted and 0% of window area is covered with shade.
Static-II: Throughout the year, the shade is always fully pulled down and 100% of window area is covered with shade.
ii. Rule-based Group (RBC) RBC-I: The shade will be pulled down to cover 100% of the window area when incident solar radiation intensity exceeds 100 W/m 2 .Otherwise, the shade will remain retracted and provide no shading effect.iii.SGR-Optimal group SGR-Optimal control strategy is used to regulate shade adjustment.Hourly values of meteorological parameters is extracted from the EPW input weather data file and used as weather forecast value.Figure 5 shows the base case and the four alternative adjustment states of shading.SGR-Optimal control strategy is used to generate a schedule for hourly-based optimal shade adjustment.The optimized schedule is generated with R language programming.Except the building operation time, the shade state is set to a 100% pull-down state.The generated schedule is then applied in the shading state control of EnergyPlus.
Among the studied cases, the Static case group stands for the extreme conditions which are mainly used as reference for comparison and unlike to happen in the real operation.And the Rule-based group presents the conventional shading control mode and the shading is triggered once the threshold value is reached.The SGR-Optimal group implemented the proposed control strategy and its energy performance is compared with the Static case group and Rule-based group to assess the outcome.
Simulation outputs of annual cooling load and lighting power consumption are gathered.Annual air conditioning system's energy consumption is converted from annual cooling load with an overall COP of 3. The overall energy consumption of lighting and air conditioning systems is then added.

Parameters calculation
Five sets of view factors are calculated, each corresponding to a different pull-down degrees.Indoor space is separated into seven sub-surfaces for shading states with pull-down degrees of 0% and 100%: the wall surrounding the window, the window, the ceiling, the back wall, the right-side wall, the floor and the left-side wall, with reflectance values of 0.7, 0.1, 0.8, 0.7, 0.7, 0.3 and 0.7, respectively.Indoor space is separated into eight sub-surfaces for the other three shading states with pull-down degrees of 25%, 50%, and 75%: the floor, the wall surrounding the window, the portion of the window covered by shading,   the portion of the window not covered by shading, the back wall, the ceiling, the right-side wall and the left-side wall, with reflectance values of 0.3, 0.7, 0.7, 0.1, 0.7, 0.8, 0.7 and 0.7, respectively.Using Ip Seng Iu's PRF/RTS Generator software, the time series of RTF required in SGR-Optimal workflow is determined according to building dimensions and material information.Table 5 displays the calculation outcomes.On weekdays, HVAC system operates from 8 am to 6 pm, with t on = 8, t off = 17.Table 6 lists the time series for the cumulative ratio CR t .
When the shade is set to base case, window area is completely covered and the transmitted radiation is all in diffuse form.In that scenario, the beam radiation ratio is zero while Eqs.( 20) and ( 21) are feasible.
When the shade is in state i, partial area of the window area is covered by shade while the rest area is still exposed.The transmitted solar radiation via the covered part is in diffuse form, which is applicable to CR non-solar,t .The direct radiation transmitted via the exposed portion of exterior window is still in direct form, and applicable to CR solar,t .
Software WINDOW is used to calculate the SHGC and τ vis of the multi-layer shading-glazing system at discrete incident angles as shown in Table 7 [35].According to the incident angle, the values of SHGC and τ vis at time t are determined by linear interpolation.IAC for direct and diffuse radiation both have values of 0.6.Radiation ratio of solar heat F r,SHG is equal to 0.46, and the radiation ratio of lighting F r,L is equal to 0.48 [26].

Result
Figure 6 depicts the overview of energy consumption results under various control strategies.It reveals that the energy performance of lighting and air conditioning systems varies among all the five control strategies.The level of daylight replacement directly affects the energy consumption.Besides the amount of solar heat gain introduced via the shading-glazing system, indoor heat gain from lighting system's heat dissipation also influences the energy consumption of air conditioning system.The overall level of energy consumption represents the total energy performance of both lighting and air-conditioning systems.Table 8 lists the annual energy consumption values of lighting and air conditioning systems for the compared shading control strategies per area.
The Static case group maintains a constant shading state and represents the extreme shading operation.Throughout the year, Static-I's shading device is inactive.In Static-I, both the solar heat obtained via the shadingglazing system and the introduced daylight for indoor supplemented lighting are at their highest levels compared to the other four cases.In Static-II, the shading is kept activated all of the year and completely covered the window areas.Due to the visible transmittance of the chosen shading material is zero and Static-II's lack of daylight access, the solar heat gain through the shadingglazing system is at its lowest.
Static-II and Static-I consume the most and the least energy, respectively, among the simulated cases with the same orientation.Taking Static-II as a comparison, It is worth noted that while the utilization of shading systems on building fenestrations has long been known as an effective strategy to minimize façade's heat gain and result in lower cooling energy consumption, in Fig. 6, with the application of the shading system and the automatic control strategy of RBC-I and RBC-II, their overall energy consumption is higher than that of Static-I, where the shading device is inactive.The reason is that while the utilization of shading devices of RBC-I and RBC-II groups can lead to less heat entering the space, their lighting system's energy consumption is higher than that of the Static-I.In the Static-I, as the shading is inactive, the entering natural light itself can fulfill the lighting demand most of the time instead of turning on lighting system and the lighting system's energy consumption is the minimum.Therefore when it comes to compare the overall energy consumption of lighting and air-conditioning systems among the RBC-I, RBC-II and Static-I groups, the energy consumption of the former two groups (RBC-I and RBC-II) are higher than the Static-I group.This result also indicates the importance of balancing the energy consumption of air conditioning and lighting systems in shading control.
Compared with the static control of constant shading state in the Static case group, RBC-I, RBC-II and SGR-Optimal groups apply automatic control strategy.RBC-I and RBC-II adopt rule-based control strategies, and the thermal threshold is selected as the control target, while SGR-Optimal is a performance-based control strategy and SGR is used as the control target to determine the pull-down degree of shading.The lighting system's energy consumption values of these three cases are between the values of Static-I and Static-II, and decrease in the order of RBC-I, RBC-II, and SGR-Optimal.Compared with Static-I, the cooling energy consumption values of these three cases are significantly reduced, and close to Static-II.The energy saving percentages of lighting and air conditioning systems are listed in Table 9.
The three automatic control strategies' deviations in total energy consumption from the Static case group are shown in Table 10 below.From Fig. 6, it can be seen that Static-II consumes the most overall energy for each orientation.Even if the Rule-based groups' (RBC-I, RBC-II) overall energy consumptions are lower than those of Fig. 6 The annual energy consumption (kWh) for lighting and air conditioning under various control strategies

Conclusion
In order to solve the contradiction between lighting and air conditioning systems' energy consumption in the cooling season, the overall influence of shading adjustment state on building energy consumption is quantitatively analyzed.Using the fully activated state of shading as benchmark, effective shading states at the specific point in time is identified.
To assess the overall impact of real-time shading control state on building energy consumption, SGR overall energy-efficient shading control metric is proposed.Operation time when cooling system stops.r 0 ,r 1 …r 23 Series of Radiant Time Factors (RTF).

CR t
Cumulative ratio for the radiant portion.CR non-solar,t CR t that generated from the Non-solar RTF.CR solar,t CR t that generated from the solar RTF.

Fig. 2
Fig. 2 Workflow of shading state optimization

Fig. 5
Fig. 5 Base case and the four optional adjustment states of shading

Table 1
Details of building material

Table 2
Parameters of double glazing element

Table 3
Roller shade's optical and thermal properties

Table 4
Operation parameters of the building model

Table 5
Radiant time series factor

Table 6
Time series of CR t

Table 7
WINDOW calculation output report for the shading-glazing system

Table 8
Annual energy consumption of lighting and air conditioning systems per area (kWh/m 2 )

I Static-II RBC-I RBC-II SGR-Optimal
And a dynamic shading optimized control strategy has been suggested to use based on SGR performance.Case Accumulative cooling energy consumption caused by lighting heat dissipation and solar heat gain under state i at time t (W) E Total,t,0 Sum of lighting energy consumption accumulative cooling energy consumption under base state at time t (W) E Total,t,i Sum of lighting energy consumption accumulative cooling energy consumption under state i at time t (W) P L,max Maximum input power of lighting system (W).P L,t,i Input power of lighting system under state i at time t (W).ΔP L,t,i Difference of input lighting power between the maximal value and the shading state i at time t (W) COP Coefficient of performance.IAC Indoor solar attenuation coefficient.CCL L,t,0 Accumulative cooling load caused by lighting heat dissipation under base state at time t (W) CCL L,t,i Accumulative cooling load caused by lighting heat dissipation under state i at time t (W) CCL SHG,t,0 Accumulative cooling load caused by solar heat gain under base state at time t (W) CCL SHG,t,i Accumulative cooling load caused by solar heat gain under state i at time t (W) CCL SHGb,t,i Accumulative cooling load caused by direct solar heat gain under state i at time t (W) CCL SHGd,t,i Accumulative cooling load caused by diffuse solar heat gain under state i at time t (W) ΔCCL L,t,i Difference of accumulative cooling load caused by lighting heat dissipation between the shading state i and the state i at time t (W) ΔCCL SHG,t,I Difference of accumulative cooling load caused by solar heat gain between the shading state i and the base state at time t (W) E save,t,i Energy consumption saved by introducing daylight-assisted illumination under state i at time t (W) E gain,t,i Energy consumption gained by the accompanying solar radiation under state i at time t (W) ΔE Total,t,i Difference of energy performance between shading adjustment state i and the base state (W) SGR t,i Save-Gain-Ratio under state i at time t.SGR t,max Maximum Save-Gain-Ratio under all optional states at time t.E t Total incident illuminance at time t (lux).E ds,t Incident sky diffuse illuminance at time t (lux).Incident direct illuminance at time t (lux).E dg,t Incident ground reflected diffuse illuminance at time t (lux).E d,t Incident diffuse illuminance at time t (lux).E bt,t,i Transmitted beam illuminance under state i at time t (lux).E dt,t,i Transmitted beam illuminance under state i at time t (lux).τ vis,b (t,i) Beam visible transmittance under state i at time t (lux).τ vis,d (t,i) Diffuse visible transmittance under state i at time t (lux).[E] N Illuminance matrix of N surfaces.[T'] (N×N) Illuminance transfer factor matrix among N surfaces.[M O ] N Initial luminous exitance matrix of N surfaces.E A,t,i Daylight illuminance at the daylight reference point under state i at time t (lux) C A-p Configuration factor between the daylight reference point and the interior surface N Interior surface.E set Target illuminance value (lux).F r,L Radiant ratio of lighting heat dissipation.F r,SHG Radiant ratio of solar heat gain.t on Operation time when cooling system begins.t off