Illumination radiation heat load dynamic prediction and compensation control method and system based on multispectral perception

Abstract

The invention provides an illumination radiation heat load dynamic prediction and compensation control method and system based on multispectral perception, and the method comprises the steps: collecting the illumination intensity of each indoor wave band in real time through a multispectral sensor, constructing a spectrum space coupling mapping model through combining the geometric structure, material absorptivity and incident angle information of a heat absorption surface, and carrying out the dynamic prediction and compensation of the illumination radiation heat load. Calculating an equivalent radiation heat disturbance quantity; thermodynamic parameters are fused to establish a thermodynamic prediction model, and the future room temperature change trend is predicted; identifying whether the disturbance type is a stationary type or a mutation type by analyzing the dynamic relationship between the thermal disturbance sequence and the temperature response; and a heat pump compensation control instruction is generated, and advanced adjustment is achieved. The problems that a traditional temperature control system is delayed in response to illumination thermal disturbance and inaccurate in recognition are solved, and the thermal comfort and the energy efficiency level are remarkably improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of heat pump control, and particularly relates to a light radiation heat load dynamic prediction and compensation control method and system based on multi-spectrum perception. BACKGROUND

[0002] With the improvement of building energy-saving standards and the increasing demand of users for living comfort, heat pump systems, as indoor temperature control means with high energy efficiency and stability, have become the most common environmental regulation equipment in residential, office and public buildings. Traditional heat pump systems mainly rely on real-time measurement results of temperature sensors for feedback control, such as common ON / OFF control, proportional-integral-derivative (PID) control and part of model control based on simplified prediction algorithm. Although such control methods can maintain good temperature balance under general steady-state conditions, their performance is significantly reduced in scenes with significant changes in light. Indoor temperature is not only determined by air heat exchange process, but also by the radiation heat effect of light, which accounts for an important proportion in indoor heat balance. In particular, in the case of direct sunlight, curtain blocking or sudden light, the uneven change of light spectrum composition will cause instantaneous heat disturbance. Traditional temperature control systems are limited by the hysteresis of single-point temperature measurement and cannot timely reflect the temperature rise or drop caused by the conversion of different wavelengths of light energy. Therefore, the phenomenon of "temperature has not changed but heat pump has over-reacted" or "temperature has risen but control signal has not been adjusted" often occurs, resulting in system oscillation, decreased comfort and increased energy consumption. Some improvement schemes in the prior art attempt to introduce light information through an illuminance sensor or an infrared photosensitive element, but due to the lack of spectral resolution capability, they cannot distinguish the heat effect of different light sources or different wavebands, making it difficult to distinguish between strong cold source and weak heat source in the signal, and still cannot effectively correct the heat pump operation strategy. In addition, most of the existing methods are post-response type control, which lacks a predictive compensation mechanism for heat disturbance caused by sudden light changes, and it is difficult to maintain the continuity of heat balance and comfort in complex dynamic light environments. Therefore, in the multi-spectrum and non-uniform radiation scene, the traditional temperature control method generally has the problems of insufficient spectral recognition ability, heat disturbance response lag and poor energy consumption optimization effect. SUMMARY

[0003] The purpose of the present application is to design a light radiation heat load dynamic prediction and compensation control method and system based on multi-spectrum perception, which can improve the prediction accuracy and response speed of the heat pump system to the change of light heat load, reduce energy consumption fluctuations and improve indoor thermal comfort.

[0004] To achieve the above purpose, in the first aspect of the present application, a light radiation heat load dynamic prediction and compensation control method based on multi-spectrum perception is provided, which comprises: Real-time collection of light intensity data of multiple wave bands through a multispectral sensor deployed indoors; construction of a spectral space coupling mapping model combining the geometric structure, material absorption rate, and light incidence angle information of each heat-absorbing surface in the room, calculation of the equivalent radiative heat disturbance caused by light at the current time; Taking the equivalent radiative heat disturbance as a thermal disturbance input, combining the current indoor and outdoor temperatures and the room thermodynamic parameters, constructing a thermal dynamics prediction model to predict the indoor temperature variation trend in a future preset time window; Based on the sequence of equivalent radiative heat disturbances in a continuous time period and the corresponding predicted temperature responses, identifying the current heat disturbance type as stable or sudden; According to the predicted temperature variation trend and the disturbance type, dynamically generating a compensation control instruction for the heat pump terminal, and driving the heat pump to perform corresponding adjustment actions.

[0005] Further, the multispectral sensor covers the visible light to near-infrared wave band, the sampling frequency is not less than once per second, and the spectral power density data of each wave band is output.

[0006] Further, the heat-absorbing surface includes curtains, walls, floors, or furniture surfaces, the geometric structure and orientation of which are obtained through spatial three-dimensional modeling during the deployment stage, and the material absorption rate is pre-set according to the standard material spectrum database.

[0007] Further, the calculation of the equivalent radiative heat disturbance introduces an incidence angle correction based on Lambert's cosine law to reflect the influence of different illumination angles on the surface heat absorption efficiency.

[0008] Further, the thermal dynamics prediction model introduces a spectral concentration regularization term based on the basic heat balance equation to suppress the temperature prediction overshoot caused by the concentration of high-energy wave bands.

[0009] Further, the disturbance type identification is based on the ratio of the heat disturbance change rate to the predicted temperature response rate, and combines the spectral energy distribution smoothness index for noise suppression and disturbance effectiveness discrimination.

[0010] Further, the disturbance type identification adopts an adaptive dynamic threshold mechanism, which automatically adjusts the threshold according to the statistical characteristics of the disturbance response index in the historical operation period.

[0011] Further, the compensation control instruction adopts a double-channel control strategy: under stable disturbance, the heat pump output is adjusted linearly according to the predicted deviation; under sudden disturbance, an enhanced gain mechanism is activated to improve the adjustment amplitude and response speed.

[0012] Further, the heat pump terminal adjustment includes at least one of compressor load, fan speed, or water valve opening, and the control instruction is transmitted to the execution unit through a standard building automation protocol.

[0013] In a second aspect of the present application, a multi-spectral perception-based illumination radiant heat load dynamic prediction and compensation control system is provided, comprising: a thermal disturbance modeling module, configured to collect illumination intensity data of multiple wave bands in real time through a multi-spectral sensor deployed indoors; construct a spectral space coupling mapping model by combining the geometric structure, material absorption rate and light incidence angle information of each heat-absorbing surface in the room, and calculate the equivalent radiant heat disturbance amount caused by illumination at the current time; a temperature prediction module, configured to take the equivalent radiant heat disturbance amount as a thermal disturbance input, fuse the current indoor and outdoor temperatures and the room thermodynamic parameters, construct a thermal dynamics prediction model, and predict the indoor temperature variation trend in a future preset time window; a disturbance identification module, configured to identify the current thermal disturbance type as a steady type or a sudden change type based on the sequence of equivalent radiant heat disturbance amounts in a continuous time period and the corresponding predicted temperature responses; a control execution module, configured to dynamically generate a compensation control instruction for the heat pump terminal according to the predicted temperature variation trend and the disturbance type, and drive the heat pump to perform corresponding adjustment actions.

[0014] The present application has at least the following beneficial technical effects: To solve the above problems, the present application provides a multi-spectral perception-based illumination radiant heat load dynamic prediction and compensation control method and system, which collects different wave band illumination intensity data in real time through a multi-spectral sensor, combines the geometric structure, material absorption rate and incidence angle information of the main heat-absorbing surfaces in the room, establishes a spectral space coupling mapping model, accurately calculates the equivalent radiant heat disturbance amount at the current time, and thus realizes quantitative evaluation of thermal energy at the spectral level. On this basis, a thermal dynamics prediction model that fuses the illumination disturbance and heat exchange characteristics is constructed, the heat capacity, heat transfer coefficient and spectral concentration regular term are used to predict the temperature variation in the future time window, and the stability and interpretability of the model under non-uniform radiation are significantly improved. To realize dynamic response, the present application further introduces a disturbance type identification module, analyzes the time fluctuation characteristics of the multi-spectral thermal disturbance sequence and the rate ratio of the predicted temperature response, judges whether the current disturbance is of a steady type or a sudden change type, and combines the spectral smoothing regular term to reduce noise interference. The control strategy part generates a heat pump control amount based on the predicted temperature and the disturbance type, adopts a double-channel control mechanism, executes gentle energy-saving adjustment under stable disturbance, triggers a rapid compensation mode under sudden disturbance, and realizes advanced temperature control response. The present application realizes a whole-process closed-loop control from multi-spectral thermal disturbance perception, dynamic thermal modeling, disturbance identification to terminal execution, can significantly improve the prediction accuracy and response speed of the heat pump system to illumination heat load changes, reduce energy consumption fluctuations, and improve indoor thermal comfort. BRIEF DESCRIPTION OF DRAWINGS

[0015] The application is further illustrated by the accompanying drawings, but the embodiments in the drawings do not constitute any limitation to the application, and other embodiments can be obtained by those skilled in the art without creative labor on the basis of the following drawings.

[0016] Figure 1 Flow chart of the illumination radiant heat load dynamic prediction and compensation control method based on multi-spectrum perception of the application.

[0017] Figure 2 Framework diagram of the illumination radiant heat load dynamic prediction and compensation control system based on multi-spectrum perception of the application. DETAILED DESCRIPTION

[0018] The embodiments of the application are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference signs represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary and are only used to explain the application, and cannot be understood as a limitation to the application.

[0019] In one or more embodiments, as shown in Figure 1 a method for illumination radiant heat load dynamic prediction and compensation control based on multi-spectrum perception is disclosed, which comprises the following steps: S1: Real-time acquisition of illumination intensity data of multiple wavebands by a multi-spectrum sensor deployed indoors; construction of a spectral space coupling mapping model combining the geometric structure, material absorption rate and light incidence angle information of each heat-absorbing surface in the room, and calculation of the equivalent radiant heat disturbance caused by illumination at the current time; Specifically, the purpose of this step is to obtain the actual radiant heat disturbance caused by multi-waveband spectrum acting on different heat-absorbing surfaces under the current indoor illumination condition. This heat disturbance value is not only affected by the intensity and wavelength distribution of the light source, but also closely related to the spatial position of the light irradiation, the incidence angle and the heat absorption capacity of the surface material of the irradiated object. Therefore, the method of only relying on "illumination intensity" or "empirical heat weight" cannot accurately reflect the real heat effect. This step realizes high-precision estimation of radiant heat load by introducing a structure absorption mapping method without adding additional sensors.

[0020] Firstly, a multi-spectrum sensor (such as a waveband detection array based on CMOS, commonly used model is AS7265x or IMX290 type with filter structure) installed indoors is used to real-time collect the power density of the current indoor illumination at each waveband at a frequency of once per second. The system presets to divide the spectrum into wavebands, covering from the visible light region (400-700 nm) to the near-infrared region (700-1100 nm), wherein each corresponding to a central wavelength, such as nm, nm, up to nm. For example, the light intensity measured by the 5th band (860 nm) at a certain moment is W / m².

[0021] Secondly, the parameters of the main heat-absorbing surfaces in the room need to be determined. The system uses a laser scanning ranging device and a space reconstruction module to construct a three-dimensional space model of the room at the initial deployment stage, and calibrates the heat-absorbing related surfaces (such as curtains, walls, desks, floors, etc.), and assigns a unique identifier to each surface, a total of types. For each , the system records its area , the orientation angle (the angle with the sensor coordinate system) and the absorption rate of the surface to different wave bands . These absorption rate data are set by consulting public material spectral library (such as ASTMG173 or NREL database). For example, the absorption rate of the curtain surface in the near-infrared wave band may be , while the absorption rate of the white-painted wall surface in the visible light wave band nm may be only .

[0022] In order to further consider the influence of the light incidence angle on the heat absorption efficiency, the system calculates the approximate incidence angle of each wave band to each surface based on the preset main light source orientation (such as the window orientation is southwest) combined with the current moment sun azimuth angle (obtained by reversing the geographical position and time) and the normal angle of each surface. The Lambert cosine law is used for incidence angle energy correction, thereby forming the wave band-surface level real heat absorption power estimation item.

[0023] Finally, the system uses the following formula to perform superposition calculation on all wave bands and all heat-absorbing surfaces, to obtain the current equivalent radiant heat load: ; wherein represents the total amount of equivalent radiant heat disturbance generated by the current indoor lighting conditions at time point . is the light intensity of the th wave band measured by the multi-spectral sensor; is the absorption rate of the surface to this wave band, determined by material preset; is the angle of the wave band light incident to the surface , calculated by a geometric model; is the area of the surface , calibrated in the deployment stage.

[0024] For example, in a typical afternoon scenario, when western sunlight shines into a room through a glass window, the sensor detects a significant increase in near-infrared intensity, with the angle of incidence approaching perpendicular. At this time, the heat absorption term on the curtain surface is greatly amplified in the aforementioned model, causing the system to calculate... The value was significantly higher than that of the same period on a cloudy day (which may only be 35), thus providing key driving information for subsequent heat forecasting and compensation adjustment.

[0025] S2: Using the equivalent radiative heat disturbance as the thermal disturbance input, and integrating the current indoor and outdoor temperatures and room thermodynamic parameters, a thermodynamic prediction model is constructed to predict the indoor temperature change trend within a preset time window in the future. Specifically, the goal of this step is to obtain estimates of multispectral thermal perturbations. Based on this, the evolution trend of indoor temperature in the near future is predicted. This predictive capability is one of the key foundations of the "dynamic compensation control of solar radiation heat load" in this scheme, because it enables the control system to take pre-emptive adjustment measures based on the intensity and direction of solar radiation disturbances before room temperature changes occur, thereby effectively overcoming the problem of response lag in traditional temperature control systems. Unlike traditional thermal models, this step not only retains the interpretable structure of thermal equilibrium energy conservation, but also incorporates the multispectral dynamic characteristics from solar radiation disturbances and introduces an innovative "spectral concentration regularization term" to penalize the risk of nonlinear thermal response caused by high-energy band concentration, thereby improving the stability and adaptability of the prediction model.

[0026] The input for this step includes the equivalent radiative heat load output from the previous step. And the current indoor temperature and outdoor temperature . The temperature is collected in real time by a high-precision thermistor temperature sensor located at the center of the control area, with an accuracy of ±0.1℃ and a sampling period of 1 second; The temperature sensor is located on the wall and can be accessed via an external meteorological data platform (such as a local LoRa gateway). The refresh cycle is 60 seconds. The output of step one is calculated using spectral-structure mapping and represents the instantaneous radiative thermal disturbance input.

[0027] Considering that indoor temperature changes are affected by a combination of factors such as heat capacity, heat exchange, and light disturbance, a basic model for heat balance prediction is first constructed: ; In this model, This refers to heat capacity, measured in J / K. It can be estimated during the deployment phase based on room volume and material properties. For example, for a medium-sized office, take... ; Background heat source input includes heat generated by personnel activities and equipment operation, which can be based on daily average historical values, such as 70W; The heat transfer coefficient, measured in W / K, reflects the heat exchange capacity between the building envelope and the outside environment, and is typically 3.2. The prediction time step is set to 5 minutes (300 seconds). This formula provides a basic expression for deriving future temperature change trends from the current temperature, but if we stop here, we cannot fully reflect the true impact of multispectral input on the rate of temperature rise.

[0028] To improve the performance of the prediction model under complex illumination perturbations, especially in scenarios with high-energy instantaneous concentration in the infrared band, this step further introduces a "spectral concentration regularization term." This term is designed to penalize the concentration of energy from high-energy bands rather than uniform diffusion. The input ensures that the model does not overestimate the temperature rise trend under conditions of high thermal perturbation concentration. A concentration index is defined. The normalized standard deviation (standard deviation divided by the mean) of the spectral intensity of the current band reflects the dispersion of the band's power density. ; in, For the first The spectral power density for each band, expressed in W / m². The total number of bands, This represents the average value of the band at that moment. The larger the value, the more concentrated the thermal disturbance is in a few high-energy bands, and the stronger the non-uniformity. Multiply this term by the regularization weight. (Empirically set at 1.2~2.0), forming a prediction correction factor. This is used to down-adjust the rate of temperature rise, resulting in the final prediction expression: ; In this corrected expression, all variables can be derived from the pre-sensor and modeling parameters. The calculation process only involves the mean, variance, and simple multiplication and division operations, and it has real-time deployment capabilities. Taking a certain experiment as an example, when Concentrated in the near-infrared band, ,but The corresponding predicted temperature rise will be reduced to the initial estimate. This effectively prevents temperature prediction overshoot caused by sudden light disturbances.

[0029] S3: Based on the equivalent radiative thermal disturbance sequence and its corresponding predicted temperature response over a continuous time period, identify whether the current thermal disturbance type is stationary or abrupt. Specifically, this step identifies the variation pattern of light-induced thermal disturbances over time, determining whether they are stationary or abrupt disturbances, thus providing a preliminary decision signal for the selection of control strategies in the next step. This step is based on the temperature prediction results output from the previous stage. With light and heat disturbance By using time series feature analysis to identify disturbance patterns, the controller can perform "advance compensation" before temperature deviations occur. Unlike the traditional method of using fixed thresholds to judge changes in illumination, this step combines the dynamic characteristics of spectral disturbances with the predictive response capability of thermal models to propose a physically interpretable "disturbance-response coupling analysis method" that can reflect the real impact of thermal disturbances on indoor temperature dynamics.

[0030] The input includes the predicted temperature sequence output from the second step. and the equivalent thermal disturbance output in the first step The system first constructs a time window. Used to capture disturbances in the past The fluctuation characteristics at each sampling time. The window length is set according to the control frequency, and is usually taken as... This refers to the disturbance record over the past 5 minutes. All data comes from the internal output of the real-time sensor measurement or prediction module, without involving manual input or additional sampling.

[0031] To improve the stability of disturbance identification and avoid misjudging short-time noise as abrupt changes, this step innovatively introduces the "thermal response normalized disturbance intensity" index. It couples the rate of change of thermal perturbation with the rate of change of temperature prediction and adds a spectral smoothing regularization term. This is to eliminate the effects of high-frequency jitter. The metric is defined as follows: ; In this formula, the numerator measures the cumulative change of the light-heat perturbation within the sliding window, while the denominator reflects the magnitude change of the temperature prediction response, and is expressed by adding terms. Regular smoothing is introduced to suppress short-term spectral measurement noise. The weighting coefficient is typically taken as... ; The term representing the smoothing of the spectral energy distribution, used to constrain the uniformity of spectral power variation, is defined as: ; in For the first The spectral power density of each band is obtained from real-time sampling data from a multispectral sensor. The introduction of this feature is a unique design element for multispectral thermal disturbance scenarios. This is because ordinary illuminance changes may not lead to abrupt temperature changes, but rather to a nonlinear thermal response caused by concentrated enhancement in specific wavelength bands (such as the near-infrared high-energy band). This is achieved by incorporating [a specific element] into the disturbance intensity calculation. The system can automatically distinguish between "band energy redistribution" and "actual energy input increase", thus significantly reducing the probability of misjudgment.

[0032] This reflects the relative intensity of the effect of light disturbance on room temperature change. The larger the value, the greater the magnitude of the current thermal disturbance change relative to the temperature response magnitude, meaning the disturbance leads the system response and exhibits abrupt change characteristics. The system automatically generates dynamic thresholds based on historical operating data or the initialization phase. (usually set to the past hour) (2.5 times the average value), the judgment logic is as follows: ; Output variables This label indicates the type of the current disturbance, with values ​​of "transient" or "stable". For example, in a scenario where afternoon light intensifies, when sunlight suddenly shines into a room from a shaded area through glass, The temperature rose nearly twofold in an instant, while the predicted temperature change was only [a fraction of the predicted change]. ℃, at this time Exceeding the adaptive threshold The system immediately determined This provides a signal for subsequent control strategies to trigger the rapid compensation mode in advance.

[0033] This step relativizes the rate of change of spectral perturbation with the predicted temperature response rate, while incorporating a spectral smoothing regularization term to suppress noise, thus achieving coupled perturbation identification between multidimensional sensing and thermodynamic prediction. This method can perform real-time identification based solely on time series data without any model assumptions, exhibiting extremely high reproducibility and deployment efficiency. Its output not only determines the perturbation state but also provides a basis for subsequent control strategy selection, enabling the system to proactively determine response strategies before illumination perturbations occur.

[0034] S4: Based on the predicted temperature change trend and disturbance type, dynamically generate compensation control commands for the heat pump terminal and drive the heat pump to perform corresponding adjustment actions.

[0035] Specifically, this step is used to determine the predicted temperature obtained in the previous stage. and disturbance type labels Dynamically generate the adjustment and control quantity of the heat pump terminal. This system guides the operation of equipment such as fan speed, water valve opening, and compressor load, thereby achieving proactive compensation and control of solar radiation heat disturbances. This control command is not calculated from the current temperature deviation, but is entirely based on a combined analysis of predicted information and disturbance trends, a key step in realizing a "feedforward regulation mechanism." Through this method, the system can respond to changes in solar radiation trends before the temperature changes significantly, preventing temperature fluctuations from exceeding the set range and fundamentally compensating for the lag problem of traditional feedback control in highly dynamic disturbance environments.

[0036] The input for this step includes: the predicted temperature value output from the second step. The disturbance type label output in the third step and the target temperature set by the user. . The results are calculated by combining a thermodynamic prediction model with spectral perturbation data, and are predictions of indoor temperature change trends over a future period of time (e.g., 5 minutes). The label is derived through a thermal disturbance response identification mechanism, indicating whether the disturbance is stable or transient. The system reads the target comfort temperature set by the user on the temperature control panel or remote interface, typically accurate to 0.1°C, from local storage or periodically synchronizes it.

[0037] Based on the input data above, the system first calculates the prediction deviation. This is used to indicate the degree of deviation between the predicted temperature and the target temperature. ; This deviation is the core control basis for this step. A positive deviation indicates anticipated overheating, and cooling should begin earlier; a negative deviation indicates anticipated undercooling, and heating should be activated; a deviation close to zero indicates that the room temperature trend is stabilizing, and the current operating state can be maintained. Compared to traditional real-time deviation (… This prediction bias has a significant advantage in response speed and can significantly reduce the energy consumption fluctuation of the temperature control system.

[0038] In generating specific control strategies, the system sets up two adjustment paths. The first path is suitable for... In this scenario, the system employs a linear proportional control strategy, where the control quantity is proportional to the prediction deviation, resulting in a smooth response and suitability for stable operation. The second path is suitable for... In such cases, the system will introduce a disturbance response enhancement mechanism to improve the adjustment speed and suppress the hysteresis heating effect caused by sudden thermal disturbances. To unify the management of the two paths, the system will unify the expression of the control quantity calculation as follows: ; in, This represents the temperature control proportional coefficient, with units of W / ℃. It is set during deployment based on room volume, heat capacity, and terminal equipment capabilities, and typically ranges from 80 to 150. It is the disturbance enhancement gain, with a setting range of 0.5~2.0, used to enhance the control amplitude under sudden disturbance conditions; This is an indicator function; it takes the value 1 when the disturbance type is a sudden change, and 0 otherwise. The unit of this control quantity is W, and it can be directly input into the heat pump controller as the heat pump output power setpoint.

[0039] For example, in a typical application scenario, the current prediction is that the temperature rise in the next 5 minutes will be... ℃, the set temperature is 25.5℃, , If the disturbance type is stable, then the output control quantity is... W (indicates a steady increase in cooling); if the disturbance type is transient, the output is... W (Advance Cooling). This strategy allows for proactive adjustment before thermal disturbances significantly affect room temperature, effectively preventing room temperature overshoot.

[0040] Finally, this step outputs the variables. As a control quantity, it is directly transmitted to the execution unit of the heat pump terminal equipment. It can be written into actuators such as fan coil units, water valve controllers or compressor control boards through Modbus, KNX or BACnet protocols to complete physical adjustment actions.

[0041] In one or more embodiments, such as Figure 2 As shown, a dynamic prediction and compensation control system for solar radiation heat load based on multispectral sensing is disclosed. The system includes: The thermal disturbance modeling module is used to collect light intensity data of multiple bands in real time through multispectral sensors deployed indoors; combined with the geometric structure, material absorptivity and light incident angle information of each heat-absorbing surface indoors, a spectral spatial coupling mapping model is constructed to calculate the equivalent radiative thermal disturbance caused by light at the current moment. The temperature prediction module is used to take the equivalent radiative heat disturbance as the thermal disturbance input, integrate the current indoor and outdoor temperatures and room thermodynamic parameters, construct a thermodynamic prediction model, and predict the indoor temperature change trend within a preset time window in the future. The disturbance identification module is used to identify whether the current thermal disturbance type is stationary or abrupt based on the equivalent radiative heat disturbance quantity sequence and its corresponding predicted temperature response over a continuous time period. The control execution module is used to dynamically generate compensation control commands for the heat pump terminal based on the predicted temperature change trend and disturbance type, and drive the heat pump to perform corresponding adjustment actions.

[0042] It is worth noting that the specific workflow of the dynamic prediction and compensation control system for light radiation heat load based on multispectral sensing provided in this embodiment of the invention is the same as that of the dynamic prediction and compensation control method for light radiation heat load based on multispectral sensing described in the above embodiments, and will not be repeated here.

[0043] This invention also provides a device for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps described in the embodiments of the method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing. Figure 1 The steps S1 to S4 described above; or, when the processor executes the computer program, it implements the functions of each module in the above system embodiments.

[0044] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the multispectral sensing-based dynamic prediction and compensation control device for light radiation heat load.

[0045] The multispectral sensing-based dynamic prediction and compensation control device for solar radiation heat load can be a desktop computer, laptop, handheld computer, or cloud server, etc. This device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the multispectral sensing-based dynamic prediction and compensation control device for solar radiation heat load can also include input / output devices, network access devices, buses, etc.

[0046] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASACs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. This processor is the control center of the multispectral sensing-based dynamic prediction and compensation control device for solar radiation heat load, connecting all parts of the device via various interfaces and lines.

[0047] The memory can be used to store the computer program and / or modules. The processor, by running or executing the computer program and / or modules stored in the memory and calling the data stored in the memory, realizes various functions of the dynamic prediction and compensation control device for light radiation heat load based on multispectral sensing. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created based on the operation of the air conditioner controller, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0048] The module integrated into the dynamic prediction and compensation control device for light radiation heat load based on multispectral sensing, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0049] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0050] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing, characterized in that, The method includes: In real time, light intensity data of multiple bands are collected by multispectral sensors deployed indoors; combined with the geometric structure, material absorptivity and light incident angle information of each heat-absorbing surface indoors, a spectral spatial coupling mapping model is constructed to calculate the equivalent radiative heat disturbance caused by light at the current moment. Using the equivalent radiative heat disturbance as the thermal disturbance input, and integrating the current indoor and outdoor temperatures and room thermodynamic parameters, a thermodynamic prediction model is constructed to predict the indoor temperature change trend within a preset time window in the future. Based on the equivalent radiative thermal disturbance sequence over a continuous time period and its corresponding predicted temperature response, the current thermal disturbance type is identified as either stationary or abrupt. Based on the predicted temperature change trend and disturbance type, compensation control commands for the heat pump terminal are dynamically generated, and the heat pump is driven to perform corresponding adjustment actions.

2. The method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing according to claim 1, characterized in that, The multispectral sensor covers the visible to near-infrared bands, has a sampling frequency of no less than once per second, and outputs spectral power density data for each band.

3. The method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing according to claim 1, characterized in that, The heat-absorbing surface includes curtains, walls, floors, or furniture surfaces. Its geometry and orientation are obtained through spatial three-dimensional modeling during the deployment phase, and the material absorption rate is preset based on a standard material spectral database.

4. The method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing according to claim 1, characterized in that, The calculation of the equivalent radiative heat disturbance incorporates an incident angle correction based on Lambert's cosine law to reflect the influence of different irradiation angles on the surface heat absorption efficiency.

5. The method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing according to claim 1, characterized in that, The thermodynamic prediction model introduces a spectral concentration regularization term on the basis of the basic heat balance equation to suppress temperature prediction overshoot caused by energy concentration in the high-energy band.

6. The method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing according to claim 1, characterized in that, The identification of the perturbation type is based on the ratio of the thermal perturbation change rate to the predicted temperature response rate, and combined with the spectral energy distribution smoothness index to determine noise suppression and perturbation effectiveness.

7. The method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing according to claim 1, characterized in that, The disturbance type identification adopts an adaptive dynamic threshold mechanism, which automatically adjusts the threshold based on the statistical characteristics of the disturbance response index during historical runtime.

8. The method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing according to claim 1, characterized in that, The compensation control command adopts a dual-channel control strategy: under stable disturbances, the heat pump output is linearly adjusted according to the predicted deviation; under abrupt disturbances, the enhanced gain mechanism is activated to improve the adjustment amplitude and response speed.

9. The method for dynamic prediction and compensation control of solar radiation heat load based on multispectral sensing according to claim 1, characterized in that, The heat pump terminal regulation includes at least one of compressor load, fan speed, or water valve opening, and the control commands are transmitted to the execution unit through a standard building automation protocol.

10. A dynamic prediction and compensation control system for solar radiation heat load based on multispectral sensing, characterized in that, The system includes: The thermal disturbance modeling module is used to collect light intensity data of multiple bands in real time through multispectral sensors deployed indoors; combined with the geometric structure, material absorptivity and light incident angle information of each heat-absorbing surface indoors, a spectral spatial coupling mapping model is constructed to calculate the equivalent radiative thermal disturbance caused by light at the current moment. The temperature prediction module is used to take the equivalent radiative heat disturbance as the thermal disturbance input, integrate the current indoor and outdoor temperatures and room thermodynamic parameters, construct a thermodynamic prediction model, and predict the indoor temperature change trend within a preset time window in the future. The disturbance identification module is used to identify whether the current thermal disturbance type is stationary or abrupt based on the equivalent radiative heat disturbance quantity sequence and its corresponding predicted temperature response over a continuous time period. The control execution module is used to dynamically generate compensation control commands for the heat pump terminal based on the predicted temperature change trend and disturbance type, and drive the heat pump to perform corresponding adjustment actions.

Images