Large model assisted human settlement space light environment twinborn modeling and body intelligent control method

Abstract

The invention discloses a large-model-assisted human settlement space light environment twinborn modeling and body intelligent control method. The method comprises the steps of S1, performing data acquisition on human settlement space light environment parameters; s2, preprocessing the light environment parameters to obtain twinborn data, and establishing a light environment twinborn model; s3, integrating values of the twinborn model to a cue word template to generate cue words; s4, sending the cue word to the large model platform for solving, and generating a control strategy; s5, converting the control strategy into a control signal, and sending the control signal to a light source driver; and S6, repeatedly executing the steps S1 to S5 until the value of the twinborn model is within the standard value range. A new normal form is provided for intelligent control of the living space light environment, the accuracy of light environment parameter adjustment can be improved through large-model-assisted light environment adjustment, high adaptability is achieved, and the method is suitable for various scenes such as intelligent homes and office environments.

Description

Technical Field

[0001] The present invention relates to the field of residential light environment control, and in particular to a large-model-assisted embodied intelligent control method and system for residential space light environment. Background Art

[0002] With the increasing demand for intelligent living spaces, traditional light environment control methods (such as timer switches and fixed brightness adjustment) have limitations such as being unable to dynamically respond to environmental parameters, lacking refined adjustment strategies, and having difficulty meeting the differentiated needs of different users for light intensity and color temperature.

[0003] With the popularization of the Internet of Things, single-chip microcomputers are generally used as economical and practical control nodes. However, the computing power of single-chip microcomputers is limited. They are not capable of embodying intelligent control of the light environment in human living spaces with complex requirements. DSP, ARM (high-performance microprocessors), etc. are needed.

[0004] How to build an economical control system for the light environment parameters of human living spaces and create a healthy, comfortable, energy-saving and intelligent light environment for human living spaces has become a hot issue and has attracted much attention. Summary of the invention

[0005] The purpose of the present invention is to provide a large-model-assisted embodied intelligent control method and system for the light environment of a human living space, so as to improve the accuracy of adjusting the light environment parameters.

[0006] To this end, on the one hand, the present invention provides an embodied intelligent control method for the light environment of a human living space assisted by a large model, comprising the following steps: S1, data collection of light environment parameters of a human living space; S2, preprocessing of the light environment parameters to obtain twin data, and establishment of a light environment twin model; S3, integration of the value of the twin model into a prompt word template to generate a prompt word; S4, sending the prompt word to the large model platform for solution to generate a control strategy; S5, converting the control strategy into a control signal, and sending it to the light source driver to adjust the control parameters of the lamp; S6, repeating the above steps S1 to S5 until the value of the twin model is within the standard value range.

[0007] According to another aspect of the present invention, there is provided an embodied intelligent control system for the light environment of a human living space assisted by a large model, comprising a plurality of LED lamps, an embodied intelligent controller for controlling the light environment parameters of the plurality of LED lamps, and a large model platform, wherein the embodied intelligent controller interacts with the large model platform to implement the steps of the above-mentioned large model-assisted embodied intelligent control method for the light environment of a human living space.

[0008] Based on big models, digital twins, and embodied intelligence technologies, the present invention proposes a big model-assisted twin modeling and embodied intelligent control method for the light environment of human settlements. The present invention uses multiple sensors to dynamically sense the light environment parameters of human settlements and implements twin modeling. Based on the big model, the light environment control strategy of human settlements is obtained, and the multi-parameters of the light environment of human settlements are refined and personalized, ensuring the health, comfort, and energy saving of human settlements.

[0009] The present invention provides a new paradigm for intelligent control of the light environment in human settlements. Large-scale model-assisted light environment adjustment in human settlements can not only improve the accuracy of light environment parameter adjustment, but also has strong adaptability and is suitable for a variety of scenarios such as smart homes and office environments.

[0010] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 It is a model block diagram of the embodied intelligent control system of the light environment of the human living space of the present invention; Figure 2 It is a flow chart of the embodied intelligent control method of the light environment of a human living space of the present invention; Figure 3 It is a schematic diagram of the interaction between the embodied intelligent controller of the living space light environment designed by the present invention and the large model through prompt words; Figure 4 It is a flow chart of the control strategy for adjusting the light environment parameters of the living space in the rule description of the prompt words designed by the present invention; Figure 5 It is a structural block diagram of the embodied intelligent controller for the residential light environment of the present invention. DETAILED DESCRIPTION

[0012] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0013] The model framework of the embodied intelligent control system of the living space light environment of the present invention is as follows Figure 1 As shown in the figure, it consists of three parts: living space light environment system, embodied intelligent controller and large model platform.

[0014] The living space light environment system includes a plurality of LED light sources distributed in the living space and a light source driver corresponding to each LED light source.

[0015] In the present invention, a plurality of LED light sources are already distributed in the living space, and have a set brightness distribution, such as a uniform brightness distribution, or an artificially set highly uneven brightness distribution.

[0016] The embodied intelligent controller proposed in the present invention has a perception, thinking and execution unit. The perception unit is one or more sensors used to collect light environment parameters. For example, a light intensity sensor is used to collect light intensity parameters. The thinking unit includes a twin data module, a twin model module and a prompt word template module. The execution unit is used to convert the control strategy into a control signal.

[0017] The large model platform obtains the desired results, namely the control strategy, through natural language interaction and standardized questioning methods.

[0018] The present invention delegates computational tasks with large computational intensity to a large model platform for processing, forming a computing mode that combines edge computing with cloud computing. In this way, the computational intensity of the embodied intelligent controller itself is small, and it can be built using an ordinary single-chip microcomputer, meeting the requirements of an economical and applicable system.

[0019] The interaction mode of each unit in the embodied intelligent control system of the living space light environment of the present invention is as follows: 1) The embodied intelligent controller dynamically senses and collects the parameters of the human living light environment X={ x 1 , x 2 ,... x i ..., x n}, where x i Represents the illuminance collected at i points of the light environment of the living space; 2) The embodied intelligent controller pre-processes the collected light environment parameters to obtain twin data of the light environment parameters Y ={ y 1 , y 2 ,... y i ..., y n}; The preprocessing of the present invention is mainly to remove abnormal data, and interval limiting filtering can be used to remove abnormal data. The formula is as follows: (1) In the formula, if x i Less than threshold λ ,but y i The value is 0; if x iGreater than or equal to threshold λ , and is less than or equal to the threshold ε, but y i equal x i ;like x i Greater than threshold ε ,but y i The value is 0; the threshold λ , ε Represents the lower and upper limits of the interval limiting filter, which come from manual experience or manual settings.

[0020] 3) The embodied intelligent controller obtains the twin model of the light environment through modeling F ; The present invention adopts linear function to carry out twin modeling, and the formula is as follows: (2) In the formula, α i Represents the light environment parameters after preprocessing y i The coefficients of α i Can take value ,in d i Indicates i The distance between a light source and a reference point in the human-occupied space.

[0021] 4) Embodied intelligent controller will twin the light environment model F Template with prompt words M Integration, generation of prompt words for embodied intelligent control of the light environment in human living spaces P , and send prompt words P To large model platform; 5) The large model platform generates a light environment control strategy for the living space based on the prompt words C , and send the control strategy to the embodied intelligent controller, C= 0 is negative adjustment, C= 1 is positive regulation; 6) The embodied intelligent controller execution unit will control the strategy C Converted into control signal S , sending the light source driver of each LED light source to control the illumination parameters of each LED light source.

[0022] like Figure 2 As shown, the embodied intelligent control process of the light environment of the living space proposed by the present invention includes the following steps S1-S6: S1, embodied intelligent controller data collection and perception of living space light environment parameters; S2, embodied intelligent controller to carry out digital twin modeling of the light environment of human living space; S3. Based on the twin model of the human living light environment, the embodied intelligent controller integrates the twin data with the prompt word template to generate prompt words; S4, the prompt words are sent to the large model platform for solution and generation of control strategy; S5, the embodied intelligent controller converts the control strategy into a control signal and sends it to the light source driver; (3) In the formula, if C =0, the control signal S To reduce the light intensity k lm, if C =1, the control signal S Increase the light intensity k lm, here k is the adjustment step size. k The value is 10 lm, and can also be set to other values.

[0023] The above control signal is an average value, which is distributed to each lamp to achieve synchronous increase or decrease of the light environment parameters of all lamps in the light environment.

[0024] S6, repeat the above steps S1 to S5 until F Within the standard value range.

[0025] like Figure 3 As shown in the figure, the prompt word structure for embodied intelligent control of the light environment in the living space is as follows: Prompt word (P) = role definition + rule description + question content + output format Role definition: Clarify the roles and responsibilities of the large model, such as "light environment control assistant for living space."

[0026] Rule description: Define the calculation rules, such as F Indicates the indoor illuminance, definition W Indicates the standard value of indoor illuminance, defined as C Indicates the switch status of the indoor light source. C =1 means the indoor light source is adjusted in the positive direction, C =0 means negative adjustment of indoor light source, the adjustment state of indoor light source C Indoor light intensity F and W to determine, if F >W, then C=0, if F <W, then C =1.

[0027] Question content: Define the question content and output requirements. For example: Currently, the indoor light intensity is F , the standard value is D , please give the value of the adjustment state C of the indoor light source according to the above rules.

[0028] Output format: defines the format requirements for large model generation control strategy. For example: C =1, or C =0.

[0029] Among them, the process of generating control strategies for the large model platform is as follows: Figure 4 As shown, it specifically includes the following steps S41-S46.

[0030] S41, the large model analyzes the light environment parameters of the living space from the prompt words; S42, determining the range to which the living light environment parameter belongs; if the light environment parameter deviates negatively from the standard value (i.e., is less than the standard value), proceeding to S43; if the light environment parameter deviates positively from the standard value, proceeding to S44; if the light environment parameter does not deviate, proceeding to S45; S43, adjust the light source in the positive direction, and go to S46; S44, adjust the light source in the negative direction, and go to S46; S45: No operation, go to S46; S46. Output the operation result.

[0031] like Figure 5 As shown, the structure of the embodied intelligent control system for human living light environment in this embodiment is selected as follows:

[0032] The embodied intelligent controller is composed of a microprocessor, a network interface, a multi-channel light source driver, an A / D conversion module and a power supply module. The microprocessor can be an esp8266 single-chip microcomputer.

[0033] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A large-model-assisted embodied intelligent control method for the light environment of a human living space, characterized in that: Includes steps: S1. Data collection of light environment parameters of human living space; S2. Preprocess the light environment parameters to obtain twin data and establish a light environment twin model; S3, integrating the value of the twin model into the prompt word template to generate a prompt word; S4, sending the prompt words to the large model platform for solving and generating a control strategy; S5, converting the control strategy into a control signal and sending it to the light source driver to adjust the control parameters of the lamp; S6. Repeat the above steps S1 to S5 until the value of the twin model is within the standard value range.

2. The large model-assisted embodied intelligent control method for the light environment of a living space according to claim 1 is characterized in that: In step S1, a photometer is used to collect light environment parameters of a living space.

3. The large model-assisted embodied intelligent control method for the light environment of a living space according to claim 1, characterized in that: In step S2, a linear function is used to carry out twin modeling, and the formula is as follows: （1） In the formula, α i Represents the light environment parameters after preprocessing y i The coefficient of .

4. The large-model-assisted embodied intelligent control method for the light environment of a living space according to claim 3 is characterized in that: The preprocessing method of light environment parameters is to use interval limiting filtering to eliminate abnormal data.

5. The large-model-assisted embodied intelligent control method for the light environment of a living space according to claim 1 is characterized in that: In step S3, the prompt word template consists of four parts: role definition, rule description, question content and output format. The role definition is used to indicate the role and responsibilities played by the large model. The rule description is used to define the calculation rules of the large model, that is, to define the correspondence between the prompt words and the generated control strategies. The question content is used to define the question content. The output format is used to define the format requirements for the large model to generate the control strategy.

6. The large model-assisted embodied intelligent control method for the light environment of a human living space according to claim 5 is characterized in that: The calculation rule is that if the value of the twin model deviates negatively from the given standard value, the light source is adjusted positively; if the value of the twin model deviates positively from the given standard value, the light source is adjusted negatively.

7. The large model-assisted embodied intelligent control method for the light environment of a living space according to claim 4 is characterized in that: The control strategy is represented by C, where C=1 represents positive regulation of the light source, and C=0 represents negative regulation of the light source.

8. The large model-assisted embodied intelligent control method for the light environment of a human living space according to claim 1 is characterized in that: In step S4, the base model of the large model selected is the GLM-4 model.

9. The large model-assisted embodied intelligent control method for the light environment of a human living space according to claim 1, characterized in that: In step S5, the control signal is a positive control step length of the light parameter or a negative control step length of the light parameter corresponding to the control strategy.

10. An embodied intelligent control system for the light environment of a living space, characterized in that: The invention comprises a plurality of LED lamps, an embodied intelligent controller for controlling the light environment parameters of the plurality of LED lamps, and a large model platform. The embodied intelligent controller interacts with the large model platform to implement the steps of the large model-assisted embodied intelligent control method for the light environment of a human living space according to any one of claims 1 to 9.

Images