Energy-saving control method of museum lighting system combined with visitor behavior prediction

Abstract

The present invention discloses an energy-saving control method for a museum lighting system combined with prediction of visitor behavior, and relates to the field of energy-saving control technology. The method comprises: collecting the behavior parameters of museum visitors in real time according to a sensor network to obtain visitor monitoring results; collecting the lighting parameters of M museum partitions to obtain M partition lighting data; predicting visitor behavior to obtain M partition visitor prediction results; performing lighting adaptability detection to determine the lighting partitions to be adjusted; making lighting expectation constraint decisions, establishing a lighting strategy space, and performing energy consumption minimization optimization according to the lighting strategy space to generate a lighting adjustment plan; and performing lighting control on the lighting partitions to be adjusted according to the lighting adjustment plan. The method solves the technical problem in the prior art that the museum lighting system lacks dynamic lighting adjustment, resulting in energy waste. By predicting visitor behavior and dynamically adjusting the lighting of each partition of the museum accordingly, the technical effect of energy saving and consumption reduction is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy-saving control, and in particular to an energy-saving control method for a museum lighting system in combination with prediction of visitor visiting behavior. Background Art

[0002] As an important place for cultural inheritance and knowledge dissemination, museums attract millions of visitors every year. Museums are gradually introducing intelligent means in management and services to enhance the visiting experience and optimize operating costs. Traditional museum lighting systems usually use fixed time periods or simple sensors to control lighting, which is difficult to accurately adapt to the real-time visiting behavior of tourists, resulting in unnecessary energy consumption during periods with fewer tourists or idle time. Especially in large museums, lighting needs vary greatly in different areas and time periods. A single lighting control strategy not only affects energy saving effects, but may also lead to poor lighting effects in some areas, affecting the visiting experience. Summary of the invention

[0003] The present application provides an energy-saving control method for a museum lighting system in combination with prediction of visitor visitor behavior, which solves the technical problem in the prior art that the museum lighting system lacks dynamic lighting adjustment, resulting in energy waste.

[0004] In view of the above problems, the present application provides an energy-saving control method for a museum lighting system combined with prediction of visitor visiting behavior.

[0005] The present application provides a museum lighting system energy-saving control method combined with visitor visit behavior prediction, the method comprising:

[0006] According to the sensor network, the behavior parameters of the museum's visitors are collected in real time to obtain visitor monitoring results; the museum is divided according to the museum layout diagram to obtain M museum partitions, and the lighting parameters of the M museum partitions are collected in real time according to the museum lighting system to obtain M partition lighting data, wherein M is a positive integer greater than 1; based on the predetermined future time zone, the visitor behavior of the M museum partitions is predicted according to the visitor monitoring results to obtain M partition visitor prediction results; based on the M partition visitor prediction results and the M partition lighting data, the M museum partitions are tested for lighting adaptability according to the lighting adaptability detection module to determine the lighting partitions to be adjusted; based on the lighting partitions to be adjusted, lighting expectation constraint decisions are made, a lighting strategy space is established, and energy consumption minimization is optimized according to the lighting strategy space to generate a lighting adjustment plan; based on the museum lighting system, lighting control is performed on the lighting partitions to be adjusted according to the lighting adjustment plan.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] First, according to the sensor network, the tourist behavior parameters of the museum are collected in real time to obtain the tourist monitoring results. Then, the museum is divided according to the museum layout diagram to obtain M museum partitions, and the lighting parameters of the M museum partitions are collected in real time according to the museum lighting system to obtain M partition lighting data, where M is a positive integer greater than 1. Next, based on the predetermined future time zone, the M museum partitions are predicted for the tourist visiting behavior according to the tourist monitoring results to obtain the M partition tourist prediction results. Then, based on the M partition tourist prediction results and the M partition lighting data, the M museum partitions are tested for lighting adaptability according to the lighting adaptability detection module to determine the lighting partitions to be adjusted. Further, based on the lighting partitions to be adjusted, the lighting expectation constraint decision is made, the lighting strategy space is established, and the energy consumption is minimized and optimized according to the lighting strategy space to generate the lighting adjustment plan. Finally, based on the museum lighting system, the lighting control is performed on the lighting partitions to be adjusted according to the lighting adjustment plan. The technical problem that the museum lighting system lacks dynamic lighting adjustment in the prior art, resulting in energy waste, is solved. By predicting the tourist visiting behavior and dynamically adjusting the lighting of each partition of the museum accordingly, the technical effect of energy saving and consumption reduction is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic flow chart of a method for energy-saving control of a museum lighting system in combination with prediction of visitor behavior provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of a flow chart for predicting visitor behavior in a method for energy-saving control of a museum lighting system combined with visitor behavior prediction provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] The present application solves the technical problem in the prior art that the museum lighting system lacks dynamic lighting adjustment, resulting in energy waste, by providing an energy-saving control method for the museum lighting system combined with prediction of visitor visiting behavior.

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.

[0015] Examples, such as Figure 1 As shown, the embodiment of the present application provides a museum lighting system energy-saving control method combined with visitor visit behavior prediction, wherein the method includes:

[0016] Based on the sensor network, the museum's visitor behavior parameters are collected in real time to obtain visitor monitoring results.

[0017] Through the sensor network deployed in the museum, visitor behavior parameters can be collected in real time, including but not limited to the number of visitors, location, movement speed, stay time, path trajectory and other information. Among them, the sensor network includes infrared sensors, cameras, thermal imaging equipment or Wi-Fi / Bluetooth signal receivers, etc., which can monitor and record visitor activities without affecting the privacy of visitors. Through data collection and analysis, the visitor monitoring results are formed, which provides basic data support for subsequent visitor behavior prediction and lighting adjustment.

[0018] The museum is divided according to the museum layout diagram to obtain M museum partitions, and lighting parameters of the M museum partitions are collected in real time according to the museum lighting system to obtain M partition lighting data, wherein M is a positive integer greater than 1.

[0019] Based on the museum layout diagram, the museum can be divided into M partitions according to factors such as exhibition hall area, exhibit layout, channel distribution, etc., where M represents the number of museum partitions after division and is a positive integer greater than 1. After the division is completed, the lighting parameters of each partition are collected in real time through the museum's lighting system, including lighting brightness, color temperature, enabled status, etc., to form M partition lighting data. Through the real-time collection of partition lighting data, the actual lighting status of each partition can be better monitored, providing a data basis for subsequent lighting adaptability detection and adjustment.

[0020] Based on the predetermined future time zone, visitor visit behaviors of the M museum partitions are predicted according to the visitor monitoring results to obtain visitor prediction results of the M partitions.

[0021] Set a future target time zone (such as one hour) and predict tourist behavior based on the visiting demand in that time zone. Specifically, based on the existing tourist monitoring results, such as the real-time number of tourists, stay time, flow trend and other data of each zone, use machine learning or statistical analysis methods to predict the visitor behavior of M museum zones, and obtain the tourist prediction results of each zone in the future time zone. The tourist prediction results can include the expected number of tourists, flow path and stay time of each zone in the target time zone. Through this prediction information, the visiting popularity distribution of each zone can be understood in advance, so as to provide data support for the reasonable adjustment of the lighting demand of each zone, and realize intelligent and precise lighting control.

[0022] Furthermore, if Figure 2 As shown, based on the predetermined future time zone, the visitor behavior of the M museum partitions is predicted according to the visitor monitoring results, and the M partition visitor prediction results are obtained, including:

[0023] According to the M museum partitions, the mth museum partition is extracted, wherein m is a positive integer; based on the mth museum partition, feature recognition is performed on the visitor monitoring result to determine the visitor monitoring result of the mth partition; based on the predetermined future time zone, visitor visiting behavior of the mth museum partition is predicted according to the visitor monitoring result of the mth partition to obtain a first result of visitor prediction for the partition; based on the predetermined future time zone, the visitor visiting behavior of the mth museum partition in the same zone is mined to obtain a second result of visitor prediction for the partition; the first result of visitor prediction for the partition and the second result of visitor prediction for the partition are integrated to generate a visitor prediction result for the mth partition, and the visitor prediction result for the mth partition is added to the M visitor prediction results for the partition.

[0024] Specifically, according to the M museum partitions, the tourist monitoring data of each partition is extracted one by one. It is assumed that the mth partition is currently extracted, where m is a positive integer; in the mth partition, feature recognition is performed on the real-time collected tourist monitoring results, and behavioral data related to the partition, such as the number of tourists, flow frequency, and residence time, are extracted to generate the tourist monitoring results of the mth partition; using the tourist monitoring results of the mth partition, a prediction model is constructed. The prediction model can be a statistical model (such as time series analysis, regression analysis, etc.) or a machine learning model (such as neural network, decision tree, etc.); the prediction model is trained and verified using historical tourist monitoring data to ensure the accuracy and reliability of the model; the trained prediction model is applied to the predetermined future time zone, and the M The visitor behavior of the museum's partitions is predicted to obtain the first result of the partition tourist prediction, which includes the prediction of the number of tourists, visiting routes and stay time in the future time zone; then the visitor behavior of the m-th partition is mined, that is, the possible tourist behavior characteristics in the same time period in the future time zone are deeply mined to obtain the second result of the partition tourist prediction; the first result and the second result of the partition tourist prediction are integrated to generate the m-th partition tourist prediction result, which more comprehensively reflects the visiting behavior of the partition in the predetermined future time zone; the tourist prediction result of the m-th partition is added to the tourist prediction results of M partitions to form an overall museum partition visiting behavior prediction, providing data support for subsequent lighting control decisions.

[0025] Furthermore, based on the predetermined future time zone, the visiting behavior of tourists in the same time zone is mined for the m-th museum zone to obtain a second result of zone tourist prediction, including:

[0026] According to the predetermined future time zone, multiple historical time zones with the same characteristics are set; based on the multiple historical time zones with the same characteristics, the historical tourist visiting behavior of the mth museum partition is queried to obtain multiple historical tourist behavior query results; based on the multiple historical tourist behavior query results, a central trend analysis is performed to generate the second result of the partition tourist prediction.

[0027] Based on the scheduled future time zones, historical time zones with similar characteristics are selected as references, such as similar time periods, dates or holidays, etc., to ensure the reference value of historical data. The setting of multiple historical time zones with the same characteristics enables the prediction results to better match the behavioral characteristics of future time zones. Based on the selected multiple historical time zones with the same characteristics, the tourist visit behavior data of the mth partition in these historical time zones are queried to obtain multiple historical tourist behavior query results. The historical tourist behavior query results include the number of tourists, visiting routes, average stay time, congestion, etc., and provide past visits in similar time zones. Perform a central trend analysis on multiple historical tourist behavior query results, such as calculating the mean, median, mode, etc. of the number of historical tourists, identifying the common trends of tourist behavior, and thus generating a second result of the partition tourist prediction.

[0028] Based on the tourist prediction results of the M partitions and the lighting data of the M partitions, the lighting adaptability detection module is used to perform lighting adaptability detection on the M museum partitions to determine the lighting partitions to be adjusted.

[0029] The lighting adaptability detection module receives the tourist prediction results of M partitions and the lighting data of M partitions. The tourist prediction results include the expected number of tourists and activity distribution in each partition in the predetermined future time zone, while the lighting data reflects the current lighting parameters such as brightness, color temperature, power status, etc. The lighting adaptability detection module calculates the lighting adaptability of each partition by matching the lighting parameters with the tourist prediction results. The lighting adaptability index reflects whether the current lighting settings are suitable for the needs of future tourist activities, such as whether the partitions with denser traffic have sufficient brightness, or whether there is excessive lighting in the partitions with low traffic. According to the lighting adaptability, the partitions that do not meet the target adaptability requirements are screened out as the lighting partitions to be adjusted. The lighting partitions to be adjusted usually show a significant gap between the tourist prediction results and the current lighting settings, such as insufficient brightness in the partition, excessive lighting, or color temperature that does not meet the expected color temperature.

[0030] Furthermore, based on the tourist prediction results of the M partitions and the lighting data of the M partitions, the lighting adaptability detection module is used to perform lighting adaptability detection on the M museum partitions to determine the lighting partitions to be adjusted, including:

[0031] The lighting adaptability detection module includes a lighting adaptability calculation module and a lighting adaptability judgment module; according to the M partition lighting data, the mth partition lighting data corresponding to the mth museum partition is extracted, and energy consumption analysis is performed based on the mth partition lighting data to determine the lighting energy consumption of the mth partition; based on the predetermined future time zone, the sunlight characteristics of the mth museum partition are predicted to determine the predicted sunlight information of the mth partition; the mth partition tourist prediction result, the mth partition lighting data, the mth partition lighting energy consumption and the mth partition predicted sunlight information are input into the lighting adaptability calculation module to obtain the lighting adaptability of the mth partition; the mth partition lighting adaptability is input into the lighting adaptability judgment module, wherein the lighting adaptability judgment module includes a predetermined lighting adaptability; if the mth partition lighting adaptability is less than / equal to the predetermined lighting adaptability, the mth museum partition is added to the lighting partition to be adjusted.

[0032] The lighting adaptability detection module includes a lighting adaptability calculation module and a lighting adaptability judgment module. The lighting adaptability calculation module is responsible for calculating the lighting adaptability of each museum partition based on input data (including visitor prediction results, lighting data, energy consumption analysis and sunshine prediction information). The lighting adaptability judgment module receives the output results of the lighting adaptability calculation module and judges whether each partition needs to adjust the lighting based on a predetermined lighting adaptability threshold.

[0033] Extract the mth partition lighting data corresponding to the mth museum partition from the M partition lighting data; perform energy consumption analysis based on the mth partition lighting data to determine the lighting energy consumption of the partition, including the total energy consumption, energy consumption distribution, etc.; use meteorological data, historical sunshine records and other information, combined with the time range of the predetermined future time zone, to predict the sunshine characteristics of the mth museum partition and determine the sunshine information (such as sunshine intensity, duration, etc.) of the mth partition in the future time zone; integrate the mth partition tourist prediction result, the mth partition lighting data, the mth partition lighting energy consumption and the mth partition predicted sunshine information to form a complete input data set, input the integrated input data set into the lighting fitness calculation module, and the lighting fitness calculation module calculates the lighting fitness of the mth partition based on these data; in the lighting fitness judgment module, set a predetermined lighting fitness threshold, which is used to judge whether each partition needs to adjust the lighting; compare the lighting fitness score of the mth partition with the predetermined lighting fitness threshold; if the lighting fitness score of the mth partition is less than or equal to the predetermined lighting fitness threshold, add the partition to the list of lighting partitions to be adjusted.

[0034] Furthermore, the tourist prediction result of the mth partition, the lighting data of the mth partition, the lighting energy consumption of the mth partition and the predicted sunshine information of the mth partition are input into the lighting adaptability calculation module to obtain the lighting adaptability of the mth partition, including:

[0035] The lighting fitness calculation module includes a lighting state fitness analytical model, a lighting energy consumption fitness analytical model and a lighting fitness weighted model; the tourist prediction result of the m-th partition, the lighting data of the m-th partition and the predicted sunshine information of the m-th partition are input into the lighting state fitness analytical model to obtain the m-th lighting state fitness; the tourist prediction result of the m-th partition, the lighting energy consumption of the m-th partition and the predicted sunshine information of the m-th partition are input into the lighting energy consumption fitness analytical model to obtain the m-th lighting energy consumption fitness; the m-th lighting state fitness and the m-th lighting energy consumption fitness are input into the lighting fitness weighted model to output the lighting fitness of the m-th partition.

[0036] The lighting fitness calculation module includes a lighting state fitness analytical model, a lighting energy consumption fitness analytical model and a lighting fitness weighted model; the lighting state fitness analytical model is responsible for evaluating the degree of match between the lighting state and tourist needs. The lighting state fitness analytical model receives the tourist prediction results of the mth partition, the lighting data of the mth partition and the predicted sunshine information of the mth partition as input, and then calculates the fitness of the mth lighting state based on the number of tourists, visit time, stay time, lighting brightness, color temperature and sunshine intensity; the lighting energy consumption fitness analytical model is responsible for evaluating the rationality of the current lighting energy consumption. The lighting energy consumption fitness analytical model The tourist prediction result of the m-th zone, the lighting energy consumption of the m-th zone and the predicted sunshine information of the m-th zone are received as inputs, and the m-th lighting energy consumption adaptability is calculated by comprehensively considering the tourist needs, lighting energy consumption and sunshine conditions, so as to reflect whether the energy consumption of the lighting system is at a reasonable level while meeting the tourist needs; the lighting fitness weighted model is responsible for integrating the lighting state fitness and the lighting energy consumption fitness to obtain the final lighting fitness. The lighting fitness weighted model receives the m-th lighting state fitness and the m-th lighting energy consumption fitness as inputs, and obtains the lighting fitness of the m-th zone through weighted calculation according to the preset weight distribution.

[0037] Collect a large amount of data about lighting conditions, including lighting brightness, color temperature, color rendering index and other parameters, extract features related to the fitness of the lighting conditions from the collected data, such as light intensity distribution, light uniformity, light source stability, etc., and use machine learning models (such as neural networks) for training. The input of the model is the extracted features, and the output is the fitness index of the lighting condition. After the training is completed, the lighting condition fitness analysis model is obtained.

[0038] Collect energy consumption data of the lighting system, including energy consumption in different time periods and under different lighting conditions; analyze the features in the energy consumption data, such as the power of the lighting equipment, usage time, light intensity, etc.; use machine learning to build an energy consumption fitness analysis model. The input of the model is the energy consumption characteristics, and the output is the energy consumption fitness index. The model is trained with a training data set to find the optimal parameters and model structure; after the training is completed, the lighting energy consumption fitness analysis model is obtained.

[0039] The tourist prediction results of the mth partition, the lighting data of the mth partition, and the predicted sunshine information of the mth partition are input into the lighting state fitness analytical model. The lighting state fitness analytical model evaluates the state of the lighting system based on the tourists' visiting behavior and sunshine conditions, and calculates the mth lighting state fitness. Similarly, the tourist prediction results of the mth partition, the lighting energy consumption of the mth partition, and the predicted sunshine information of the mth partition are input into the lighting energy consumption fitness analytical model. The lighting energy consumption fitness analytical model comprehensively considers the tourists' needs and lighting energy consumption, as well as the influence of sunshine conditions on energy consumption, and calculates the mth lighting energy consumption fitness. The mth lighting state fitness and the mth lighting energy consumption fitness are input into the lighting fitness weighted model, which obtains the lighting fitness of the mth partition through weighted calculation according to the preset weights.

[0040] Based on the lighting partition to be adjusted, a lighting expectation constraint decision is made, a lighting strategy space is established, and energy consumption minimization is optimized according to the lighting strategy space to generate a lighting adjustment plan.

[0041] According to the future visit demand, predicted sunshine information and current lighting adaptability of each partition to be adjusted, the corresponding lighting expectation constraints are set. These constraints include the minimum brightness, color temperature requirements, coverage, etc. that need to be achieved in the partition to ensure that the adjusted lighting effect can meet the visitor's needs. Based on the lighting expectation constraints, a lighting strategy space is constructed, that is, all lighting configuration combinations allowed under the premise of meeting the expectation constraints. The configurations in the strategy space may include brightness level, switch status, color temperature adjustment, combinations of different lamps, etc. Each combination represents a different energy consumption level and lighting effect. By listing and analyzing these combinations, multiple lighting solutions that meet the needs can be found for each partition.

[0042] Energy consumption is minimized and optimized within the lighting strategy space to find the optimal configuration that meets the lighting expectations and can effectively save energy. The optimization process can be implemented through algorithms, such as genetic algorithms, simulated annealing, particle swarm optimization, etc., to gradually screen out the solution that achieves the best balance between lighting fitness and energy consumption. The optimal configuration generated by optimization is used as the final lighting adjustment solution, and the specific lighting adjustment parameters of each partition to be adjusted are clarified, including brightness, color temperature, switch status, etc.

[0043] Furthermore, based on the lighting partition to be adjusted, a lighting expectation constraint decision is made to establish a lighting strategy space, including:

[0044] Perform an expected analysis of the lighting state according to the lighting partition to be adjusted to determine the expected lighting state; load the lighting configuration constraint information of the lighting partition to be adjusted; perform random lighting decisions on the lighting partition to be adjusted according to the lighting configuration constraint information to obtain a lighting decision space; based on the expected lighting state, perform an optimization analysis on the lighting decision space to generate the lighting strategy space, wherein the number of lighting strategies in the lighting strategy space satisfies the lighting strategy number constraint.

[0045] Specifically, the lighting state expectation analysis is performed on the needs of the lighting partition to be adjusted to determine the expected lighting state of the partition. The expected lighting state includes specific lighting targets such as illuminance, color temperature, and uniformity to ensure that the adjusted lighting can meet the visitor needs of the partition and achieve the ideal visual effect. After analyzing the expected lighting state, the lighting configuration constraint information of the partition is loaded. These constraint information include the physical limitations, power limitations, dimming range, switch status, etc. of the lighting equipment. Based on the loaded lighting configuration constraint information, a series of feasible lighting configuration combinations are generated for the lighting partition to be adjusted by a randomization method to form a lighting decision space. Each combination represents a feasible lighting configuration. Through a variety of configuration options, it covers possible lighting settings and provides a sufficient selection basis for subsequent optimization analysis. In the generated lighting decision space, an optimization analysis is performed based on the expected lighting state to screen out configurations that meet the expected state and are superior in energy consumption. Through the lighting strategy quantity constraint, the number of strategies in the final lighting strategy space is limited to ensure computational efficiency and diversity of choices.

[0046] Furthermore, based on the desired lighting state, performing optimization analysis on the lighting decision space to generate the lighting strategy space includes:

[0047] According to the lighting decision space, a first lighting decision is randomly extracted; based on the first lighting decision, a lighting state of the lighting partition to be adjusted is predicted to obtain a first predicted lighting state; whether the first predicted lighting state satisfies the expected lighting state is determined; if the first predicted lighting state satisfies the expected lighting state, the first lighting decision is set as a first lighting strategy, and the first lighting strategy is added to the lighting strategy space; according to the expected lighting state, the lighting decision space is continued to be optimized and analyzed to construct the lighting strategy space.

[0048] In the lighting decision space, a lighting decision is randomly extracted as the current candidate configuration, which is called the first lighting decision. The first lighting decision contains a specific lighting parameter combination, such as brightness, color temperature, and enabled state. Based on the first lighting decision, the lighting state of the lighting partition to be adjusted is predicted to obtain a first predicted lighting state. The first predicted lighting state refers to the actual lighting effect of the first lighting decision in the future time zone obtained through simulation, including parameters such as lighting uniformity, color temperature, and coverage. It is judged whether the first predicted lighting state meets the expected lighting state. If all parameters of the first predicted lighting state meet or exceed the expected setting requirements, it is considered that the lighting decision meets the lighting needs and is suitable as a candidate strategy. If the first predicted lighting state meets the expected lighting state, the first lighting decision is set as the first lighting strategy, and the strategy is added to the lighting strategy space as one of the configuration solutions that meet the requirements. According to the expected lighting state, the above steps are repeated to continue to extract candidate decisions in the lighting decision space for lighting state prediction and verification, and gradually build the lighting strategy space. The construction process of the space will continue until the preset lighting strategy quantity constraint is met or the analysis of all feasible lighting decisions is completed.

[0049] Furthermore, performing an expected analysis of the lighting state according to the lighting partition to be adjusted to determine the expected lighting state includes:

[0050] According to the partition tourist prediction results and partition predicted sunlight information corresponding to the lighting partition to be adjusted, the lighting scene condition information is constructed; based on the lighting partition to be adjusted, lighting record retrieval is performed to obtain a lighting scene condition information record set and an adaptive lighting state record set; based on the lighting scene condition information record set and the adaptive lighting state record set, integrated fusion learning is performed to establish a lighting state expected prediction model; the lighting scene condition information is input into the lighting state expected prediction model to obtain the expected lighting state.

[0051] According to the tourist prediction results of the lighting partition to be adjusted and the predicted sunshine information of the partition, the lighting scene condition information is generated. The information includes the predicted number of tourists, distribution location, sunshine intensity and expected light coverage, etc., which comprehensively reflects the actual lighting demand and external light conditions of the partition in the future time zone. Based on the historical lighting control data of the partition to be adjusted, similar lighting scene condition information record sets and adaptive lighting state record sets under these conditions are retrieved. The adaptive lighting state record set includes multiple historical lighting state information greater than the predetermined lighting adaptability, and these multiple historical lighting state information correspond one to one with multiple historical lighting scene condition information in the lighting scene condition information record set. The lighting scene condition information record set and the adaptive lighting state record set are input into the integrated fusion learning model, and trained by machine learning methods (such as deep learning or integrated learning algorithms) to establish a lighting state expectation prediction model. The lighting state expectation prediction model can identify the optimal lighting configuration mode under different scene conditions and form the ability to predict the expected lighting state. The current lighting scene condition information is input into the lighting state expectation prediction model. The model generates the expected lighting state based on the knowledge obtained through training. The expected lighting state includes specific lighting parameters (such as brightness, color temperature, and distribution uniformity) to meet the visiting needs and ambient light conditions in the future time zone.

[0052] Based on the museum lighting system, lighting control is performed on the lighting partition to be adjusted according to the lighting adjustment scheme.

[0053] The optimized lighting adjustment scheme is input into the museum's lighting control system. The lighting adjustment scheme contains detailed lighting parameter configurations for the partitions to be adjusted, such as brightness, color temperature, switch status, etc., to ensure that the lighting settings of each partition meet the expected lighting status and energy consumption optimization goals. The museum lighting system adjusts the lighting status of the partitions to be adjusted one by one according to the adjustment scheme. The system adjusts the light intensity, color temperature and opening time of each partition in real time according to the set parameters to ensure that the light in each area meets the needs of tourists and minimizes unnecessary energy consumption.

[0054] In summary, the embodiments of the present application have at least the following technical effects:

[0055] First, according to the sensor network, the tourist behavior parameters of the museum are collected in real time to obtain the tourist monitoring results. Then, the museum is divided according to the museum layout diagram to obtain M museum partitions, and the lighting parameters of the M museum partitions are collected in real time according to the museum lighting system to obtain M partition lighting data, where M is a positive integer greater than 1. Next, based on the predetermined future time zone, the M museum partitions are predicted for the tourist visiting behavior according to the tourist monitoring results to obtain the M partition tourist prediction results. Then, based on the M partition tourist prediction results and the M partition lighting data, the M museum partitions are tested for lighting adaptability according to the lighting adaptability detection module to determine the lighting partitions to be adjusted. Further, based on the lighting partitions to be adjusted, the lighting expectation constraint decision is made, the lighting strategy space is established, and the energy consumption is minimized and optimized according to the lighting strategy space to generate the lighting adjustment plan. Finally, based on the museum lighting system, the lighting control is performed on the lighting partitions to be adjusted according to the lighting adjustment plan. The technical problem that the museum lighting system lacks dynamic lighting adjustment in the prior art, resulting in energy waste, is solved. By predicting the tourist visiting behavior and dynamically adjusting the lighting of each partition of the museum accordingly, the technical effect of energy saving and consumption reduction is achieved.

[0056] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0058] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A museum lighting system energy-saving control method based on visitor behavior prediction, characterized in that: The method comprises: Based on the sensor network, the museum's visitor behavior parameters are collected in real time to obtain visitor monitoring results; Divide the museum according to the museum layout diagram to obtain M museum partitions, and collect lighting parameters of the M museum partitions in real time according to the museum lighting system to obtain M partition lighting data, where M is a positive integer greater than 1; Based on the predetermined future time zone, tourist visit behaviors of the M museum partitions are predicted according to the tourist monitoring results to obtain tourist prediction results of the M partitions, including: According to the M museum partitions, extract the mth museum partition, where m is a positive integer; Performing feature recognition on the visitor monitoring result based on the mth museum partition to determine the visitor monitoring result of the mth partition; Based on the predetermined future time zone, predicting the visitor behavior of the mth museum partition according to the mth partition visitor monitoring result, and obtaining a first partition visitor prediction result; Based on the predetermined future time zone, the mth museum partition is subjected to the same-zone tourist visit behavior mining to obtain a second partition tourist prediction result; The first result of the zoned tourist prediction and the second result of the zoned tourist prediction are integrated to generate an m-th zoned tourist prediction result, and the m-th zoned tourist prediction result is added to the M zoned tourist prediction results; based on the M zoned tourist prediction results and the M zoned lighting data, the lighting adaptability detection module is used to perform a lighting adaptability detection on the M museum zones to determine the lighting zones to be adjusted; Based on the lighting partition to be adjusted, a lighting expectation constraint decision is made, a lighting strategy space is established, and energy consumption is minimized and optimized according to the lighting strategy space to generate a lighting adjustment plan; Based on the museum lighting system, lighting control is performed on the lighting partition to be adjusted according to the lighting adjustment scheme.

2. The energy-saving control method for a museum lighting system combined with prediction of visitor behavior according to claim 1, characterized in that: Based on the predetermined future time zone, the mth museum partition is subjected to the same-zone tourist visit behavior mining to obtain a second partition tourist prediction result, including: According to the predetermined future time zone, setting a plurality of historical time zones with the same characteristics; Based on the multiple historical time zones with the same characteristics, query the historical visitor behavior of the mth museum partition to obtain multiple historical visitor behavior query results; A central tendency analysis is performed based on the plurality of historical tourist behavior query results to generate the second result of the zoned tourist prediction.

3. The energy-saving control method for a museum lighting system combined with prediction of visitor behavior according to claim 1, characterized in that: Based on the tourist prediction results of the M partitions and the lighting data of the M partitions, a lighting adaptability detection module is used to perform lighting adaptability detection on the M museum partitions to determine the lighting partitions to be adjusted, including: The lighting adaptability detection module includes a lighting adaptability calculation module and a lighting adaptability judgment module; Extract the mth partition lighting data corresponding to the mth museum partition according to the M partition lighting data, and perform energy consumption analysis according to the mth partition lighting data to determine the lighting energy consumption of the mth partition; Predicting the sunshine characteristics of the mth museum partition based on the predetermined future time zone, and determining the predicted sunshine information of the mth partition; Inputting the tourist prediction result of the mth partition, the lighting data of the mth partition, the lighting energy consumption of the mth partition and the predicted sunshine information of the mth partition into the lighting adaptability calculation module to obtain the lighting adaptability of the mth partition; Inputting the illumination adaptability of the mth partition into the illumination adaptability judgment module, wherein the illumination adaptability judgment module includes a predetermined illumination adaptability; If the lighting adaptability of the m-th zone is less than / equal to the predetermined lighting adaptability, the m-th museum zone is added to the lighting zones to be adjusted.

4. The energy-saving control method for a museum lighting system combined with prediction of visitor behavior according to claim 3, characterized in that: Inputting the tourist prediction result of the m-th partition, the lighting data of the m-th partition, the lighting energy consumption of the m-th partition and the predicted sunshine information of the m-th partition into the lighting adaptability calculation module to obtain the lighting adaptability of the m-th partition, including: The lighting adaptability calculation module includes a lighting state adaptability analytical model, a lighting energy consumption adaptability analytical model and a lighting adaptability weighted model; Inputting the tourist prediction result of the m-th partition, the lighting data of the m-th partition and the predicted sunshine information of the m-th partition into the lighting state fitness analytical model to obtain the m-th lighting state fitness; Inputting the tourist prediction result of the m-th subarea, the lighting energy consumption of the m-th subarea and the predicted sunshine information of the m-th subarea into the lighting energy consumption adaptability analytical model to obtain the m-th lighting energy consumption adaptability; The mth lighting state fitness and the mth lighting energy consumption fitness are input into the lighting fitness weighting model, and the mth partition lighting fitness is output.

5. The energy-saving control method for a museum lighting system combined with prediction of visitor behavior according to claim 1, characterized in that: Based on the lighting partition to be adjusted, a lighting expectation constraint decision is made to establish a lighting strategy space, including: Performing an expected analysis of the lighting state according to the lighting partition to be adjusted to determine the expected lighting state; Loading the lighting configuration constraint information of the lighting partition to be adjusted; Performing random lighting decisions on the lighting partition to be adjusted according to the lighting configuration constraint information to obtain a lighting decision space; Based on the expected lighting state, the lighting decision space is optimized and analyzed to generate the lighting strategy space, wherein the number of lighting strategies in the lighting strategy space satisfies a lighting strategy number constraint.

6. The energy-saving control method for a museum lighting system combined with prediction of visitor behavior according to claim 5, characterized in that: Based on the desired lighting state, performing optimization analysis on the lighting decision space to generate the lighting strategy space includes: According to the lighting decision space, randomly extracting a first lighting decision; Predicting the lighting state of the lighting partition to be adjusted based on the first lighting decision to obtain a first predicted lighting state; determining whether the first predicted lighting state satisfies the expected lighting state; If the first predicted lighting state satisfies the expected lighting state, setting the first lighting decision as a first lighting strategy, and adding the first lighting strategy to the lighting strategy space; According to the desired lighting state, the lighting decision space is continuously optimized and analyzed to construct the lighting strategy space.

7. The energy-saving control method for a museum lighting system combined with prediction of visitor behavior according to claim 5, characterized in that: Performing an expected analysis of the lighting state according to the lighting partition to be adjusted to determine the expected lighting state includes: Constructing lighting scene condition information according to the zone tourist prediction result and zone predicted sunshine information corresponding to the lighting zone to be adjusted; Perform lighting record retrieval based on the lighting partition to be adjusted to obtain a lighting scene condition information record set and an adapted lighting state record set; Performing integrated fusion learning based on the lighting scene condition information record set and the adapted lighting state record set to establish a lighting state expectation prediction model; The lighting scene condition information is input into the lighting state expectation prediction model to obtain the expected lighting state.

Images