A method and system for management analysis based on big data

Abstract

The present invention discloses a management and analysis method and system based on big data, which relates to the technical field of big data analysis. The method collects park lighting demand data in real time through an integrated sensor group, transmits it to a big data management platform through a wireless network, and integrates it with other data sources to generate a complete data set. Based on the lighting demand prediction algorithm model, the lighting demand coefficient Lreq(t) is calculated in real time, and the output lighting demand nonlinear coefficient Lfxx(t) is calculated in combination with the nonlinear term and the parameter interaction to refine the prediction result. In the adaptive power regulation, the system dynamically adjusts the street lamp power according to the difference between the lighting demand nonlinear coefficient Lfxx(t) and the actual output light Lout(t), to ensure proper lighting, improve energy efficiency, avoid excessive lighting, especially when there is sufficient natural light, less traffic, and related holiday needs, to achieve significant energy-saving effects.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a method and system for management analysis based on big data. Background Art

[0002] Smart cities optimize urban infrastructure and public services through information technology and intelligent management. Among the many areas of smart cities, intelligent lighting systems are an important component, especially in public places such as parks, where intelligent lighting can significantly improve energy efficiency. In the park intelligent lighting system, the intelligent control of street lights can automatically adjust the lighting intensity according to real-time environmental changes such as natural light intensity, traffic flow and population density, reduce energy consumption, improve lighting quality, and ensure the safety and comfort of citizens' nighttime activities.

[0003] The main problems facing the current park lighting system at this stage are energy waste and low lighting efficiency. Traditional lighting systems cannot dynamically adjust the lighting intensity according to real-time needs. They are usually controlled according to preset time and brightness without considering changes in environmental conditions. This fixed lighting method leads to excessive lighting in some cases, such as during the day and in clear weather, which wastes electricity resources; while in other cases, such as when the weather is gloomy or when there are dense crowds, insufficient lighting affects the user experience and safety of citizens. Although some lighting systems have preliminary automated control functions, they often lack data-driven dynamic optimization mechanisms, making it difficult to achieve intelligent real-time adjustment. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for management analysis based on big data, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented by the following technical scheme: comprising the following steps:

[0006] S1. Install a sensor group on the top of the park's buildings to collect the park's lighting demand data in real time, and build a big data management platform. At the same time, set up a big data collection tool in the big data management platform to collect nonlinear items in real time;

[0007] S2. Connect the integrated sensor group to the big data management platform using a wireless communication network, receive the park lighting demand data in real time on the big data management platform, integrate it with nonlinear items into a lighting demand data set, extract time series features from the lighting demand data set, obtain a lighting demand feature vector set, and then construct a time series database to store the lighting demand feature vector set in chronological order;

[0008] S3, construct a lighting demand prediction algorithm model, extract a lighting demand feature vector set, input it into the lighting demand prediction algorithm model, calculate and output the lighting demand coefficient Lreq(t), and then combine the nonlinear term of the lighting demand feature vector set to calculate and output the lighting demand nonlinear coefficient Lfxx(t), so as to further refine the lighting demand prediction;

[0009] S4. Based on the obtained illumination demand nonlinear coefficient Lfxx(t) and the actual output illumination value Lout(t) of the park street lamp, a preliminary comparative evaluation is performed to analyze the illumination output of the park street lamp over time, and based on the evaluation results, an adaptive power adjustment mechanism for the park street lamp is executed to output the optimized power value Padj(t) of the street lamp to perform preliminary adaptive power adjustment on the park street lamp;

[0010] S5. After the preliminary adaptive power adjustment is completed, the optimized power value Padj(t) of the street lamp, the nonlinear coefficient of the lighting demand Lfxx(t) and the actual output lighting value Lout(t) are correlated to calculate the output feedback index FBI(t), and the preset energy consumption threshold T is compared with the feedback index FBI(t) for secondary evaluation to analyze the energy consumption output of the park street lamps, and an iterative mechanism is generated based on the results of the secondary comparative evaluation.

[0011] Preferably, S1 includes S11 and S12;

[0012] S11. Install a sensor group on the top of the building in the park area, and set the collection frequency of the integrated sensor group to collect data once every hour to collect the park's lighting demand data;

[0013] The integrated sensor set includes an ambient light sensor, an ambient noise sensor, an ambient light pollution monitor, a radar speed gun, and an infrared detection device;

[0014] The park lighting demand data include sky brightness index SLI, ambient noise level NLV, light pollution index LPI, traffic speed TVS and population density PCD;

[0015] S12. Build a big data management platform and set up a big data collection tool inside the big data management platform to collect nonlinear items in real time;

[0016] Big data collection tools include API application programming interfaces and social media crawlers;

[0017] The nonlinear items include weather data WTH, holiday time data EVT, and energy consumption history data PWR;

[0018] Weather data WTH and energy consumption history data PWR are integrated with the third-party meteorological data platform and the park energy consumption management system EMS through the API application program interface for real-time collection and acquisition;

[0019] Holiday time data EVT crawls keywords and geographic location information on social media through social media crawlers, and then uses natural language processing and big data analysis technology to predict emergencies that affect lighting demand.

[0020] Preferably, S2 includes S21, S22 and S23;

[0021] S21. The integrated sensor group is integrated and connected with the big data management platform through NB-loT wireless communication technology, and the park lighting demand data collected by the integrated sensor group is transmitted to the big data management platform through the MQTT data transmission protocol;

[0022] S22. In the big data management platform, the park lighting demand data is collected and received in real time, and integrated with the nonlinear items in the big data management platform to obtain a lighting demand data set, and time series feature extraction is performed on the lighting demand data set based on the collection time to obtain a lighting demand feature vector set;

[0023] The lighting demand feature vector set includes the sky brightness index SLI(t) at time t, the ambient noise level NLV(t) at time t, the light pollution index LPI(t) at time t, the traffic speed TVS(t) at time t, the population density PCD(t) at time t, and the nonlinear term;

[0024] The nonlinear terms include the weather data WTH(t) at time t, the holiday time data EVT(t) at time t, and the energy consumption history data PWR(t) at time t;

[0025] S23. By constructing a time series database and setting a real-time data table and a historical data table in the time series database, the lighting demand feature vector set is passed through the write port of the time series database and written into the real-time data table, and the lighting demand feature vector set in the original real-time data table is automatically transferred to the historical data table, and arranged in chronological order in the historical data table.

[0026] Preferably, S3 includes S31 and S32;

[0027] S31, constructing a lighting demand prediction algorithm model, extracting a lighting demand feature vector set in a real-time data table through a write port of a time series database, inputting the lighting demand prediction algorithm model, and calculating and outputting a lighting demand coefficient Lreq(t);

[0028] The light demand coefficient Lreq(t) is calculated and output by the following light demand prediction algorithm model;

[0029] Lreq(t)=a1·SLI(t) -1 +a2·log(NLV(t))+a3·eLPI(t) +a4·TVS(t) 0.5 +a5·PCD(t);

[0030] Wherein, e represents exponential function, log represents logarithmic function, a1, a2, a3, a4 and a5 represent the preset weight values ​​of sky brightness index SLI, ambient noise level NLV, light pollution index LPI, traffic speed TVS and population density PCD respectively, and a1+a2+a3+a4+a5=1, and the specific value is set by the user, SLI(t) -1 Indicates that the sky brightness is inversely proportional to the lighting demand. The lighting demand decreases at dawn. log(NLV(t)) indicates that the effect of noise level on lighting demand is increasing, but as the noise increases, the demand growth gradually tends to ease. LPI(t) It means that the influence of light pollution index on lighting demand decreases exponentially. The stronger the external light source, the lower the lighting demand. TVS(t) 0.5 It means that the faster the vehicle speed, the slower the increase in lighting demand, which reflects the nonlinear relationship between vehicle speed and lighting demand.

[0031] Preferably, S32, based on the obtained lighting demand coefficient Lreq(t), combined with the nonlinear terms and parameter interactions in the lighting demand feature vector set, calculate and output the lighting demand nonlinear coefficient Lfxx(t) to optimize the accuracy of the prediction;

[0032] The nonlinear coefficient of light demand Lfxx(t) is calculated and output by the following algorithm formula;

[0033] Ltxx(t)=Lreq(t)+b1·(PCD(t)·log(TVS(t)))+b2·WTH(t)+b3·EVT(t)+b4·PWR(t);

[0034] Wherein, (PCD(t) log(TVS(t))) represents the interaction between the traffic speed TVS and the population density PCD. In areas with dense traffic, even if the vehicle speed is slow, strong lighting is still required to ensure safety. b1, b2, b3 and b4 represent the preset weight values ​​of the interaction between the traffic speed TVS and the population density PCD, the weather data WTH, the holiday time data EVT and the energy consumption history data PWR, respectively. And b1+b2+b3+b4=1, and the specific values ​​are set by the user.

[0035] Preferably, S4 includes S41 and S42;

[0036] S41. Use the built-in light intensity sensor of the park street lamp to detect and obtain the actual output light Lout(t) of the park street lamp at the current moment in real time, and then make a preliminary comparative evaluation of the actual output light Lout(t) and the obtained nonlinear coefficient of light demand Lfxx(t), and analyze the current light output of the park street lamp and the predicted light demand. The specific evaluation contents are as follows;

[0037] When the actual output light Lout(t)> the nonlinear coefficient of light demand Lfxx(t), it means that the park road lighting has excessive lighting, and the adaptive adjustment mechanism is executed at this time;

[0038] When the nonlinear coefficient of light demand Lfxx(t) represents the expected effect of lighting load on park roads, no adaptive power regulation is required;

[0039] When the actual output light Lout(t) is less than the nonlinear coefficient of light demand Lfxx(t), it indicates that the lighting of the park road is insufficient, and the adaptive adjustment mechanism is executed at this time.

[0040] Preferably, S42, when the street lamps in the park area are preliminarily compared and evaluated to have over-lighting and under-lighting, the adaptive power adjustment mechanism is automatically executed. The adaptive power adjustment mechanism calculates the output street lamp optimized power value Padj(t) by introducing the illumination deviation between the illumination demand nonlinear coefficient Lfxx(t) and the actual output illumination value Lout(t) of the park street lamp, and preliminarily automatically adjusts the park street lamp power;

[0041] The optimized power value of the street lamp Padj(t) is calculated and output by the following algorithm formula;

[0042]

[0043] In the formula, Pmax represents the upper limit of street lamp power output, Lmax represents the upper limit of street lamp illumination intensity, and γ represents the efficiency adjustment coefficient, which is used to control the nonlinear change of power output and ensure the balance between power and illumination requirements. Indicates lighting deviation.

[0044] Preferably, S5 includes S51 and S52;

[0045] S51, based on the street lamp optimization power value Padj(t) obtained by preliminary automatic adjustment of the park street lamp power, combined with the illumination deviation, perform a related calculation to output the feedback index FBI(t), and analyze the comprehensive energy consumption of the park street lamp after the preliminary adaptive power adjustment;

[0046] The feedback indicator FBI(t) is calculated and output by the following algorithm formula;

[0047]

[0048] In the formula, λ represents the energy consumption weight coefficient, and its specific value is set by the user.

[0049] Preferably, S52, based on the user's preset energy consumption threshold T in combination with the park's energy consumption standard, and then compared with the obtained feedback index FBI(t), a secondary comparative evaluation is performed to analyze the energy consumption of the park street lamps after the preliminary adaptive power adjustment, and an iterative mechanism is generated based on the secondary comparative evaluation results. The specific evaluation results are as follows;

[0050] When the feedback index FBI(t)≤energy consumption threshold T, it means that the energy consumption of the park street lamp is normal. At this time, no adjustment is required and continuous monitoring is required;

[0051] When the feedback indicator FBI(t)>energy consumption threshold T, it means that the energy consumption of the park street lamp is abnormal. At this time, the adaptive power adjustment mechanism is iteratively executed until the energy consumption returns to normal and the iteration is stopped.

[0052] A management and analysis system based on big data, including a lighting demand acquisition module, a data transmission and feature extraction module, a lighting demand prediction and analysis module, an energy efficiency optimization power output module, and an energy consumption feedback and analysis module;

[0053] The light demand collection module collects the park's light demand data in real time by installing a sensor group on the top of the park's buildings, and builds a big data management platform. At the same time, a big data collection tool is set up in the big data management platform to collect nonlinear items in real time;

[0054] The data transmission and feature extraction module connects the integrated sensor group with the big data management platform through the wireless communication network, receives the park lighting demand data in real time in the big data management platform, integrates it with the nonlinear terms into a lighting demand data set, extracts time series features from the lighting demand data set, obtains a lighting demand feature vector set, and then constructs a time series database to store the lighting demand feature vector set in chronological order;

[0055] The lighting demand prediction and analysis module constructs a lighting demand prediction algorithm model, extracts the lighting demand feature vector set, inputs it into the lighting demand prediction algorithm model, calculates and outputs the lighting demand coefficient Lreq(t), and then combines the nonlinear term of the lighting demand feature vector set to calculate and output the lighting demand nonlinear coefficient Lfxx(t), so as to further refine the lighting demand prediction;

[0056] The energy efficiency optimization power output module performs a preliminary comparative evaluation and analysis of the time output of the park street lamps based on the obtained nonlinear coefficient of the light demand Lfxx(t) and the actual output light value Lout(t) of the park street lamps, and based on the evaluation results, executes the park street lamp adaptive power adjustment mechanism, outputs the street lamp optimized power value Padj(t) to perform preliminary adaptive power adjustment on the park street lamps;

[0057] After the energy consumption feedback and analysis module completes the initial adaptive power adjustment, it correlates the optimized power value Padj(t) of the street lamp, the nonlinear coefficient of illumination demand Lfxx(t) and the actual output illumination value Lout(t) to calculate the output feedback index FBI(t), and performs a secondary comparative evaluation between the preset energy consumption threshold T and the feedback index FBI(t) to analyze the energy consumption output of the park street lamps, and generates an iterative mechanism based on the results of the secondary comparative evaluation.

[0058] The present invention provides a method and system for management analysis based on big data, which has the following beneficial effects:

[0059] (1) This method collects the park's lighting demand data in real time by installing a sensor group on the top of the park's buildings, and transmits this data to the big data management platform through a wireless communication network. In the big data management platform, the park's lighting demand data is integrated with other data sources to generate a complete lighting demand data set. By constructing a lighting demand prediction algorithm model, this method can calculate the output lighting demand coefficient Lreq(t) and the nonlinear coefficient of the lighting demand Lfxx(t) in real time, and further refine the prediction results. When performing adaptive power regulation, the system adjusts the power output of the street lamp based on the comparison between the current lighting demand and the actual output to ensure that the lighting intensity is just right. This ability to accurately match lighting demand with actual light output greatly improves energy efficiency, avoids over-lighting, and reduces the overall energy consumption of the park. Especially during periods of sufficient natural light or in areas with less traffic, the system can automatically reduce power to achieve significant energy-saving effects.

[0060] (2) This method builds a feedback mechanism to continuously monitor the comparison between the actual output power Lout(t) of the street lamp and the nonlinear coefficient of the lighting demand Lfxx(t), generates the feedback index FBI(t) and sets the energy consumption threshold TTT. When the feedback index FBI(t) exceeds the energy consumption threshold T, the system automatically executes the adaptive power adjustment mechanism and iteratively optimizes the power output of the street lamp until the energy consumption returns to normal. The feedback mechanism can conduct a detailed analysis of the energy consumption in each time period to ensure that the energy consumption is always at a reasonable level. In addition, through secondary comparative evaluation, the system can make fine adjustments to abnormal energy consumption or insufficient lighting, automatically generate an iterative mechanism, and continuously optimize energy consumption performance. This mechanism reduces the reliance on manual intervention and reduces the maintenance cost of the park lighting system. More importantly, it ensures that the system is in an efficient operation state for a long time, avoids the negative impact of excessive or insufficient lighting on activities and safety in the park, and ensures the sustainability and stability of the park lighting system. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a schematic diagram of the steps of a method for management analysis based on big data of the present invention;

[0062] Figure 2 The present invention is a system flow diagram of management analysis based on big data. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] Example 1

[0065] See also Figure 1 The present invention provides a management analysis method based on big data. To achieve the above purpose, the present invention is implemented by the following technical scheme: comprising the following steps:

[0066] S1. Install a sensor group on the top of the park's buildings to collect the park's lighting demand data in real time, and build a big data management platform. At the same time, set up a big data collection tool in the big data management platform to collect nonlinear items in real time;

[0067] S2. Connect the integrated sensor group to the big data management platform using a wireless communication network, receive the park lighting demand data in real time on the big data management platform, integrate it with nonlinear items into a lighting demand data set, extract time series features from the lighting demand data set, obtain a lighting demand feature vector set, and then construct a time series database to store the lighting demand feature vector set in chronological order;

[0068] S3, construct a lighting demand prediction algorithm model, extract a lighting demand feature vector set, input it into the lighting demand prediction algorithm model, calculate and output the lighting demand coefficient Lreq(t), and then combine the nonlinear term of the lighting demand feature vector set to calculate and output the lighting demand nonlinear coefficient Lfxx(t), so as to further refine the lighting demand prediction;

[0069] S4. Based on the obtained illumination demand nonlinear coefficient Lfxx(t) and the actual output illumination value Lout(t) of the park street lamp, a preliminary comparative evaluation is performed to analyze the illumination output of the park street lamp over time, and based on the evaluation results, an adaptive power adjustment mechanism for the park street lamp is executed to output the optimized power value Padj(t) of the street lamp to perform preliminary adaptive power adjustment on the park street lamp;

[0070] S5. After the preliminary adaptive power adjustment is completed, the optimized power value Padj(t) of the street lamp, the nonlinear coefficient of the lighting demand Lfxx(t) and the actual output lighting value Lout(t) are correlated to calculate the output feedback index FBI(t), and the preset energy consumption threshold T is compared with the feedback index FBI(t) for secondary evaluation to analyze the energy consumption output of the park street lamps, and an iterative mechanism is generated based on the results of the secondary comparative evaluation.

[0071] In this embodiment, the method can collect the park's lighting demand data in real time by installing a sensor group on the top of the park building and building a big data management platform in steps S1 and S2, and integrate it with other nonlinear items to generate a time series feature vector set, and store it in a time series database, providing a solid data foundation for subsequent lighting demand prediction. Subsequently, in step S3, by constructing a lighting demand prediction algorithm model, the lighting demand coefficient Lreq(t) and the lighting demand nonlinear coefficient Lfxx(t) can be accurately calculated, so as to make a detailed prediction of the lighting demand under different environments, greatly improving the accuracy of the prediction. In S4, based on the comparison between the predicted demand coefficient and the actual lighting output, the adaptive power adjustment mechanism is started to generate the optimized power Padj(t) to ensure that the lighting intensity matches the real-time demand, effectively avoiding excessive lighting and insufficient lighting. Finally, through the feedback and secondary evaluation mechanism of S5, the energy consumption performance can be continuously monitored, and the power output can be further optimized through the iterative mechanism to ensure that the energy consumption is maintained within a reasonable range. Compared with traditional lighting, this approach not only significantly reduces energy consumption and operating costs, but also improves automation, lighting quality and response efficiency through intelligent adaptive power regulation and big data-driven prediction. Through these improvements, park lighting can operate more efficiently and environmentally friendly, while providing users in the park with a more comfortable and safe nighttime environment.

[0072] Example 2

[0073] This embodiment is explained in Example 1, please refer to Figure 1 , specifically: S1 includes S11 and S12;

[0074] S11. Install a sensor group on the top of the building in the park area, and set the collection frequency of the integrated sensor group to collect once every hour to collect the park's lighting demand data;

[0075] The integrated sensor set includes an ambient light sensor, an ambient noise sensor, an ambient light pollution monitor, a radar speed gun, and an infrared detection device;

[0076] The park lighting demand data include sky brightness index SLI, ambient noise level NLV, light pollution index LPI, traffic speed TVS and population density PCD;

[0077] S12. Build a big data management platform and set up a big data collection tool inside the big data management platform to collect nonlinear items in real time;

[0078] Big data collection tools include API application programming interfaces and social media crawlers;

[0079] The nonlinear items include weather data WTH, holiday time data EVT, and energy consumption history data PWR;

[0080] Weather data WTH and energy consumption history data PWR are integrated with the third-party meteorological data platform and the park energy consumption management system EMS through the API application program interface for real-time collection and acquisition;

[0081] Holiday time data EVT crawls keywords and geographic location information on social media through social media crawlers, and then uses natural language processing and big data analysis technology to predict emergencies that affect lighting demand.

[0082] In this embodiment, the method ensures the real-time collection of the park lighting demand data in the park by installing an integrated sensor group in S11. By automatically collecting data once every hour, the latest lighting demand information can be continuously obtained to provide accurate data support for subsequent prediction and adjustment. In S12, by building a big data management platform and setting up big data collection tools, nonlinear items can be integrated in real time, further enriching the dimension of lighting demand analysis. The API application program interface makes it possible to seamlessly integrate with third-party meteorological data platforms and park energy consumption management, while social media crawlers mine emergency data related to lighting needs through natural language processing technology, thereby improving the responsiveness. Compared with traditional static lighting control technology, this method realizes dynamic, data-driven lighting demand prediction and adjustment by integrating multiple sensors and data sources, which not only improves energy efficiency and lighting quality, but also has the flexibility to deal with emergencies. Ultimately, the method significantly reduces energy waste, optimizes lighting output, and improves the overall user experience and safety of the park.

[0083] Example 3

[0084] This embodiment is explained in Example 2. Please refer to Figure 1 , specifically: S2 includes S21, S22 and S23;

[0085] S21. The integrated sensor group is integrated and connected with the big data management platform through NB-loT wireless communication technology, and the park lighting demand data collected by the integrated sensor group is transmitted to the big data management platform through the MQTT data transmission protocol;

[0086] S22. In the big data management platform, the park lighting demand data is collected and received in real time, and integrated with the nonlinear items in the big data management platform to obtain a lighting demand data set, and time series feature extraction is performed on the lighting demand data set based on the collection time to obtain a lighting demand feature vector set;

[0087] The lighting demand feature vector set includes the sky brightness index SLI(t) at time t, the ambient noise level NLV(t) at time t, the light pollution index LPI(t) at time t, the traffic speed TVS(t) at time t, the population density PCD(t) at time t, and the nonlinear term;

[0088] The nonlinear terms include the weather data WTH(t) at time t, the holiday time data EVT(t) at time t, and the energy consumption history data PWR(t) at time t;

[0089] S23. By constructing a time series database and setting a real-time data table and a historical data table in the time series database, the lighting demand feature vector set is passed through the write port of the time series database and written into the real-time data table, and the lighting demand feature vector set in the original real-time data table is automatically transferred to the historical data table, and arranged in chronological order in the historical data table.

[0090] In this embodiment, the method uses NB-IoT wireless communication technology to connect the integrated sensor group with the big data management platform in S21, and uses the MQTT protocol for data transmission, thereby ensuring low-power and high-efficiency data communication, and ensuring real-time uploading and reliable reception of lighting demand data. Then, in S22, the collected park lighting demand data is integrated with nonlinear factors to generate a comprehensive lighting demand data set. By extracting the time series features of the data set, the lighting demand feature vector set at each time point can be obtained, thereby achieving an accurate characterization of the lighting demand. In S23, the construction of a time series database makes data storage and management more efficient. Through the separate management of real-time data tables and historical data tables, current data can be processed in real time, while retaining historical data for subsequent analysis and optimization. This data management method provides the ability to track lighting demand over a long period of time, which facilitates subsequent prediction and tuning. Compared with traditional lighting, this method achieves more accurate data processing and decision support through real-time wireless communication, nonlinear feature integration and time series management. Compared with static or semi-dynamic data processing methods, this method not only improves the reliability and speed of data transmission, but also significantly enhances the ability to handle complex environmental variables, improves the intelligence level and energy efficiency optimization capabilities of park lighting, and ultimately achieves more precise lighting control and reduces energy consumption and operating costs.

[0091] Example 4

[0092] This embodiment is explained in Example 3, please refer to Figure 1 , specifically: S3 includes S31 and S32;

[0093] S31, constructing a lighting demand prediction algorithm model, extracting a lighting demand feature vector set in a real-time data table through a write port of a time series database, inputting the lighting demand prediction algorithm model, and calculating and outputting a lighting demand coefficient Lreq(t);

[0094] The light demand coefficient Lreq(t) is calculated and output by the following light demand prediction algorithm model;

[0095] Lreq(t)=a1·SLI(t) -1 +a2·log(NLV(t))+a3·e LPI(t) +a4·TVS(t) 0.5 +a5·PCD(t);

[0096] Wherein, e represents exponential function, log represents logarithmic function, a1, a2, a3, a4 and a5 represent the preset weight values ​​of sky brightness index SLI, ambient noise level NLV, light pollution index LPI, traffic speed TVS and population density PCD respectively, and a1+a2+a3+a4+a5=1, and the specific value is set by the user, SLI(t) -1 Indicates that the sky brightness is inversely proportional to the lighting demand. The lighting demand decreases at dawn. log(NLV(t)) indicates that the effect of noise level on lighting demand is increasing, but as the noise increases, the demand growth gradually tends to ease. LPI(t) It means that the influence of light pollution index on lighting demand decreases exponentially. The stronger the external light source, the lower the lighting demand. TVS(t) 0.5 It means that the faster the vehicle speed, the slower the increase in lighting demand, which reflects the nonlinear relationship between vehicle speed and lighting demand.

[0097] S32, based on the obtained lighting demand coefficient Lreq(t), combined with the nonlinear terms and parameter interactions in the lighting demand feature vector set, calculate and output the lighting demand nonlinear coefficient Lfxx(t) to optimize the accuracy of the prediction;

[0098] The nonlinear coefficient of light demand Lfxx(t) is calculated and output by the following algorithm formula;

[0099] Ltxx(t)=Lreq(t)+b1·(PCD(t)·log(TVS(t)))+b2·WTH(t)+b3·EVT(t)+b4·PWR(t);

[0100] Wherein, (PCD(t) log(TVS(t))) represents the interaction between the traffic speed TVS and the population density PCD. In areas with dense traffic, even if the vehicle speed is slow, strong lighting is still required to ensure safety. b1, b2, b3 and b4 represent the preset weight values ​​of the interaction between the traffic speed TVS and the population density PCD, the weather data WTH, the holiday time data EVT and the energy consumption history data PWR, respectively. And b1+b2+b3+b4=1, and the specific values ​​are set by the user.

[0101] In this embodiment, the method extracts the lighting demand feature vector set from the write port of the time series database in S31, inputs the lighting demand prediction algorithm model, and calculates the lighting demand coefficient Lreq(t). And according to the weight value set by the user, it accurately reflects the different effects of these factors on the lighting demand, especially the processing of nonlinear relationships, so that the prediction is more precise and reasonable. Then, in S32, the lighting demand nonlinear coefficient Lfxx(t) is further calculated by combining the nonlinear terms and parameter interactions, and the prediction accuracy of the model for lighting demand in complex scenes is optimized. Compared with the traditional simple linear model, this method significantly improves the accuracy and adaptability of lighting demand prediction by introducing nonlinear calculation and interaction of multidimensional environmental parameters. Through the accurate lighting demand coefficient Lreq(t) and lighting demand nonlinear coefficient Lfxx(t), it can more flexibly adapt to dynamically changing environmental conditions, especially in areas with large traffic or dense flow of people, ensuring accurate prediction of lighting demand. This improvement not only reduces the problem of over-lighting or under-lighting, but also improves the responsiveness to complex environments, greatly enhances the intelligence level and energy efficiency performance of lighting management, and ultimately achieves more efficient energy use and a better lighting experience.

[0102] Example 5

[0103] This embodiment is explained in Example 4. Please refer to Figure 1 , specifically: S4 includes S41 and S42;

[0104] S41. Use the built-in light intensity sensor of the park street lamp to detect and obtain the actual output light Lout(t) of the park street lamp at the current moment in real time, and then make a preliminary comparative evaluation of the actual output light Lout(t) and the obtained nonlinear coefficient of light demand Lfxx(t), and analyze the current light output of the park street lamp and the predicted light demand. The specific evaluation contents are as follows;

[0105] When the actual output light Lout(t)> the nonlinear coefficient of light demand Lfxx(t), it means that the park road lighting has excessive lighting, and the adaptive adjustment mechanism is executed at this time;

[0106] When the nonlinear coefficient of light demand Lfxx(t) represents the expected effect of lighting load on park roads, no adaptive power regulation is required;

[0107] When the actual output light Lout(t) is less than the nonlinear coefficient of light demand Lfxx(t), it indicates that the lighting of the park road is insufficient, and the adaptive adjustment mechanism is executed at this time.

[0108] S42, when the preliminary comparison and evaluation shows that the street lamps in the park area have over-lighting and under-lighting, the adaptive power adjustment mechanism is automatically executed. The adaptive power adjustment mechanism calculates the output street lamp optimized power value Padj(t) by introducing the illumination deviation between the illumination demand nonlinear coefficient Lfxx(t) and the actual output illumination value Lout(t) of the park street lamp, and performs preliminary automatic adjustment of the park street lamp power;

[0109] The optimized power value of the street lamp Padj(t) is calculated and output by the following algorithm formula;

[0110]

[0111] In the formula, Pmax represents the upper limit of street lamp power output, Lmax represents the upper limit of street lamp illumination intensity, and γ represents the efficiency adjustment coefficient, which is used to control the nonlinear change of power output and ensure the balance between power and illumination requirements. Indicates lighting deviation.

[0112] In this embodiment, the method monitors the actual light output Lout(t) of the street lamp in real time through the built-in light intensity sensor of the street lamp in S41, and compares and evaluates it with the nonlinear coefficient Lfxx(t) of the light demand. Through this comparison, the phenomenon of excessive lighting or insufficient lighting can be discovered in time, and the adaptive power adjustment mechanism is automatically triggered. When excessive lighting or insufficient lighting is detected, adjustments are made immediately to avoid energy waste or insufficient lighting problems. In S42, based on the light deviation, the optimized power Padj(t) is calculated, and the power output of the street lamp is dynamically adjusted through the adaptive power adjustment mechanism to ensure that the lighting intensity is always accurately matched with the demand. Compared with traditional fixed power lighting, this method significantly improves the accuracy and response speed of lighting through real-time monitoring and automatic adjustment. The adaptive adjustment mechanism enables flexible response to various environmental changes, ensures that the light intensity in each time period is in the best state, and avoids energy waste caused by excessive lighting and safety hazards caused by insufficient lighting. Through this intelligent adjustment, the method not only improves energy utilization efficiency, but also optimizes the user experience, especially by providing appropriate lighting intensity during both busy and quiet hours, significantly improving the overall efficiency and environmental protection of park lighting.

[0113] Example 6

[0114] This embodiment is explained in Example 2. Please refer to Figure 1 , specifically: S5 includes S51 and S52;

[0115] S51, based on the street lamp optimization power value Padj(t) obtained by preliminary automatic adjustment of the park street lamp power, combined with the illumination deviation, perform a related calculation to output the feedback index FBI(t), and analyze the comprehensive energy consumption of the park street lamp after the preliminary adaptive power adjustment;

[0116] The feedback indicator FBI(t) is calculated and output by the following algorithm formula;

[0117]

[0118] In the formula, λ represents the energy consumption weight coefficient, and its specific value is set by the user.

[0119] S52, based on the preset energy consumption threshold T by the user in combination with the park energy consumption standard, and then compared with the obtained feedback index FBI(t), a secondary comparative evaluation is performed to analyze the energy consumption of the park street lamps after the initial adaptive power adjustment, and an iterative mechanism is generated based on the secondary comparative evaluation results. The specific evaluation results are as follows;

[0120] When the feedback index FBI(t)≤energy consumption threshold T, it means that the energy consumption of the park street lamp is normal. At this time, no adjustment is required and continuous monitoring is required;

[0121] When the feedback indicator FBI(t)>energy consumption threshold T, it means that the energy consumption of the park street lamp is abnormal. At this time, the adaptive power adjustment mechanism is iteratively executed until the energy consumption returns to normal and the iteration is stopped.

[0122] In this embodiment, the method calculates the feedback index FBI(t) according to the optimized power value Padj(t) of the street lamp after the preliminary adaptive power adjustment combined with the illumination deviation in S51, which is used to evaluate the comprehensive energy consumption after adjustment. Through this feedback mechanism, the system can quantitatively analyze the energy consumption performance after each adjustment, providing a basis for further optimization. In S52, the system compares and evaluates the feedback index FBI(t) with the preset energy consumption threshold T for a second time to determine whether the energy consumption is normal. If abnormal energy consumption is detected, the system automatically starts the iterative adjustment mechanism and repeatedly optimizes the power output until the energy consumption returns to normal. Compared with the fixed power or manual adjustment method of the traditional lighting system, this method realizes continuous self-adjustment and improvement through intelligent feedback and iterative optimization, significantly improving the energy efficiency management level of the lighting system. This mechanism enables the system to not only respond quickly to environmental changes, but also has the ability to optimize autonomously to ensure that energy consumption is always within a reasonable range. Through automated adjustment and optimization, the system effectively reduces manual intervention, improves management efficiency, and significantly reduces operating costs and energy waste. Ultimately, the park lighting system can achieve the goals of high efficiency, energy saving and sustainable operation while providing the best lighting effect.

[0123] Example 7

[0124] See also Figure 1 and Figure 2 , a management and analysis system based on big data, including a lighting demand acquisition module, a data transmission and feature extraction module, a lighting demand prediction and analysis module, an energy efficiency optimization power output module and an energy consumption feedback and analysis module;

[0125] The light demand collection module collects the park's light demand data in real time by installing a sensor group on the top of the park's buildings, and builds a big data management platform. At the same time, a big data collection tool is set up in the big data management platform to collect nonlinear items in real time;

[0126] The data transmission and feature extraction module connects the integrated sensor group with the big data management platform through the wireless communication network, receives the park lighting demand data in real time in the big data management platform, integrates it with the nonlinear terms into a lighting demand data set, extracts time series features from the lighting demand data set, obtains a lighting demand feature vector set, and then constructs a time series database to store the lighting demand feature vector set in chronological order;

[0127] The lighting demand prediction and analysis module constructs a lighting demand prediction algorithm model, extracts the lighting demand feature vector set, inputs it into the lighting demand prediction algorithm model, calculates and outputs the lighting demand coefficient Lreq(t), and then combines the nonlinear term of the lighting demand feature vector set to calculate and output the lighting demand nonlinear coefficient Lfxx(t), so as to further refine the lighting demand prediction;

[0128] The energy efficiency optimization power output module performs a preliminary comparative evaluation and analysis of the time output of the park street lamps based on the obtained nonlinear coefficient of the light demand Lfxx(t) and the actual output light value Lout(t) of the park street lamps, and based on the evaluation results, executes the park street lamp adaptive power adjustment mechanism, outputs the street lamp optimized power value Padj(t) to perform preliminary adaptive power adjustment on the park street lamps;

[0129] After the energy consumption feedback and analysis module completes the initial adaptive power adjustment, it correlates the optimized power value Padj(t) of the street lamp, the nonlinear coefficient of illumination demand Lfxx(t) and the actual output illumination value Lout(t) to calculate the output feedback index FBI(t), and performs a secondary comparative evaluation between the preset energy consumption threshold T and the feedback index FBI(t) to analyze the energy consumption output of the park street lamps, and generates an iterative mechanism based on the results of the secondary comparative evaluation.

[0130] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A management analysis method based on big data, characterized in that: The following steps are involved: S1. Install a sensor group on the top of the park's buildings to collect the park's lighting demand data in real time, and build a big data management platform. At the same time, set up a big data collection tool in the big data management platform to collect nonlinear items in real time; S2. Connect the integrated sensor group to the big data management platform using a wireless communication network, receive the park lighting demand data in real time on the big data management platform, integrate it with nonlinear items into a lighting demand data set, extract time series features from the lighting demand data set, obtain a lighting demand feature vector set, and then construct a time series database to store the lighting demand feature vector set in chronological order; S3, construct a lighting demand prediction algorithm model, extract a lighting demand feature vector set, input it into the lighting demand prediction algorithm model, calculate and output the lighting demand coefficient Lreq(t), and then combine the nonlinear term of the lighting demand feature vector set to calculate and output the lighting demand nonlinear coefficient Lfxx(t), so as to further refine the lighting demand prediction; S4. Based on the obtained illumination demand nonlinear coefficient Lfxx(t) and the actual output illumination value Lout(t) of the park street lamp, a preliminary comparative evaluation is performed to analyze the illumination output of the park street lamp over time, and based on the evaluation results, an adaptive power adjustment mechanism for the park street lamp is executed to output the optimized power value Padj(t) of the street lamp to perform preliminary adaptive power adjustment on the park street lamp; S5. After the preliminary adaptive power adjustment is completed, the optimized power value Padj(t) of the street lamp, the nonlinear coefficient of the lighting demand Lfxx(t) and the actual output lighting value Lout(t) are correlated to calculate the output feedback index FBI(t), and the preset energy consumption threshold T is compared with the feedback index FBI(t) for secondary evaluation to analyze the energy consumption output of the park street lamps, and an iterative mechanism is generated based on the results of the secondary comparative evaluation.

2. The method for management and analysis based on big data according to claim 1, characterized in that: Said S1 includes S11 and S12; S11. Install a sensor group on the top of the building in the park area, and set the collection frequency of the integrated sensor group to collect once every hour to collect the park's lighting demand data; The integrated sensor group includes an ambient light sensor, an ambient noise sensor, an ambient light pollution monitor, a radar speedometer, and an infrared detection device; The park lighting demand data include sky brightness index SLI, ambient noise level NLV, light pollution index LPI, traffic speed TVS and population density PCD; S12. Build a big data management platform and set up a big data collection tool inside the big data management platform to collect nonlinear items in real time; The big data collection tools include API application programming interfaces and social media crawlers; The nonlinear items include weather data WTH, holiday time data EVT and energy consumption history data PWR.

3. The method for management analysis based on big data according to claim 2, characterized in that: The S2 includes S21, S22 and S23; S21. The integrated sensor group is integrated and connected with the big data management platform through NB-loT wireless communication technology, and the park lighting demand data collected by the integrated sensor group is transmitted to the big data management platform through the MQTT data transmission protocol; S22. In the big data management platform, the park lighting demand data is collected and received in real time, and integrated with the nonlinear items in the big data management platform to obtain a lighting demand data set, and time series feature extraction is performed on the lighting demand data set based on the collection time to obtain a lighting demand feature vector set; The lighting demand feature vector set includes the sky brightness index SLI(t) at time t, the ambient noise level NLV(t) at time t, the light pollution index LPI(t) at time t, the traffic speed TVS(t) at time t, the population density PCD(t) at time t and the nonlinear term; The nonlinear term includes weather data WTH(t) at time t, holiday time data EVT(t) at time t, and energy consumption history data PWR(t) at time t; S23. By constructing a time series database and setting a real-time data table and a historical data table in the time series database, the lighting demand feature vector set is passed through the write port of the time series database and written into the real-time data table, and the lighting demand feature vector set in the original real-time data table is automatically transferred to the historical data table, and arranged in chronological order in the historical data table.

4. The method for management analysis based on big data according to claim 3, characterized in that: The S3 includes S31 and S32; S31, constructing a lighting demand prediction algorithm model, extracting a lighting demand feature vector set in a real-time data table through a write port of a time series database, inputting the lighting demand prediction algorithm model, and calculating and outputting a lighting demand coefficient Lreq(t); The light demand coefficient Lreq(t) is calculated and output by the following light demand prediction algorithm model; Lreq(t)=a1·SLI(t) -1 +a2·log(NLV(t))+a3·e LPI(t) +a4·TVS(t) 0.5 +a5·PCD(t); In the formula, e represents an exponential function, log represents a logarithmic function, a1, a2, a3, a4 and a5 represent preset weight values ​​of the sky brightness index SLI, the ambient noise level NLV, the light pollution index LPI, the traffic speed TVS and the population density PCD, respectively, and a1+a2+a3+a4+a5=1, and the specific values ​​are set by the user.

5. The method for management analysis based on big data according to claim 4, characterized in that: S32, based on the obtained lighting demand coefficient Lreq(t), combined with the nonlinear terms and parameter interactions in the lighting demand feature vector set, calculate and output the lighting demand nonlinear coefficient Lfxx(t); The illumination demand nonlinear coefficient Lfxx(t) is calculated and outputted by the following algorithm formula; Ltxx(t)=Lreq(t)+b1·(PCD(t)·log(TVS(t)))+b2·WTH(t)+b3·EVT(t)+b4·PWR(t); Wherein, (PCD(t) log(TVS(t))) represents the interaction between the traffic speed TVS and the population density PCD, b1, b2, b3 and b4 represent the preset weight values ​​of the interaction between the traffic speed TVS and the population density PCD, the weather data WTH, the holiday time data EVT and the energy consumption history data PWR, respectively, and b1+b2+b3+b4=1, and the specific values ​​are set by the user.

6. The method for management analysis based on big data according to claim 5, characterized in that: The S4 includes S41 and S42; S41. Use the built-in light intensity sensor of the park street lamp to detect and obtain the actual output light Lout(t) of the park street lamp at the current moment in real time, and then make a preliminary comparative evaluation of the actual output light Lout(t) and the obtained nonlinear coefficient of light demand Lfxx(t), and analyze the current light output of the park street lamp and the predicted light demand. The specific evaluation contents are as follows; When the actual output light Lout(t)> the nonlinear coefficient of light demand Lfxx(t), it means that the park road lighting has excessive lighting, and the adaptive adjustment mechanism is executed at this time; When the nonlinear coefficient of light demand Lfxx(t) represents the expected effect of lighting load on park roads, no adaptive power regulation is required; When the actual output light Lout(t) is less than the nonlinear coefficient of light demand Lfxx(t), it indicates that the lighting of the park road is insufficient, and the adaptive adjustment mechanism is executed at this time.

7. The method for management and analysis based on big data according to claim 6, characterized in that: S42. When the preliminary comparison and evaluation shows that the street lamps in the park area have over-lighting and under-lighting phenomena, the adaptive power adjustment mechanism is automatically executed. The adaptive power adjustment mechanism calculates the output street lamp optimized power value Padj(t) by introducing the illumination deviation between the illumination demand nonlinear coefficient Lfxx(t) and the actual output illumination value Lout(t) of the park street lamp, and performs preliminary adaptive power adjustment on the park street lamp power. The street lamp optimization power value Padj(t) is calculated and output by the following algorithm formula; In the formula, Pmax represents the upper limit of street lamp power output, Lmax represents the upper limit of street lamp illumination intensity, and γ represents the efficiency adjustment coefficient. Indicates lighting deviation.

8. The method for management analysis based on big data according to claim 7, characterized in that: The S5 includes S51 and S52; S51, based on the street lamp optimization power value Padj(t) obtained by preliminary automatic adjustment of the park street lamp power, combined with the illumination deviation, perform a related calculation to output the feedback index FBI(t), and analyze the comprehensive energy consumption of the park street lamp after the preliminary adaptive power adjustment; The feedback index FBI(t) is calculated and outputted by the following algorithm formula; In the formula, λ represents the energy consumption weight coefficient, and its specific value is set by the user.

9. The method for management analysis based on big data according to claim 8, characterized in that: S52, based on the user's preset energy consumption threshold T in combination with the park's energy consumption standard, and then compared with the obtained feedback index FBI(t), a secondary comparative evaluation is performed to analyze the energy consumption of the park street lamps after the preliminary adaptive power adjustment, and an iterative mechanism is generated based on the secondary comparative evaluation results. The specific evaluation results are as follows; When the feedback index FBI(t)≤energy consumption threshold T, it means that the energy consumption of the park street lamp is normal. At this time, no adjustment is required and continuous monitoring is required; When the feedback indicator FBI(t)>energy consumption threshold T, it means that the energy consumption of the park street lamp is abnormal. At this time, the adaptive power adjustment mechanism is iteratively executed until the energy consumption returns to normal and the iteration is stopped.

10. A management analysis system based on big data, applied to a management analysis method based on big data according to any one of claims 1 to 9, characterized in that: It includes a lighting demand collection module, a data transmission and feature extraction module, a lighting demand prediction and analysis module, an energy efficiency optimization power output module, and an energy consumption feedback and analysis module; The light demand collection module collects the park's light demand data in real time by installing a sensor group on the top of the park's buildings, and builds a big data management platform. At the same time, a big data collection tool is set in the big data management platform to collect nonlinear items in real time; The data transmission and feature extraction module connects the integrated sensor group with the big data management platform by using a wireless communication network, receives the park lighting demand data in real time in the big data management platform, integrates it with the nonlinear term into a lighting demand data set, extracts time series features from the lighting demand data set, obtains a lighting demand feature vector set, and then constructs a time series database to store the lighting demand feature vector set in chronological order; The lighting demand prediction and analysis module constructs a lighting demand prediction algorithm model, extracts a lighting demand feature vector set, inputs it into the lighting demand prediction algorithm model, calculates and outputs a lighting demand coefficient Lreq(t), and then calculates and outputs a lighting demand nonlinear coefficient Lfxx(t) in combination with the nonlinear term of the lighting demand feature vector set, so as to further refine the lighting demand prediction; The energy efficiency optimization power output module performs preliminary comparative evaluation and analysis of the time illumination output of the park street lamp based on the obtained illumination demand nonlinear coefficient Lfxx(t) and the actual output illumination value Lout(t) of the park street lamp, and based on the evaluation result, executes the park street lamp adaptive power adjustment mechanism, outputs the street lamp optimized power value Padj(t) to perform preliminary adaptive power adjustment on the park street lamp; After the energy consumption feedback and analysis module completes the preliminary adaptive power adjustment, it correlates the optimized power value Padj(t) of the street lamp, the nonlinear coefficient of the lighting demand Lfxx(t) and the actual output lighting value Lout(t) to calculate the output feedback index FBI(t), and performs a secondary comparative evaluation between the preset energy consumption threshold T and the feedback index FBI(t), analyzes the energy consumption output of the park street lamps, and generates an iterative mechanism based on the secondary comparative evaluation results.

Images