Energy-saving management method and system of intelligent lighting street lamp

Abstract

The application relates to the technical field of data prediction analysis, in particular to an efficient energy-saving management method and system of intelligent lighting street lamps, which comprises the following steps: acquiring initial fitting flow of each time according to flow of all dates at the same time in a historical time period, clustering all dates according to flow distribution in different dates, adjusting the initial fitting flow by using the category containing the largest number of dates in the clustering result, fitting the obtained final fitting flow to obtain a fitting curve, acquiring default power of each time according to the fitting curve, adjusting the default power according to the difference between real-time flow and the fitting value on the fitting curve and an error threshold to obtain target power of the current time, and adjusting the power of the intelligent street lamp in real time according to the target power. The application can meet the brightness demand of the flow of people and the flow of vehicles, realize energy saving, reduce the adjustment frequency, and protect the service life of the street lamp.

Description

Technical Field

[0001] This invention relates to the field of data prediction and analysis technology, specifically to an energy-saving management method and system for smart streetlights. Background Technology

[0002] Smart streetlights can intelligently adjust and manage their brightness, duration, and energy consumption based on real-time conditions and needs, making them an important component of smart cities. Especially in the face of energy pressures on current urban public lighting systems, smart streetlights can effectively control road lighting energy consumption, extend streetlight lifespan, and reduce maintenance and management costs, representing an inevitable trend in the construction of a modern energy-efficient society.

[0003] In practical smart street light adjustment, the current pedestrian and vehicle traffic is typically monitored in real time using sensor modules. The lighting power of the smart streetlights is then adjusted based on this monitoring data to achieve energy-saving management. However, this method involves excessively frequent adjustments to the lighting power of the smart streetlights, which negatively impacts energy efficiency and the lifespan of the streetlights. Summary of the Invention

[0004] To address the aforementioned problems, this invention provides an energy-saving management method and system for smart streetlights.

[0005] The present invention provides an energy-saving management method for intelligent streetlights, which adopts the following technical solution:

[0006] One embodiment of the present invention provides an energy-saving management method for smart streetlights, the method comprising the following steps:

[0007] Collect traffic data for each date and time within a historical time period; obtain the initial fitted traffic for each time moment based on the traffic data for the same time moment across all dates;

[0008] Cluster all dates based on the traffic distribution across different dates, and select the cluster with the most dates in the clustering results as the optimal cluster. For each time point, obtain the difference between the traffic at that time point and the initial fitted traffic at the same time point for each date in the optimal cluster, and construct the difference sequence for that time point from all the differences. Based on the difference sequence for each time point, obtain the correlation between the initial fitted traffic and the optimal cluster at each time point.

[0009] The reference flow is obtained based on the flow at the same time on all dates; the initial fitted flow at each time is adjusted based on the reference flow at each time and the uncorrelation between the initial fitted flow at each time and the optimal cluster, to obtain the final fitted flow at each time; the final fitted flow at each time is fitted to obtain the fitted curve.

[0010] Obtain the default power at each moment based on the fitted curve; set the error threshold at each moment; collect the real-time traffic at the current moment on the current date; adjust the default power based on the difference between the real-time traffic at the current moment and the fitted value at the corresponding moment on the fitted curve, as well as the error threshold, to obtain the target power at the current moment.

[0011] The power of the smart streetlights is adjusted in real time according to the target power.

[0012] Preferably, the specific steps for obtaining the initial fitted flow for each time moment based on the flow at the same time on all dates are as follows:

[0013] Divide all dates within the historical time period into special dates and non-special dates; obtain the weights of special dates and non-special dates:

[0014]

[0015]

[0016] in, This indicates the weight of each specific date; This indicates the weight of each non-special date; Indicates the quantity of special dates; Indicates the number of days included in a historical time period; It is an exponential function with the natural constant as its base;

[0017] The initial fitted flow for each time step is obtained based on the weight of each special date, the weight of non-special dates, and the flow rate at each time step for each date:

[0018]

[0019] in, Indicates the first time in all moments The initial fitted flow rate at time 1. Take all [1, Integers in ] Indicates the number of all moments; Indicates the first [number]th ... Heavenly Traffic at any given moment; Indicates the number of days included in a historical time period; Indicates the first The weight of the heavens, when the first When the day is a special date When the first When the day is not a special date .

[0020] Preferably, the specific steps for clustering all dates based on traffic distribution across different dates are as follows:

[0021] For each traffic data point at each date, obtain the proportion of traffic at that time to the total traffic of that date, as the traffic percentage for that time. Based on the traffic percentage for each date at each time, obtain the difference in traffic percentage for each date relative to the whole. Cluster the differences in traffic percentage for all dates relative to the whole within the historical time period, and divide all dates into multiple clusters.

[0022] Preferably, the specific steps for obtaining the difference in traffic share of each date relative to the overall traffic share based on the traffic share of each date at each time point are as follows:

[0023]

[0024] in, Indicates the first [number]th ... The difference in traffic share for each date relative to the overall total. Indicates the number of days included in a historical time period. This represents the number of times across all moments. Indicates the first [number]th ... The date Traffic percentage at any given moment Indicates the first [number]th ... The date Traffic percentage at any given moment.

[0025] Preferably, the specific steps for obtaining the reference traffic at a given time based on the traffic at the same time on all dates are as follows:

[0026] For each moment, obtain the different traffic values ​​for all dates at that moment, count the number of dates corresponding to each traffic value, divide the number of dates corresponding to each traffic value by the total number of dates in the historical time period to obtain the date distribution density of each traffic value, and select the traffic value with the highest date distribution density as the reference traffic for that moment.

[0027] Preferably, the specific steps for obtaining the correlation between the initial fitted flow and the optimal cluster at each time step based on the difference sequence at each time step are as follows:

[0028]

[0029] in, Indicates the first time in all moments The initial fitted flow at each time point is uncorrelated with the optimal cluster. Indicates the first A sequence of differences at each time step. Indicates the first A sequence of differences at each time step. Take all [1, [Middle and] Dissimilar integers, This represents the number of times across all moments. Represents the covariance function. This represents the variance function.

[0030] Preferably, the specific steps for adjusting the initial fitted flow at each time step based on the reference flow at each time step, the initial fitted flow at each time step, and the lack of correlation between the initial fitted flow at each time step and the optimal cluster are as follows:

[0031]

[0032] in, Indicates the first time in all moments The final fitted flow rate at time step, Indicates the first Reference flow rate at any time Indicates the first time in all moments The initial fitted flow rate at time t, Indicates the first time in all moments The initial fitted flow at each time point is uncorrelated with the optimal cluster.

[0033] Preferably, the specific steps for obtaining the default power at each moment based on the fitted curve are as follows:

[0034]

[0035] in, Indicates the first time in all moments The default power at any given moment. This indicates the minimum power required for the smart streetlight to maintain its lowest brightness. This indicates the maximum power of the smart street light. Indicates the first The fitted value on the fitted curve at time _____. This represents the minimum fitted value on the fitted curve. This represents the largest fitted value on the fitted curve.

[0036] Preferably, the specific steps for adjusting the default power based on the difference between the real-time traffic at the current moment and the fitted value on the fitted curve at the corresponding moment, as well as the error threshold, to obtain the target power at the current moment are as follows:

[0037] The real-time adjustment coefficient is obtained based on the real-time traffic at each moment:

[0038]

[0039] in, Indicates the current date. Real-time adjustment coefficient at any moment This indicates that the street light sensor module detected the current date. Real-time traffic at any given moment Indicates the first line on the fitted curve The fitted value at time 10:00. Indicates the first Error threshold at any given time;

[0040] Calculate the target power for each moment of the current date based on the real-time adjustment coefficient and the default power:

[0041]

[0042] in, Indicates the current date. Target power at any given time Indicates the first time in all moments The default power at any given moment. This indicates the minimum power required for the smart streetlight to maintain its lowest brightness. This indicates the current maximum power of the smart street light. This represents the hyperbolic tangent function.

[0043] The present invention also proposes an energy-saving management system for smart streetlights, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any one of the energy-saving management methods for smart streetlights.

[0044] The beneficial effects of the technical solution of this invention are as follows: This invention uses the category with the most dates in the clustering results to adjust the initial fitted flow rate, making the final fitted flow rate more consistent with the overall flow rate level at each moment in the historical time period. This makes the subsequent adjustment of street light power based on the fitted curve corresponding to the final fitted flow rate more reliable. Traditional methods directly adjust street light power based on real-time flow rate, resulting in excessively frequent adjustments that affect energy-saving effects and street light lifespan. This invention sets an error threshold and adjusts the default power based on the difference between the real-time flow rate and the fitted value on the fitted curve, along with the error threshold, to obtain the target power for the current moment. Real-time adjustment of street light power based on the target power reduces the frequency of adjustments compared to traditional methods, improves energy-saving effects, and protects the lifespan of street lights. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart illustrating the steps of an energy-saving management method for a smart street light according to the present invention. Detailed Implementation

[0047] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an energy-saving management method for smart streetlights proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0049] The following description, in conjunction with the accompanying drawings, details a specific scheme for an energy-saving management method for smart streetlights provided by this invention.

[0050] Please see Figure 1 The diagram illustrates a flowchart of an energy-saving management method for smart streetlights according to an embodiment of the present invention. The method includes the following steps:

[0051] S001. Collect traffic data for each date and time within a historical time period.

[0052] The system acquires traffic flow data for each moment of the daily period from the earliest on to the latest off of the smart streetlights on the road within a historical timeframe. The traffic flow data represents the sum of pedestrian and vehicle traffic. In this embodiment, the historical timeframe is the past two years, with each moment measured in hours. For example, if the streetlights were first turned on at 5 PM and last turned off at 8 AM the following day for the past two years, then the system acquires the hourly traffic flow data for that road from 5 PM to 8 AM the following day for the past two years. In other embodiments, implementers can set the range of the historical timeframe and the unit of measurement according to the actual implementation situation.

[0053] Daily weather data for a historical period can be obtained from meteorological websites. Heavy snow, low temperatures, strong winds, high temperatures, heavy rainfall, or continuous rainfall are considered severe weather, and the dates corresponding to these severe weather conditions are designated as special dates. It should be noted that the criteria for severe weather refer to severe weather warnings issued by meteorological departments, but implementers can also set their own standards based on actual circumstances.

[0054] Dates that fall on holidays within a historical period are also considered special dates.

[0055] This completes the collection of historical traffic flow data for roads, and also obtains specific dates.

[0056] S002. Obtain the initial fitted flow rate for each time point based on the flow rate at the same time point for all dates.

[0057] It should be noted that, in order to manage energy conservation for streetlights, it is necessary to fit the traffic flow of the road where the streetlights are located at each moment. In traditional methods, the average traffic flow at the same time on all dates in historical traffic data is usually used as the fitted traffic flow for that moment, thus performing curve fitting on the fitted traffic flow for all moments. However, due to the influence of holidays and severe weather, the traffic flow on the corresponding date will deviate from that of a normal day, making it inaccurate to directly use the average traffic flow at the same time on all dates as the fitted traffic flow for that moment. Therefore, in this embodiment of the invention, when obtaining the fitted traffic flow for each moment, the weight of traffic flow on special dates such as holidays and severe weather is reduced, while the weight of traffic flow on non-special dates is increased. This makes the fitted traffic flow for each moment more focused on the traffic flow at the corresponding time on non-special dates, thereby reducing the impact of holidays and severe weather.

[0058] Specifically, obtain the number of special dates, and then determine the weight of each special date based on the number of special dates:

[0059]

[0060] in, This indicates the weight of each specific date; Indicates the quantity of special dates; Indicates the number of days included in a historical time period; It is an exponential function with the natural constant as its base; in traditional methods, the average flow rate at the same time on all dates is used as the fitting point for that time, and the weight of each date is... This embodiment will As a negative correlation normalization function for the number of special dates, to... Limited to Within this range, the weight of each specific date is reduced.

[0061] The weight of each non-special date is obtained based on the weight of each special date:

[0062]

[0063] in, This indicates the weight of each non-special date; This indicates the weight of each specific date; Indicates the quantity of special dates; Indicates the number of days included in a historical time period; This represents the weight of all special dates. The remaining weight is then evenly distributed among all non-special dates to obtain the weight of each non-special date.

[0064] The initial fitted flow rate at each time step is obtained based on the weight of each special date and the weight of non-special dates:

[0065]

[0066] in, Indicates the first time in all moments The initial fitted flow rate at time 1. Take all [1, Integers in ] Indicates the number of all moments; Indicates the first [number]th ... Heavenly Traffic at any given moment; Indicates the number of days included in a historical time period; Indicates the first The weight of the heavens, when the first When the day is a special date When the first When the day is not a special date .

[0067] At this point, the initial fitted flow rate at each time step has been obtained.

[0068] S003. Cluster all dates according to the flow distribution on different dates, and adjust the initial fitted flow for each time moment using the category with the most dates in the clustering results to obtain the final fitted flow for each time moment.

[0069] Step S002 obtains the initial fitted flow rate for each time moment. The initial fitted flow rate is calculated by combining factors such as weather and holidays within the historical time period. The factors used in the calculation are relatively singular because even after weighting adjustments for special dates, there may still be some unexpected situations on non-special dates that cause deviations in the flow rate data, affecting the accuracy of the initial fitted flow rate for each time moment. Therefore, this embodiment corrects the initial fitted flow rate for each time moment by combining the flow rate distribution between different dates to obtain a fitted flow rate that is more consistent with the actual situation.

[0070] In this embodiment, for each traffic data point on each date, the proportion of traffic at that moment to the total traffic for that date is obtained as the traffic percentage for that moment. For example, if the traffic at 7 PM on a certain day accounts for 20% of the total traffic from 5 PM on that day to 8 AM the next day, then the traffic percentage at 7 PM on that day is 20%. On non-special dates, the traffic percentages at the same time on different dates should be similar. On special dates, the traffic percentages at the same time should differ significantly from those on non-special dates. Therefore, this embodiment calculates the difference in traffic percentages at various times for each date compared to other dates, and clusters the dates based on this difference. Dates with similar traffic percentages at different times are merged into the same cluster, allowing the initial fitted traffic for each moment to be corrected based on the distribution characteristics of the clusters, thus obtaining an initial fitted traffic that better reflects the actual situation.

[0071] Specifically, based on the traffic share of each date at each time point, we obtain the difference in traffic share for each date relative to the overall total:

[0072]

[0073] in, Indicates the first [number]th ... The difference in traffic share for each date relative to the overall total. Indicates the number of days included in a historical time period. This represents the number of times across all moments. Indicates the first [number]th ... The date Traffic percentage at any given moment Indicates the first [number]th ... The date Traffic percentage at any given moment It indicates the first The date Traffic share at each moment and the first The date The difference in traffic share at each moment, and amplified by squaring this difference, by comparing the first... The difference in traffic share between each time point on a given date and each time point on other days is obtained. The difference in traffic share for each date relative to the overall volume.

[0074] It should be noted that for dates where the traffic share at any given moment is similar to that of other dates at the same moment, the difference in traffic share relative to the overall picture is small. However, for dates with traffic deviations due to unforeseen circumstances, the difference in traffic share at any given moment on that date compared to other dates at the same moment is significant, resulting in a large difference in traffic share relative to the overall picture for that date. Therefore, this embodiment of the invention categorizes all dates into multiple categories based on the traffic difference percentage of each date relative to the overall picture, so that the initial fitted traffic can be corrected subsequently based on the traffic at each moment on each date within each category.

[0075] Specifically, the K-means clustering algorithm is used to cluster the differences in traffic share of all dates relative to the overall total traffic within a historical time period, dividing all dates into multiple clusters, where the number of clusters is... The number of clusters is obtained using the elbow method. It should be noted that the K-means clustering algorithm and the elbow method are well-known techniques and will not be described in detail here. The K-means clustering algorithm and the elbow method are merely a preferred implementation method in this embodiment, and are not specifically limited thereto. Implementers may also use other clustering algorithms and methods to determine the number of clusters. .

[0076] The cluster containing the most dates in the clustering results is selected as the optimal cluster. The optimal cluster reflects the overall traffic level at each moment within the historical time period. Theoretically, most dates within the historical time period are normal without any unexpected events; therefore, the optimal cluster can be considered a set of normal dates. Thus, this embodiment adjusts the initial fitted traffic by obtaining the difference between the traffic at each moment for each date within the optimal cluster and the initial fitted traffic at the corresponding moment, thereby obtaining a final fitted traffic that better reflects the overall traffic level at each moment within the historical time period.

[0077] Specifically, for each time point, the difference between the flow rate at that time point and the initial fitted flow rate at the same time point for each date in the optimal cluster is obtained, and all the differences at that time point are used to form the difference sequence for that time point.

[0078] The correlation between the initial fitted flow and the optimal cluster at each time step is obtained based on the difference sequence at each time step:

[0079]

[0080] in, Indicates the first time in all moments The initial fitted flow at each time point is uncorrelated with the optimal cluster. Indicates the first A sequence of differences at each time step. Indicates the first A sequence of differences at each time step. Take all [1, [Middle and] Dissimilar integers, This represents the number of times across all moments. Represents the covariance function. Represents the variance function. It indicates the first The difference sequence at time n and the first time n The correlation of the difference sequence at each time point, when The closer the difference is to 1, the more positively correlated the difference sequences between these two time points are. The closer the difference is to -1, the more negatively correlated the difference sequences between these two time points are. The closer the difference is to 0, the less correlated the difference sequences between the two points are. Reflects the first The correlation between the difference sequence at time t and the difference sequences at other times, when the t... The greater the correlation between the difference sequence at time t and the difference sequences at other times, the stronger the correlation. The initial fitted flow rate at time t is correlated with the initial fitted flow rate at other times relative to the flow rate at each time corresponding to each date in the optimal cluster, thus indicating that the initial fitted flow rate at time t is the same as that at other times. The greater the correlation between the initial fitted flow at each time point and the optimal cluster, the more important it is to use 1 minus... As the first The initial fitted flow at each time point is uncorrelated with the optimal cluster.

[0081] It should be noted that since the optimal cluster is selected based on the maximum number of dates, when the initial fitted flow at a certain moment has a large discorrelation with the optimal cluster, it indicates that the initial fitted flow at that moment deviates from the flow data of more dates in the historical time period. Therefore, this embodiment adjusts the initial fitted flow based on this deviation value. The direction of adjustment should be towards the flow with the highest date distribution density among the flow at the same time on all dates. The flow with the highest date distribution density is selected because the road flow during peak periods usually does not fluctuate too much. Therefore, the flow with the highest date distribution density among the flow at the same time should be located at the approximate center of all flows with high date distribution density at that time. Adjusting the initial fitted flow based on this position can make the final fitted curve more consistent with the flow data of ordinary dates at that time.

[0082] Specifically, for each moment, the different traffic values ​​for that moment across all dates are obtained. The number of dates corresponding to each traffic value is counted. The date distribution density of each traffic value is obtained by dividing the number of dates corresponding to each traffic value by the total number of dates in the historical time period. The traffic value with the highest date distribution density is selected as the reference traffic for that moment.

[0083] Based on the reference flow rate at each time step and the inconsistency between the initial fitted flow rate at each time step and the optimal cluster, the initial fitted flow rate at each time step is adjusted to obtain the final fitted flow rate at each time step:

[0084]

[0085] in, Indicates the first time in all moments The final fitted flow rate at time step, Indicates the first Reference flow rate at any time Indicates the first The initial fitted flow rate at time t, Indicates the first The initial fitted flow at each time point is uncorrelated with the optimal cluster. Indicates the first The difference between the reference flow rate and the initially fitted flow rate at time step. The decision was made The direction of adjustment for the initial fitted flow rate at time step [is determined], when the reference flow rate is less than the initial fitted flow rate. For negative numbers, the corresponding It is also a negative number. By adding the initial fitted flow rate to the negative number, the initial fitted flow rate is reduced, thus moving closer to the reference flow rate. When the reference flow rate is greater than the initial fitted flow rate, For positive numbers, the corresponding It is also a positive number. By adding the initial fitted flow rate to the positive number, the initial fitted flow rate is increased, thereby moving closer to the reference flow rate. The decision was made The degree of adjustment of the initial fitted flow rate at time t, when the t The greater the discorrelation between the initial fitted flow and the optimal cluster at time step 1, and the greater the difference between the reference flow and the initial fitted flow, the greater the adjustment degree. The smaller the correlation between the initial fitted flow and the optimal cluster at each time point, and the smaller the difference between the reference flow and the initial fitted flow, the smaller the degree of adjustment.

[0086] At this point, the final fitted flow rate at each time step has been obtained.

[0087] S004. Fit the final fitted flow rate at each time step to obtain the fitted curve, and set the error threshold for each time step.

[0088] The final fitted flow rate at all time points is fitted using a polynomial method with the least squares approach to obtain the fitted curve. The horizontal axis of the fitted curve represents different time points, and the vertical axis represents the final fitted flow rate. It should be noted that the least squares approach is a well-known technique and will not be described in detail here. This embodiment only uses the least squares approach as an example. Implementers can choose the fitting method according to the actual implementation situation, such as the maximum likelihood method.

[0089] The fitted curve was used as a baseline for adjusting the street light power.

[0090] It should be noted that using the fitted curve as a baseline for adjusting streetlight power aims to avoid frequent power adjustments affecting the lifespan of the streetlights. However, directly adjusting the power based on the difference between the fitted curve and the real-time traffic data detected by the streetlight sensing module would still lead to frequent power adjustments, failing to achieve the goal of protecting the streetlight's lifespan. Therefore, this embodiment sets error thresholds for each time point. When the difference between the real-time traffic data actually detected by the streetlight sensing module and the fitted value on the corresponding time point on the fitted curve does not exceed the error threshold, the streetlight power is still adjusted according to the fitted curve. For times with larger fitted values ​​on the fitted curve, this time may be during peak hours, where the actual traffic flow can vary significantly. In this case, the error threshold should be relatively larger to reduce the frequency of streetlight power adjustments during peak hours and protect the normal lifespan of the streetlights. For times with smaller fitted values ​​on the fitted curve, the traffic flow is lower, and due to the smaller base value, the range of variation is smaller. Therefore, this embodiment determines the relationship between the error thresholds at different times by considering the magnitude of the fitted values ​​on the fitted curve. Meanwhile, for periods of significant traffic fluctuation on normal dates, the error threshold is appropriately increased so that normal local traffic fluctuations do not require adjustment of street light power, thereby reducing the frequency of street light power adjustments and protecting the lifespan of the street lights.

[0091] Specifically, the error threshold for each time moment is obtained based on the maximum and minimum flow rates corresponding to different dates at each time point in the optimal cluster, as well as the fitted value of each time moment in the fitted curve.

[0092]

[0093] in, Indicates the first time in all moments Error threshold at time, Indicates the first The fitted value on the fitted curve at time _____. Represents the date of all dates in the optimal cluster. The maximum flow rate at any given moment. Represents the date of all dates in the optimal cluster. The minimum flow rate at any given moment. Represents the normalization function. Describes the minimum value function. Used to obtain and The minimum value in represents the first... The possible fluctuation range of the real-time flow at time t, based on the fitted value at the corresponding time on the fitted curve, will be used to determine the magnitude of the fluctuation range of the real-time flow at time t. Multiplying the normalized value of the fitted curve at time step n by the possible fluctuation range yields the value at time step n. The error threshold at each moment. On a normal date, the greater the flow fluctuation at that moment and the larger the fitted value on the fitted curve at that moment, the larger the error threshold at that moment. Conversely, on a normal date, the smaller the flow fluctuation at that moment and the smaller the fitted value on the fitted curve at that moment, the smaller the error threshold at that moment.

[0094] Thus, the error threshold at each moment has been obtained. It should be noted that the above method for obtaining the error threshold is only a preferred implementation method in this embodiment. The method for obtaining the error threshold is not limited, and implementers can set the error threshold according to the actual implementation situation, such as manually setting it based on experience.

[0095] S005. Obtain the default power at each moment based on the fitted curve, and adjust the default power according to the difference between the real-time flow at the current moment and the fitted value at the corresponding moment on the fitted curve, as well as the error threshold, to obtain the target power at the current moment.

[0096] Obtain the default power at each moment:

[0097]

[0098] in, Indicates the first time in all moments The default power at any given moment. This indicates the minimum power at which the smart street light maintains its lowest brightness, typically used when there is no traffic or very low traffic. This indicates the current maximum power of the smart street light. Indicates the first The fitted value on the fitted curve at time _____. This represents the minimum fitted value on the fitted curve. This represents the maximum fitted value on the fitted curve. The maximum and minimum fitted values ​​of the fitted curve are used to determine the... The magnitude of the fitted value at time 1 on the fitted curve is mapped to the range between the maximum and minimum power, serving as the value of the first fitted value. The default power at a given moment is intended to establish an initial link between power and historical traffic data. When the first... The larger the fitted value on the fitted curve at time t, the greater the value at time t. The greater the traffic flow at a given moment, the higher the brightness required on the road, and the greater the power of the streetlights accordingly.

[0099] It should be noted that the default power is based on the fitted value on the fitted curve, and the actual road brightness, i.e. the actual power, still needs to be adjusted according to the actual road traffic flow.

[0100] Specifically, the street light sensor module detects the real-time traffic flow at the current time and date, and transmits this real-time traffic flow to the server via a wireless sensor network. The server then obtains the real-time adjustment coefficient for each moment of the current date based on the real-time traffic flow detected by the street light sensor module.

[0101]

[0102] in, Indicates the current date. Real-time adjustment coefficient at any moment This indicates that the street light sensor module detected the current date. Real-time traffic at any given moment Indicates the first line on the fitted curve The fitted value at time 10:00. Indicates the first Error threshold at any given time. When the street light sensor module detects the current date... Real-time flow at time t and the fitted curve at time t When the difference between the fitted values ​​at time points is less than or equal to the error threshold, i.e. At that time, the first The real-time adjustment factor is 0, meaning no adjustment is needed at this time; the default power can be maintained. When the streetlight sensor module detects the current date... Real-time flow at time t and the fitted curve at time t When the difference between the fitted values ​​at time points exceeds the error threshold, i.e. hour, Reflects the first When the difference between the real-time flow and the fitted value exceeds the error threshold, then... As the first Real-time adjustment coefficient at any given moment.

[0103] The target power at each moment of the current date is obtained based on the real-time adjustment coefficient.

[0104]

[0105] in, Indicates the current date. Target power at any given time Indicates the current date. Real-time adjustment coefficient at any moment Indicates the first The default power at any given moment. This indicates the minimum power required for the smart streetlight to maintain its lowest brightness. This indicates the current maximum power of the smart street light. Represents the hyperbolic tangent function, used to... Limited to the range [-1, 1], when the real-time adjustment coefficient is less than 0, the real-time flow is smaller than the fitted value. If the value is less than 0, use the adjustment factor. With default power The multiplication yields an adjustment value, which is less than 0. This adjustment value, less than 0, is used to adjust the power to a lower level by adding the default power. When the real-time adjustment coefficient is greater than 0, the real-time flow rate is larger than the fitted value. If the value is greater than 0, use the adjustment factor. With default power The values ​​are multiplied to obtain an adjustment value. Since this adjustment value is greater than 0, the power is adjusted upwards by adding the default power to the adjustment value that is greater than 0. When the adjusted power exceeds the maximum power of the street light, the maximum power is used as the final target power. When the adjusted power is lower than the minimum power of the street light, the minimum power is used as the final target power.

[0106] At this point, the target power for each moment of the current date has been obtained.

[0107] S006. Adjust the power of the smart streetlights in real time according to the target power.

[0108] The server adjusts the power of the smart streetlights in real time via a wireless sensor network based on the target power at the current moment, ensuring that the current power of the smart streetlights reaches the target power. This ensures that the smart streetlights can meet the brightness requirements of pedestrian and vehicle traffic while also achieving energy conservation. At the same time, it reduces the adjustment frequency, thereby protecting the lifespan of the streetlights.

[0109] Through the above steps, energy-saving management of smart streetlights has been completed.

[0110] This invention also proposes an energy-saving management system for smart streetlights, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any one of the energy-saving management methods for smart streetlights.

[0111] This invention adjusts the initial fitted flow rate by utilizing the clustering results that contain the most dates. This makes the final fitted flow rate more consistent with the overall flow rate at each moment in the historical time period, thus increasing the reliability of subsequent adjustments to street light power based on the fitted curve corresponding to the final fitted flow rate. Traditional methods directly adjust street light power based on real-time flow, resulting in excessively frequent adjustments that negatively impact energy efficiency and street light lifespan. This invention sets an error threshold and adjusts the default power based on the difference between the real-time flow and the fitted value on the fitted curve, obtaining the target power for the current moment. Real-time adjustments to street light power are then made based on this target power, reducing the frequency of adjustments compared to traditional methods, improving energy efficiency, and protecting street light lifespan.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An energy-saving management method for intelligent streetlights, characterized in that, The method includes the following steps: Collect traffic data for each date and time within a historical time period; obtain the initial fitted traffic for each time moment based on the traffic data for the same time moment across all dates; Cluster all dates based on the traffic distribution across different dates, and select the cluster with the most dates in the clustering results as the optimal cluster. For each time point, obtain the difference between the traffic at that time point and the initial fitted traffic at the same time point for each date in the optimal cluster, and construct the difference sequence for that time point from all the differences. Based on the difference sequence for each time point, obtain the correlation between the initial fitted traffic and the optimal cluster at each time point. The reference flow is obtained based on the flow at the same time on all dates; the initial fitted flow at each time is adjusted based on the reference flow at each time and the uncorrelation between the initial fitted flow at each time and the optimal cluster, to obtain the final fitted flow at each time; the final fitted flow at each time is fitted to obtain the fitted curve. Obtain the default power at each moment based on the fitted curve; set the error threshold at each moment; collect the real-time traffic at the current moment on the current date; adjust the default power based on the difference between the real-time traffic at the current moment and the fitted value at the corresponding moment on the fitted curve, as well as the error threshold, to obtain the target power at the current moment. The power of the smart streetlights is adjusted in real time according to the target power.

2. The energy-saving management method for a smart street light according to claim 1, characterized in that, The specific steps for obtaining the initial fitted flow for each time point based on the flow at the same time on all dates are as follows: Divide all dates within the historical time period into special dates and non-special dates; obtain the weights of special dates and non-special dates: in, This indicates the weight of each specific date; This indicates the weight of each non-special date; Indicates the quantity of special dates; Indicates the number of days included in a historical time period; It is an exponential function with the natural constant as its base; The initial fitted flow for each time step is obtained based on the weight of each special date, the weight of non-special dates, and the flow rate at each time step for each date: in, Indicates the first time in all moments The initial fitted flow rate at time 1. Take all [1, Integers in ] Indicates the number of all moments; Indicates the first [number]th ... Heavenly Traffic at any given moment; Indicates the number of days included in a historical time period; Indicates the first The weight of the heavens, when the first When the day is a special date When the first When the day is not a special date .

3. The energy-saving management method for a smart street light according to claim 1, characterized in that, The specific steps involved in clustering all dates based on traffic distribution across different dates are as follows: For each traffic data point at each date, obtain the proportion of traffic at that time to the total traffic of that date, as the traffic percentage for that time. Based on the traffic percentage for each date at each time, obtain the difference in traffic percentage for each date relative to the whole. Cluster the differences in traffic percentage for all dates relative to the whole within the historical time period, and divide all dates into multiple clusters.

4. The energy-saving management method for a smart street light according to claim 3, characterized in that, The specific steps for obtaining the difference in traffic share for each date relative to the overall traffic based on the traffic share at each moment on each date are as follows: in, Indicates the first [number]th ... The difference in traffic share for each date relative to the overall total. Indicates the number of days included in a historical time period. This represents the number of times across all moments. Indicates the first [number]th ... The date Traffic percentage at any given moment Indicates the first [number]th ... The date Traffic percentage at any given moment.

5. The energy-saving management method for a smart street light according to claim 1, characterized in that, The specific steps for obtaining the reference traffic at a given time based on the traffic at the same time on all dates are as follows: For each moment, obtain the different traffic values ​​for all dates at that moment, count the number of dates corresponding to each traffic value, divide the number of dates corresponding to each traffic value by the total number of dates in the historical time period to obtain the date distribution density of each traffic value, and select the traffic value with the highest date distribution density as the reference traffic for that moment.

6. The energy-saving management method for a smart street light according to claim 1, characterized in that, The specific steps for obtaining the correlation between the initial fitted flow and the optimal cluster at each time step based on the difference sequence at each time step are as follows: in, Indicates the first time in all moments The initial fitted flow at each time point is uncorrelated with the optimal cluster. Indicates the first A sequence of differences at each time step. Indicates the first A sequence of differences at each time step. Take all [1, [Middle and] Dissimilar integers, This represents the number of times across all moments. Represents the covariance function. This represents the variance function.

7. The energy-saving management method for a smart street light according to claim 1, characterized in that, The specific steps involved in adjusting the initial fitted flow at each time step based on the reference flow at each time step, the initial fitted flow at each time step, and the incompatibility between the initial fitted flow at each time step and the optimal cluster are as follows: in, Indicates the first time in all moments The final fitted flow rate at time step, Indicates the first Reference flow rate at any time Indicates the first time in all moments The initial fitted flow rate at time t, Indicates the first time in all moments The initial fitted flow at each time point is uncorrelated with the optimal cluster.

8. The energy-saving management method for a smart street light according to claim 1, characterized in that, The specific steps for obtaining the default power at each moment based on the fitted curve are as follows: in, Indicates the first time in all moments The default power at any given moment. This indicates the minimum power required for the smart streetlight to maintain its lowest brightness. This indicates the maximum power of the smart street light. Indicates the first The fitted value on the fitted curve at any given time. This represents the minimum fitted value on the fitted curve. This represents the largest fitted value on the fitted curve.

9. The energy-saving management method for a smart street light according to claim 1, characterized in that, The specific steps for adjusting the default power based on the difference between the real-time flow at the current moment and the fitted value on the fitted curve at the corresponding moment, as well as the error threshold, to obtain the target power at the current moment are as follows: The real-time adjustment coefficient is obtained based on the real-time traffic at each moment: in, Indicates the current date. Real-time adjustment coefficient at any moment This indicates that the street light sensor module detected the current date. Real-time traffic at any given moment Indicates the first line on the fitted curve The fitted value at time 10:

00. Indicates the first Error threshold at any given time; Calculate the target power for each moment of the current date based on the real-time adjustment coefficient and the default power: in, Indicates the current date. Target power at any given time Indicates the first time in all moments The default power at any given moment. This indicates the minimum power required for the smart streetlight to maintain its lowest brightness. This indicates the current maximum power of the smart street light. This represents the hyperbolic tangent function.

10. An energy-saving management system for intelligent streetlights, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-9.

Images