Street lamp state remote monitoring system and method based on big data

Abstract

The invention discloses a street lamp state remote monitoring system and method based on big data, and relates to the technical field of street lamp monitoring, the street lamp state remote monitoring system comprises a plurality of local modules, an edge node module and a platform module, each local module is matched with each street lamp, and each local module comprises a light sensing unit, an edge node module and a platform module, monitoring the brightness and popularity of the street lamp, and generating corresponding brightness data and popularity data; and the carrier unit is used for collecting the electric power data of the street lamps and transmitting the electric power data, the brightness data and the heat data of the street lamps to the corresponding edge node modules. According to the invention, through the light sensing unit and the carrier unit, the brightness, heat and power data of the street lamp are comprehensively collected, multi-dimensional data analysis is formed, the accuracy of fault identification is improved, and through primary screening and secondary data comparison, the interference of environment change and street lamp type difference on fault judgment is reduced, and the false alarm rate is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of street lamp monitoring, and in particular to a street lamp status remote monitoring system and method based on big data. Background Art

[0002] With the rapid advancement of smart city construction, the intelligence level of urban lighting systems has gradually improved. As an important part of urban infrastructure, the stable operation of street lamps is of great significance to protecting citizens' lives and improving the city's image.

[0003] However, traditional street light status monitoring methods have many limitations. They mainly rely on manual inspections and simple fault alarm systems, which have problems such as low efficiency, long maintenance cycles, and untimely response to problems.

[0004] After searching, a Chinese patent (publication number: CN117057784A) discloses a method and system for monitoring the operating status of street lamps. The patent obtains the real-time lighting image data and street light sensing intensity data of street lamps through a sensor device, and transmits the data to a server through a ZigBee wireless communication module. The real-time lighting image data to be trained is input into the target ResNet convolutional neural network model for training to obtain the real-time status data of the street lamp, and a street lamp maintenance measures database is established. The status of the street lamp is judged according to the real-time status data of the street lamp and the street light sensing intensity data. If the street lamp is a street lamp in a state to be repaired, a street lamp maintenance measure is generated for the street lamp in a state to be repaired according to the street lamp maintenance measures database, and the street lamp maintenance measure is sent to the server and an early warning is issued. If the street lamp is a street lamp in a state to be repaired, the street lamp in a state to be repaired is monitored by a sensor device.

[0005] In the prior art, traditional monitoring systems mainly rely on a single sensor or simple circuit detection methods, and usually can only obtain partial data of street lamps. In addition, there is a lack of effective mechanisms for multi-dimensional trend analysis of data and utilization of historical fault data, resulting in low accuracy of fault identification. Therefore, the present invention proposes a remote monitoring system and method for street lamp status based on big data. Summary of the invention

[0006] The purpose of the present invention is to provide a street lamp status remote monitoring system and method based on big data to solve the problems mentioned in the above background technology.

[0007] The present invention can be implemented through the following technical solutions: A street lamp status remote monitoring system based on big data, including a local module, an edge node module and a platform module;

[0008] There are multiple local modules, and each local module is matched with each street lamp respectively, and each local module obtains the working data of the corresponding street lamp when working, and the local module includes:

[0009] The light sensing unit monitors the brightness and heat of the street lamps and generates corresponding brightness data and heat data;

[0010] The carrier unit collects the power data of the street lamps and transmits the power data, brightness data and heat data of the street lamps to the corresponding edge node modules;

[0011] The edge node module is used to receive the power data, brightness data and heat data of each street lamp in the corresponding area, and compare it with the built-in screening model to obtain the state fluctuation information of each street lamp matching the screening model, and the edge node module transmits the state fluctuation information to the platform module. The edge node includes:

[0012] The comparison unit regularly updates the screening model and compares the screening model with the power data, brightness data and heat data of each street lamp. The comparison unit marks the power data, brightness data and heat data that do not match the screening model as abnormal and adds a timestamp.

[0013] The marked and timestamped parts of the power data, brightness data, and heat data of each street lamp are the status fluctuation information of the corresponding street lamp;

[0014] A transmission unit, wherein the transmission unit transmits the power data, brightness data and heat data marked and timestamped and other normal power data, brightness data and heat data to the platform module;

[0015] After receiving the power data, brightness data and heat data from each edge node, the platform module stores the normal power data, brightness data and heat data, and matches the marked and timestamped parts of the power data, brightness data and heat data, that is, the state fluctuation information of each street lamp with the big data, predicts the failure of the street lamp, and issues a warning signal.

[0016] A further technical improvement of the present invention is that the comparison method of the screening model comprises the following steps:

[0017] S1. Preset the normal range of power data, brightness data and heat data of each street lamp in the corresponding area;

[0018] The normal ranges of power data, brightness data, and heat data are dynamically adjusted based on factors such as season, climate, and age;

[0019] S2. After the power data, brightness data, and heat data fluctuate beyond the corresponding normal range, mark them, and add a timestamp to the marked power data, brightness data, or heat data to generate state fluctuation information;

[0020] And when comparing, the power data, brightness data and heat data are analyzed independently;

[0021] S3. After generating the data fluctuation information, the edge node module adds location information to the corresponding data fluctuation information based on the actual geographical location of each street lamp;

[0022] S4, the edge node module collects power data, brightness data or heat data in the past time period of the state fluctuation information, and generates an abnormal data set;

[0023] Finally, the edge node module transmits the abnormal data set together with the power data, brightness data and heat data to the platform module.

[0024] A further technical improvement of the present invention is that: the edge node module generates secondary comparison data based on the average values ​​of the brightness data and heat data of each street lamp in the corresponding area, and comprehensively analyzes the brightness data and heat data of each street lamp in the corresponding area, thereby reducing factors such as environmental changes and differences in street lamp types, and avoiding marking a street lamp as faulty in isolation;

[0025] When the edge node module generates secondary comparison data, it smoothes the brightness data and heat data of each street lamp in the corresponding area to reduce the impact of high-discrete brightness data or heat data on the overall situation.

[0026] A further technical improvement of the present invention is that: the edge node module is provided with a buffer zone;

[0027] When the brightness data or heat data does not match the screening model but does not exceed the secondary comparison data, the edge node module marks the brightness data or heat data as "unconfirmed abnormality" and transmits it to the buffer. In the subsequent N time windows, if the brightness data or heat data of the street lamp exceeds the secondary comparison data, or the number marked as "unconfirmed abnormality" reaches the preset judgment threshold, the "unconfirmed abnormality" mark of the corresponding brightness data or heat data is converted to an abnormal mark.

[0028] A further technical improvement of the present invention is that before generating the secondary comparison data, the edge node calculates whether the change trends of the brightness data and the heat data in the corresponding area in the past time window are consistent. The calculation method of the change trend includes:

[0029] a1. Calculate the first-order difference of the brightness data and heat data of each street lamp;

[0030] a2. Based on the first-order difference of the brightness data and heat data of each street lamp, the mean of the brightness data change and heat data change of each street lamp in the area are counted, and the standard deviation of the brightness data change and the standard deviation of the heat data change are calculated;

[0031] a3. Calculate the consistency between the brightness data change trend of each street lamp in the corresponding area at the past W moments and the heat data change trend by using the Pearson correlation coefficient;

[0032] The formula is:

[0033]

[0034] Where, ρL,T(t) is the Pearson correlation coefficient;

[0035] ΔL i (t) is the brightness data change of the i-th street lamp at time t;

[0036] ΔT i (t) is the change in heat data of the i-th street lamp at time t;

[0037] μΔL(t) is the mean value of the brightness data change of all street lamps in the area at time t;

[0038] μΔT(t) is the mean value of the change in heat data of all street lamps in the area at time t;

[0039] a4. Average all the results of the Pearson correlation coefficient in a3 within W moments. The formula used is:

[0040] Where, ρL,T(t) is the Pearson correlation coefficient;

[0041] μ ρL,T (W) is the average Pearson correlation coefficient within W moments;

[0042] W end and W start is the end and start time of W moment;

[0043] a5. Set the error range e and generate the judgment condition: |μ ρL,T (W)-1<e;

[0044] If the Pearson correlation coefficient μ ρL,T If the difference between (W) and 1 is less than the preset error range e, it is considered that the change trend of the brightness data and the heat data at time W is consistent and is used for secondary comparison of data;

[0045] If the changing trends of brightness data and heat data are inconsistent, the generation of secondary comparison data is canceled.

[0046] A further technical improvement of the present invention is that: the platform module generates an abnormal detection axis of a time series for the brightness data, heat data and power data of the corresponding street lamps based on the timestamp in the state fluctuation information;

[0047] By calculating the time difference between each time stamp between the anomaly detection axes, the abnormal frequency of the corresponding street light is determined.

[0048] A further technical improvement of the present invention is that: the platform module sets a plurality of frequency threshold ranges, and the platform module sets a corresponding computing resource allocation level based on each frequency threshold range;

[0049] When the frequency threshold range is low, it indicates that the frequency of status fluctuation information of the corresponding street lamp is high. The platform module will give priority to mobilizing more computing resources to increase the monitoring intensity of the street lamp status fluctuation information and match it with possible fault modes in big data in a timely manner.

[0050] A further technical improvement of the present invention is that: the platform module constructs a multi-dimensional fault mode library based on the historical fault data of the street lamp, and associates the brightness data, heat data and power data of the street lamp with the fault;

[0051] When the platform module performs the association, the brightness data, the heat data and the power data are compared with the fault type, and data features related to the fault type are extracted, including brightness data fluctuations, heat data fluctuations and power data fluctuations.

[0052] The present invention also discloses a method for remotely monitoring street lamp status based on big data, which includes the following steps:

[0053] Step 1: Monitor each street lamp through multiple local modules to obtain the brightness data, heat data and power data of the corresponding street lamp;

[0054] Step 2: Receive the brightness data, heat data and power data of each local module in the corresponding area through the edge node module, and compare the brightness data, heat data and power data of each street lamp through the built-in screening model of the edge node module;

[0055] Step 3: The edge node module marks the power data, brightness data, and heat data that do not match the screening model as abnormal, adds a timestamp, and generates state fluctuation information;

[0056] Step 4: The platform module receives the power data, brightness data and heat data from each edge node, stores the normal power data, brightness data and heat data, and matches the marked and timestamped parts of the power data, brightness data and heat data, that is, the status fluctuation information of each street lamp with the big data, predicts the failure of the street lamp, and issues a warning signal.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] The present invention uses a light sensing unit and a carrier unit to comprehensively collect the brightness, heat and power data of street lamps, form multi-dimensional data analysis, improve the accuracy of fault identification, and reduce the interference of environmental changes and differences in street lamp types on fault judgment through primary screening and secondary comparison of data, effectively reducing the false alarm rate;

[0059] At the same time, the first-order difference and Pearson correlation coefficient are used to analyze the change trend and consistency of the data, realizing dynamic verification of abnormal data, further improving the accuracy of fault judgment. Moreover, through the frequency threshold range, the platform module can dynamically allocate computing resources, give priority to processing street lamp data with high-frequency state fluctuations, and improve the real-time response capability of the system.

[0060] Furthermore, the present invention combines historical fault data to build a fault mode library, matches real-time monitoring data with fault modes, predicts street lamp failures in advance and issues warnings, and improves the intelligent operation and maintenance capabilities of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0062] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0063] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0064] Example 1

[0065] See also Figure 1 As shown, the present invention provides a street lamp status remote monitoring system based on big data, including a local module, an edge node module and a platform module;

[0066] There are multiple local modules, and each local module is matched with each street lamp respectively, and each local module obtains the working data of the corresponding street lamp when working, and the local module includes:

[0067] The light sensing unit monitors the brightness and heat of the street lamps and generates corresponding brightness data and heat data;

[0068] The carrier unit includes a power sensor incorporated into the street lamp circuit, collects the power data of the street lamp, and transmits the power data, brightness data and heat data of the street lamp to the corresponding edge node module through the power line carrier communication technology;

[0069] The edge node module is used to receive the power data, brightness data and heat data of each street lamp in the corresponding area, and compare it with the built-in screening model to obtain the state fluctuation information of each street lamp that matches the screening model. The edge node module transmits the state fluctuation information to the platform module. The edge node includes:

[0070] The comparison unit regularly updates the screening model and compares the screening model with the power data, brightness data and heat data of each street lamp. The comparison unit marks the power data, brightness data and heat data that do not match the screening model as abnormal and adds a timestamp.

[0071] The marked and timestamped parts of the power data, brightness data, and heat data of each street lamp are the status fluctuation information of the corresponding street lamp;

[0072] A transmission unit, which transmits the power data, brightness data and heat data marked and added with a timestamp and other normal power data, brightness data and heat data to the platform module;

[0073] The comparison method of the screening model includes the following steps:

[0074] S1. Preset the normal range of power data, brightness data and heat data of each street lamp in the corresponding area;

[0075] The normal ranges of power data, brightness data, and heat data are dynamically adjusted based on factors such as season, climate, and age;

[0076] S2. After the power data, brightness data, and heat data fluctuate beyond the corresponding normal range, mark them, and add a timestamp to the marked power data, brightness data, or heat data to generate state fluctuation information;

[0077] And when comparing, the power data, brightness data and heat data are analyzed independently;

[0078] S3. After generating the data fluctuation information, the edge node module adds location information to the corresponding data fluctuation information based on the actual geographical location of each street lamp;

[0079] S4, the edge node module collects power data, brightness data or heat data in the past time period of the state fluctuation information, and generates an abnormal data set;

[0080] Finally, the edge node module transmits the abnormal data set together with the power data, brightness data, and heat data to the platform module;

[0081] The platform module receives the power data, brightness data and heat data from each edge node, stores the normal power data, brightness data and heat data, and matches the marked and timestamped parts of the power data, brightness data and heat data, that is, the state fluctuation information of each street lamp with the big data, predicts the failure of the street lamp, and issues a warning signal;

[0082] Based on the timestamp in the state fluctuation information, the platform module generates time series anomaly detection axes for the brightness data, heat data, and power data of the corresponding street lamps;

[0083] By calculating the time difference between each time stamp between the anomaly detection axes, the abnormal frequency of the corresponding street lamp is determined;

[0084] The platform module sets a plurality of frequency threshold ranges, and the platform module sets a corresponding computing resource allocation level based on each frequency threshold range;

[0085] When the frequency threshold range is low, it indicates that the frequency of state fluctuation information of the corresponding street lamp is high. The platform module will give priority to mobilizing more computing resources to increase the monitoring intensity of the state fluctuation information of the street lamp and match it with the possible fault modes in the big data in a timely manner;

[0086] In addition, the platform module builds a multi-dimensional fault mode library based on the historical fault data of street lamps, and associates the brightness data, heat data and power data of street lamps with faults;

[0087] When the platform module performs correlation, it compares the brightness data, heat data and power data with the fault type and extracts data features related to the fault type, including brightness data fluctuations, heat data fluctuations and power data fluctuations.

[0088] Example 2

[0089] A street lamp status remote monitoring system based on big data, including a local module, an edge node module and a platform module;

[0090] There are multiple local modules, and each local module is matched with each street lamp respectively, and each local module obtains the working data of the corresponding street lamp when working, and the local module includes:

[0091] The light sensing unit monitors the brightness and heat of the street lamps and generates corresponding brightness data and heat data;

[0092] The carrier unit collects the power data of the street lamps and transmits the power data, brightness data and heat data of the street lamps to the corresponding edge node modules;

[0093] The edge node module is used to receive the power data, brightness data and heat data of each street lamp in the corresponding area, and compare it with the built-in screening model to obtain the state fluctuation information of each street lamp that matches the screening model. The edge node module transmits the state fluctuation information to the platform module. The edge node includes:

[0094] The comparison unit regularly updates the screening model and compares the screening model with the power data, brightness data and heat data of each street lamp. The comparison unit marks the power data, brightness data and heat data that do not match the screening model as abnormal and adds a timestamp.

[0095] The marked and timestamped parts of the power data, brightness data, and heat data of each street lamp are the status fluctuation information of the corresponding street lamp;

[0096] A transmission unit, which transmits the power data, brightness data and heat data marked and added with a timestamp and other normal power data, brightness data and heat data to the platform module;

[0097] The comparison method of the screening model includes the following steps:

[0098] S1. Preset the normal range of power data, brightness data and heat data of each street lamp in the corresponding area;

[0099] The normal ranges of power data, brightness data, and heat data are dynamically adjusted based on factors such as season, climate, and age;

[0100] S2. After the power data, brightness data, and heat data fluctuate beyond the corresponding normal range, mark them, and add a timestamp to the marked power data, brightness data, or heat data to generate state fluctuation information;

[0101] And when comparing, the power data, brightness data and heat data are analyzed independently;

[0102] S3. After generating the data fluctuation information, the edge node module adds location information to the corresponding data fluctuation information based on the actual geographical location of each street lamp;

[0103] S4, the edge node module collects power data, brightness data or heat data in the past time period of the state fluctuation information, and generates an abnormal data set;

[0104] Finally, the edge node module transmits the abnormal data set together with the power data, brightness data, and heat data to the platform module;

[0105] Compared with Example 1, the edge node module in Example 2 generates secondary comparison data based on the average values ​​of the brightness data and heat data of each street lamp in the corresponding area, and comprehensively analyzes the brightness data and heat data of each street lamp in the corresponding area, thereby reducing factors such as environmental changes and differences in street lamp types, and avoiding marking a street lamp as faulty in isolation;

[0106] When the edge node module generates secondary comparison data, it smoothes the brightness data and heat data of each street lamp in the corresponding area to reduce the impact of high discrete brightness data or heat data on the overall situation.

[0107] Before generating secondary comparison data, the edge node calculates whether the change trends of the brightness data and heat data in the corresponding area in the past time window are consistent. The calculation method of the change trend includes:

[0108] a1. Calculate the first-order difference of the brightness data and heat data of each street lamp;

[0109] The formula for the first-order difference of brightness data is: ΔL i (t) = L i (t)-L i (t-1);

[0110] Among them, ΔL i (t) is the brightness data change of the i-th street lamp at time t;

[0111] L i (t) is the brightness data of the i-th street lamp at time t;

[0112] L i (t-1) is the brightness data of the i-th street light at time t-1;

[0113] The formula for the first-order difference of heat data is: ΔT i (t) = T i (t)-T i (t-1);

[0114] ΔT i (t) is the change in heat data of the i-th street lamp at time t;

[0115] T i (t) is the heat data of the i-th street lamp at time t;

[0116] T i(t-1) is the heat data of the i-th street lamp at time t-1;

[0117] a2. Count the mean values ​​of the brightness data changes and the heat data changes of each street lamp in the area, and calculate the standard deviation of the brightness data changes and the standard deviation of the heat data changes;

[0118] The mean formula of the brightness data change is:

[0119] Among them, μΔL(t) is the mean value of the brightness data change of all street lamps in the area at time t;

[0120] N is the number of street lights in the area;

[0121] ΔL i (t) is the brightness data change of the i-th street lamp at time t;

[0122] The standard deviation of the brightness data change is:

[0123]

[0124] Among them, σΔL(t) is the standard deviation of the brightness data change of all street lamps in the corresponding area at time t, which reflects the fluctuation range of the brightness data. The larger its value, the greater the change range;

[0125] The mean formula of the heat data change is:

[0126] Among them, μΔT(t) is the mean value of the change in heat data of all street lamps in the area at time t;

[0127] N is the number of street lights in the area;

[0128] ΔT i (t) is the brightness data change of the i-th street lamp at time t;

[0129] The standard deviation of the change in heat data is:

[0130]

[0131] Among them, σΔT(t) is the standard deviation of the change in the heat data of all street lamps in the corresponding area at time t, which reflects the fluctuation range of the heat data. The larger its value, the greater the change range;

[0132] a3. Calculate whether the change trend of the brightness data of each street lamp in the corresponding area over the past W moments is consistent with the change trend of the heat data;

[0133] In this embodiment, the Pearson correlation coefficient is used to measure whether the change trends between brightness and heat are consistent, and the formula is:

[0134]

[0135] Among them, ΔL i (t) is the brightness data change of the i-th street lamp at time t;

[0136] ΔT i (t) is the change in heat data of the i-th street lamp at time t;

[0137] μΔL(t) is the mean value of the brightness data change of all street lamps in the area at time t;

[0138] μΔT(t) is the mean value of the change in heat data of all street lamps in the area at time t;

[0139] a4. Average all the results of the Pearson correlation coefficient in a3 within W moments. The formula used is:

[0140] Among them, μ ρL,T (W) is the average Pearson correlation coefficient within W moments;

[0141] W end and W start is the end and start time of W moment;

[0142] μ ρL,T If (W) is close to 1, it means that within W moments, the change trends of brightness data and heat data are highly consistent;

[0143] If it is close to 0, it means that the relationship between the changing trends of brightness data and heat data is weak, and there may be anomalies;

[0144] a5. Set the error range e as the condition for judging the changing trend of the brightness data and heat data of each street lamp. The judgment condition is: |μ ρL,T (W)-1<e;

[0145] If the Pearson correlation coefficient μ ρL,T If the difference between (W) and 1 is less than the preset error range e, it is considered that the change trend of the brightness data and the heat data at time W is consistent and can be used for secondary comparison data;

[0146] If the change trends of brightness data and heat data are inconsistent, the step of generating secondary comparison data is cancelled;

[0147] And the edge node module is provided with a buffer zone;

[0148] When the brightness data or heat data does not match the screening model but does not exceed the secondary comparison data, the edge node module marks the brightness data or heat data as "unconfirmed abnormality" and transmits it to the buffer. In the subsequent N time windows, if the brightness data or heat data of the street lamp exceeds the secondary comparison data, or the number of "unconfirmed abnormalities" marked reaches the preset judgment threshold, the "unconfirmed abnormality" mark of the corresponding brightness data or heat data is converted to an abnormal mark;

[0149] The platform module receives the power data, brightness data and heat data from each edge node, stores the normal power data, brightness data and heat data, and matches the marked and timestamped parts of the power data, brightness data and heat data, that is, the state fluctuation information of each street lamp with the big data, predicts the failure of the street lamp, and issues a warning signal;

[0150] Based on the timestamp in the state fluctuation information, the platform module generates time series anomaly detection axes for the brightness data, heat data, and power data of the corresponding street lamps;

[0151] By calculating the time difference between each time stamp between the anomaly detection axes, the abnormal frequency of the corresponding street lamp is determined;

[0152] The platform module sets a plurality of frequency threshold ranges, and the platform module sets a corresponding computing resource allocation level based on each frequency threshold range;

[0153] When the frequency threshold range is low, it indicates that the frequency of state fluctuation information of the corresponding street lamp is high. The platform module will give priority to mobilizing more computing resources to increase the monitoring intensity of the state fluctuation information of the street lamp and match it with the possible fault modes in the big data in a timely manner;

[0154] In addition, the platform module builds a multi-dimensional fault mode library based on the historical fault data of street lamps, and associates the brightness data, heat data and power data of street lamps with faults;

[0155] When the platform module performs correlation, it compares the brightness data, heat data and power data with the fault type and extracts data features related to the fault type, including brightness data fluctuations, heat data fluctuations and power data fluctuations.

[0156] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A street lamp status remote monitoring system based on big data, including a local module, an edge node module and a platform module, characterized in that: There are multiple local modules, and each local module is matched with each street lamp respectively, and the local module includes: The light sensing unit monitors the brightness and heat of the street lamps and generates corresponding brightness data and heat data; The carrier unit collects the power data of the street lamps and transmits the power data, brightness data and heat data of the street lamps to the corresponding edge node modules; The edge node module includes: The comparison unit regularly updates the screening model and compares the screening model with the power data, brightness data and heat data of each street lamp. The comparison unit marks the power data, brightness data and heat data that do not match the screening model as abnormal and adds a timestamp. The transmission unit transmits the marked and time-stamped power data, brightness data and heat data and the normal power data, brightness data and heat data to the platform module; After receiving the power data, brightness data and heat data from each edge node, the platform module matches the abnormally marked power data, brightness data and heat data with the big data, predicts the failure of the street lamp, and issues a warning signal.

2. According to the big data-based street lamp status remote monitoring system of claim 1, it is characterized in that: The comparison method of the screening model comprises the following steps: S1. Preset the normal range of power data, brightness data and heat data of each street lamp in the corresponding area; S2. After the power data, brightness data, and heat data fluctuate beyond the corresponding normal range, mark them, and add a timestamp to the marked power data, brightness data, or heat data to generate state fluctuation information; S3. After generating the data fluctuation information, the edge node module adds location information to the corresponding data fluctuation information based on the actual geographical location of each street lamp; S4, the edge node module collects power data, brightness data or heat data in the past time period of the state fluctuation information, and generates an abnormal data set; Finally, the edge node module transmits the abnormal data set together with the power data, brightness data and heat data to the platform module.

3. According to the big data-based remote monitoring system for street lamp status according to claim 2, it is characterized in that: The edge node module generates secondary comparison data based on the average values ​​of brightness data and heat data of each street lamp in the corresponding area; And when the edge node module generates secondary comparison data, the brightness data and heat data of each street lamp in the corresponding area are smoothed.

4. According to the big data-based street lamp status remote monitoring system of claim 3, it is characterized in that: The edge node module is provided with a buffer zone; When the brightness data or heat data does not match the screening model but does not exceed the secondary comparison data, the edge node module marks the brightness data or heat data as "unconfirmed abnormality" and transmits it to the buffer; In the subsequent N time windows, if the brightness data or heat data of the street lamp exceeds the secondary comparison data, or the number marked as "unconfirmed abnormality" reaches the preset judgment threshold, the "unconfirmed abnormality" mark of the corresponding brightness data or heat data will be converted to an abnormal mark.

5. The street lamp status remote monitoring system based on big data according to claim 4 is characterized in that: Before generating the secondary comparison data, the edge node calculates whether the change trends of the brightness data and the heat data in the corresponding area in the past time window are consistent. The calculation method of the change trend includes: a1. Calculate the first-order difference of the brightness data and heat data of each street lamp; a2. Based on the first-order difference of the brightness data and heat data of each street lamp, the mean of the brightness data change and heat data change of each street lamp in the area are counted, and the standard deviation of the brightness data change and the standard deviation of the heat data change are calculated; a3. Calculate the consistency between the brightness data change trend of each street lamp in the corresponding area at the past W moments and the heat data change trend by using the Pearson correlation coefficient; The formula is: Where, ρL,T(t) is the Pearson correlation coefficient; ΔL i (t) is the brightness data change of the i-th street lamp at time t; ΔT i (t) is the change in heat data of the i-th street lamp at time t; μΔL(t) is the mean value of the brightness data change of all street lamps in the area at time t; μΔT(t) is the mean value of the change in heat data of all street lamps in the area at time t; a4. Average all the results of the Pearson correlation coefficient in a3 within W moments. The formula used is: Where, ρL,T(t) is the Pearson correlation coefficient; μ ρL,T (W) is the average Pearson correlation coefficient within W moments; W end and W start is the end and start time of W moment; a5. Set the error range e and generate the judgment conditions: | μ ρL,T (W)-1<e; If the Pearson correlation coefficient μ ρL,T If the difference between (W) and 1 is less than the preset error range e, it is considered that the change trend of the brightness data and the heat data at time W is consistent and is used for secondary comparison of data; If the changing trends of brightness data and heat data are inconsistent, the generation of secondary comparison data is canceled.

6. The street lamp status remote monitoring system based on big data according to claim 1 is characterized in that: The platform module generates an abnormal detection axis of a time series for the brightness data, heat data and power data of the corresponding street lamps based on the timestamp in the state fluctuation information; By calculating the time difference between each time stamp between the anomaly detection axes, the abnormal frequency of the corresponding street light is determined.

7. A street lamp status remote monitoring system and method based on big data according to claim 6, characterized in that: The platform module sets a plurality of frequency threshold ranges, and the platform module sets a corresponding computing resource allocation level based on each frequency threshold range; When the frequency threshold range is low, it indicates that the frequency of status fluctuation information of the corresponding street lamp is high. The platform module will give priority to mobilizing more computing resources to increase the monitoring intensity of the street lamp status fluctuation information and match it with possible fault modes in big data in a timely manner.

8. The street lamp status remote monitoring system and method based on big data according to claim 1 is characterized in that: The platform module builds a multi-dimensional fault mode library based on the historical fault data of street lamps, and associates the brightness data, heat data and power data of street lamps with faults; When the platform module performs the association, the brightness data, the heat data and the power data are compared with the fault type, and data features related to the fault type are extracted, including brightness data fluctuations, heat data fluctuations and power data fluctuations.

9. A method for remote monitoring of street lamp status based on big data according to any one of claims 1 to 8, characterized in that: The monitoring method comprises the following steps: Step 1: Monitor each street lamp through multiple local modules to obtain the brightness data, heat data and power data of the corresponding street lamp; Step 2: Receive the brightness data, heat data and power data of each local module in the corresponding area through the edge node module, and compare the brightness data, heat data and power data of each street lamp through the built-in screening model of the edge node module; Step 3: The edge node module marks the power data, brightness data, and heat data that do not match the screening model as abnormal, adds a timestamp, and generates state fluctuation information; Step 4: The platform module receives the power data, brightness data and heat data from each edge node, stores the normal power data, brightness data and heat data, and matches the marked and timestamped parts of the power data, brightness data and heat data, that is, the status fluctuation information of each street lamp with the big data, predicts the failure of the street lamp, and issues a warning signal.

Images