A LED street lamp for smart city construction
By designing the control system in the smart street light system, using the LSTM model for fault prediction and judgment, and disconnecting the electrical connection of the fault end through the opening and closing mechanism, the problem of fault location and processing in the existing technology is solved, and the early stage of street lamp failure is realized. Identification and early warning reduce operation and maintenance costs and improve system stability.
Patent Information
- Application Number
- CN202410740516.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-09
- Publication Date
- 2025-01-21
- Estimated Expiration
- 2044-06-09
AI Technical Summary
Existing smart street light systems are difficult to quickly locate and deal with when a fault occurs, which may lead to the spread of faults, safety risks and increased operation and maintenance costs.
A control system for LED street lights is designed. By obtaining the real-time parameters of the energy supply end, energy storage end and LED, the LSTM model is used for prediction and fault judgment to achieve early identification and early warning of faults, and it is actively disconnected through the opening and closing mechanism. Electrical connection at the fault end to prevent fault expansion and energy waste.
It achieves rapid and accurate identification and processing of street light faults, reduces operation and maintenance costs, improves the stability and safety of the street light system, and ensures the continuity of lighting services.
Smart Images

Figure CN118510115B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of smart street lamps, and in particular to an LED street lamp for smart city construction. Background Art
[0002] Nowadays, as an innovative product integrating advanced information technology and lighting technology, smart street lights are widely used in urban construction, aiming to improve the energy efficiency, intelligent management level and multi-functional services of urban lighting systems. In addition, the use of LED light sources can greatly reduce energy consumption compared to traditional street lights to adapt to the city's energy saving needs.
[0003] Because street lamps operate in outdoor environments, on the one hand, once a component or circuit failure, short circuit or overload occurs, if the protection mechanism and protective measures are not timely, it may cause the fault to spread or safety risks caused by continuous power supply, such as fire; on the other hand, relying on manual experience to locate the type and location of the fault is not only time-consuming and labor-intensive, but also has low accuracy, which can easily lead to excessive or insufficient maintenance, and increase operation and maintenance costs. For this reason, we propose an LED street lamp for smart city construction. Summary of the invention
[0004] 1. Technical issues to be resolved
[0005] In view of the deficiencies in the prior art, the present invention provides an LED street lamp for smart city construction, which has the advantages of fault classification protection, early warning, and continuous power supply during faults, and can effectively solve the problems in the background technology.
[0006] (II) Technical solution
[0007] To achieve the above object, the technical solution adopted by the present invention is: a control system for LED street lamps, comprising:
[0008] The acquisition module includes a voltage value V of the energy supply terminal for acquiring the voltage value V of the energy supply terminal within the k time period. s k , current value I s k , the voltage value of the energy storage end V e k , current value I e k The light intensity L of the LED k , generate parameter set A for k time periods k , and the parameter set A for the k time period k a processing unit for preprocessing; and a parameter set A for the k time periods after preprocessing k Perform noise cleaning and extract the calculation unit of the corresponding parameter features; and obtain at least the real-time parameter set A a (V sa ,I s a ,V e a ,I e a ,L a );
[0009] A main control module, wherein the main control module comprises a model building unit, a parameter prediction unit and a fault judgment unit.
[0010] Preferably, the model building unit is used for parameter sets A based on several groups of k time periods. k The corresponding parameter features extracted are used to train and optimize the LSTM model to build a prediction model;
[0011] A parameter prediction unit is used to apply the prediction model and input a real-time parameter set A a (V s a ,I s a ,V e a ,I e a ,L a ) to make predictions and output the corresponding prediction parameter set A p (V s p ,I s p ,V e p ,I e p ,L p );
[0012] A fault judgment unit is used to compare the real-time parameter set A a With the prediction parameter set A p ,The threshold method is used to identify the fault type and generate a response strategy.
[0013] Preferably, the parameter set A based on several groups of k time periods k The corresponding parameter features extracted from the training and optimization of the LSTM model are used to build a prediction model:
[0014] Data processing: for several groups of parameter sets A of k time periods k The corresponding parameter features extracted are normalized to eliminate the dimension effect;
[0015] Sequence division: dividing the corresponding parameter features of the normalization process into a training set, a validation set and a test set, with the proportions being 70%, 15% and 15%, or 60%, 20% and 20%, respectively;
[0016] Model building: Use a deep learning framework (such as TensorFlow or PyTorch) to build an LSTM model and set the appropriate number of hidden layers, number of neurons, and learning rate;
[0017] Model training: training the model through back propagation and gradient descent optimization algorithms based on the corresponding parameter features of the training set until the loss of the validation set no longer decreases significantly;
[0018] Model evaluation: evaluating the prediction accuracy of the model based on the test set;
[0019] Among them, the mean square error is used as the loss function LossFunction:
[0020]
[0021] Among them, (A p -A a )=(V s p1 -V s a1 )+(V s p2 -V s a2 )+......+(V s pN -V s aN )
[0022] or (A p -A a )=(I s p1 -I s a1 )+(I s p2 -I s a2 )+......+(I s pN -I s aN )
[0023] or (A p -A a )=(V e p1 -V e a1 )+(V e p2 -V e a2 )+......+(V epN -V e aN )
[0024] or (A p -A a )=(I e p1 -I e a1 )+(I e p2 -I e a2 )+......+(I e pN -I e aN )
[0025] or (A p -A a )=(L p1 -L a1 )+(L p2 -L a2 )+......+(L pN -L aN );
[0026] Where N is the total number of parameters in the parameter set.
[0027] Preferably, the fault judgment unit is used to compare the real-time parameter set A a With the prediction parameter set A p , use the threshold method to identify the fault type and generate a response strategy, including:
[0028] like or Generate a type of response strategy and a type of alarm information;
[0029] like or Generate second-class response strategies and second-class alarm information;
[0030] If L a -L p >δ L , three types of alarm information are generated.
[0031] As a preference, if and or and If the conditions of and are met at the same time, it is more strongly indicated that there is a fault at the energy supply end or the energy storage end, and if L a Much lower than L p, especially when both the energy supply and storage are normal, it may indicate that the LED is damaged or aged. a The trend over time shows a sharp decline or fluctuation, even if condition L is not met. a -L p >δ L , should be regarded as a potential fault signal.
[0032] Preferably, the main control module further includes a collaborative decision-making unit:
[0033] Used to call the opening and closing mechanism according to the first type of response strategy, so that the energy storage end of the current faulty street lamp is electrically connected to the energy supply end of any other street lamp in the street lamp ring network, and call the communication module to send a type of alarm information to the client;
[0034] Used to call the opening and closing mechanism according to the second type of response strategy, so that the energy supply end of the current faulty street lamp is electrically connected to the energy storage end of any other street lamp in the street lamp ring network, and the energy storage end of this street lamp is electrically connected to the LED of the faulty street lamp, call the communication module, and send the second type of alarm information to the client;
[0035] It is used to call the communication module according to the three types of response strategies and send the three types of alarm information to the client.
[0036] Preferably, the communication module is based on any one or a combination of several protocols such as Zigbee, LoRa, NB-IoT, 4G Cat.1 or PLC.
[0037] Preferably, the acquisition module comprises:
[0038] A processing unit, used to obtain at least the voltage value V of the energy supply terminal within a k time period s k , current value I s k , the voltage value of the energy storage end V e k , current value I e k The light intensity L of the LED k , generate parameter set A for k time periods k , and the parameter set A for the k time period k Perform pre-processing;
[0039] A calculation unit for the parameter set A of the preprocessed k time periods k Perform noise cleaning and extract corresponding parameter features.
[0040] Preferably, the parameter set A for the k time period k Preprocessing is performed, including:
[0041] Missing value handling: Check parameter set A for k time periods k Check whether there are null values or missing records in the data, and adopt filling strategies according to the situation, such as using previous and next value interpolation, average value filling, or filling based on time series model prediction;
[0042] Outlier detection: Use statistical methods (such as box plot analysis, Z-score or IQR) to analyze the parameter set A of k time periods. k Identify and process outliers to prevent them from interfering with the analysis results. Outliers can be marked, replaced, or deleted, depending on the cause and the set of preceding and following parameters.
[0043] Timeline calibration: Ensure that the parameter set A for k time periods k The different parameter data within the system are precisely aligned on the time axis. In the actual implementation process, given the different sampling rates of different sensors, it is necessary to resample or interpolate the time series to ensure the synchronization of the data.
[0044] Preferably, the calculation unit is used to calculate the parameter set A of the preprocessed k time period. k Perform noise cleaning and extract corresponding parameter features, including:
[0045] Feature extraction: Use the moving average algorithm to extract the feature from the parameter set A of the preprocessed k time periods k The corresponding parameter features, such as maximum value, minimum value, mean value, standard deviation, peak-to-valley difference, and time series characteristics (such as sliding average and lag characteristics) are extracted from the data. The specific calculation of the mean value is as follows:
[0046] Average voltage at the energy supply end during time period k The average voltage of the energy storage terminal Calculation:
[0047] (1)
[0048] (1) In the formula, N V is the total number of voltage acquisitions in the k time period, is the voltage measurement value at time i in time period k, N is the total number of time periods, and i is the number of each time period;
[0049] As mentioned above, for the average voltage at the energy storage end The feature extraction algorithm is equivalent to the average voltage at the energy supply end The feature extraction algorithm for the LED light intensity L k The feature extraction algorithm is also equivalent to the energy supply end V in the k time period e k Feature extraction algorithm;
[0050] Effective current at the energy supply end during time period k The effective current of the energy storage terminal Calculation:
[0051] (2)
[0052] (2) In the formula, N I is the total number of current acquisitions in the k time period, is the voltage measurement value at time i in time period k, is the RMS value of the current in time period k, N is the total number of moments, and i is the number of each moment;
[0053] As mentioned above, for the effective current of the energy storage end The feature extraction algorithm is equivalent to the effective current of the energy supply end Special diagnosis extraction algorithm.
[0054] The present invention also provides an LED street lamp for smart city construction, comprising an energy supply end, an energy storage end and an LED, characterized in that it also comprises a controller with a built-in control system of the LED street lamp and a voltage collector, a current collector, a light intensity sensor and an opening and closing mechanism electrically connected to the controller, wherein:
[0055] Voltage collector, used to collect real-time voltage parameters of the energy supply end, energy storage end or LED;
[0056] Current collector, used to collect real-time current parameters of the energy supply end, energy storage end or LED;
[0057] Light intensity sensor, used to collect the real-time light intensity of LED;
[0058] The light intensity sensor is fixedly connected to a portion of the street lamp adjacent to the LED through a housing;
[0059] The energy storage end, controller and opening and closing mechanism are fixedly connected inside the bottom section of the street lamp, which is more convenient for subsequent operation and maintenance operations by operation and maintenance personnel;
[0060] The opening and closing mechanism is used to control the electrical connection between the energy supply end, the energy storage end and the LED in at least two street lamps.
[0061] As a preference, the energy supply end includes not only solar power generation, but also wind power generation or access to the municipal power grid. Correspondingly, the acquisition module needs to add corresponding sensors.
[0062] Preferably, the opening and closing mechanism includes a n terminals, and a n One of the terminals that is electrically connected to the LED on the current street lamp is located in the center, and the other a n- 1 terminal is based on a ring array of one terminal that is electrically connected to the LED on the current street lamp to form a ring circuit;
[0063] It also includes a rotating ring and a motor, wherein the rotating ring and a n Each terminal is made of conductive material, and the rotating ring is connected to a n The wiring terminals are movable and fitted, one of the wiring terminals located in the center is fixedly connected to the rotating ring, and the rotating ring is externally connected to the motor through the insulating seat;
[0064] The relationship between the number of street lamps and the connection terminals is:
[0065] a n =2n+1.
[0066] Preferably, when there is a n When there are multiple terminals, the rotating ring can fit them all at once Terminals.
[0067] Preferably, the terminal comprises a conductive column 1 and a conductive column 2, and an insulating spacer 2 is provided between the conductive column 1 and the conductive column 2; the rotating ring comprises a conductive ring 1 and a conductive ring 2, and an insulating spacer 1 is provided between the conductive ring 1 and the conductive ring 2;
[0068] The rotating ring is based on a conductive ring and a n The conductive pillars in the two terminals are movably fitted, and the rotating ring is based on the conductive rings on the two terminals and a n The conductive posts in the two terminals form a movable fit.
[0069] Preferably, an insulating seat is provided between the motor and the insulating spacer, and the rotor of the motor is fixedly connected to the rotating ring via the insulating seat.
[0070] Preferably, the motor is electrically connected to the control system.
[0071] Preferably, the opening and closing mechanism further includes a top shell and a bottom shell, the top shell and the bottom shell constitute the outer shell of the opening and closing mechanism, and the main body of the motor is fixedly connected to the top shell.
[0072] Preferably, the conductive ring 2 is rotatably connected to the bottom shell based on a terminal located in the center.
[0073] (III) Beneficial effects
[0074] Compared with the prior art, the present invention provides a control system for an LED street lamp and an LED street lamp for smart city construction, which have the following beneficial effects:
[0075] 1. This LED street light control system can establish a comprehensive performance parameter set A by continuously monitoring and collecting detailed parameters of the energy supply end, energy storage end and LED within a k time period. k , by the performance parameter set A k Preprocessing and noise cleaning can effectively remove errors and interference signals in the monitoring process, ensure the accuracy and reliability of monitoring data, and use the prediction model to predict the real-time parameter set A. a Through analysis, potential street light failure trends can be discovered in advance, achieving early identification and early warning of street light failures;
[0076] And, based on the prediction parameter set A output by the prediction model p With real-time parameter set A a The comparative analysis can quickly and accurately identify the specific types of faults, including power supply faults, energy storage faults and LED faults, thereby forming an accurate fault location capability, allowing the operation and maintenance team to formulate targeted maintenance plans, saving a lot of inspection work, saving manpower and material resources, and making operation and maintenance operations more efficient.
[0077] 2. This kind of LED street lamp for smart city construction can actively disconnect the electrical connection between the energy supply end, the energy storage end and the LED by controlling the opening and closing mechanism through the control system, which can effectively prevent the expansion of circuit faults and avoid the energy waste caused by invalid power consumption at the fault end. At the same time, it can effectively avoid the fire risk caused by electrical short circuit, overload and other conditions, so as to ensure public safety and the construction safety of operation and maintenance personnel in the subsequent operation and maintenance process;
[0078] Correspondingly, because the electrical connection between the LED on the faulty street lamp and the faulty end can be disconnected in time, the abnormal current flow caused by the local fault can be prevented to protect other circuit components from being damaged, thereby extending the service life of the entire street lamp system and reducing operation and maintenance costs;
[0079] In addition, the opening and closing mechanism is used to change the energy supply relationship between the energy supply end, energy storage end or LED of the faulty street lamp and other normally functioning street lamps, so that other normally functioning street lamps can temporarily power the faulty street lamp to ensure the continuity of lighting services, improve the overall stability and maintainability of the street lamp system, and indirectly reduce the impact on night traffic and pedestrian safety. BRIEF DESCRIPTION OF THE DRAWINGS
[0080] Figure 1 This is a schematic diagram of the overall structure of an LED street lamp for smart city construction according to the present invention.
[0081] Figure 2 This is a schematic diagram of the partial structure of an LED street lamp for smart city construction according to the present invention.
[0082] Figure 3 This is a structural explosion diagram of an opening and closing mechanism for an LED street lamp for smart city construction according to the present invention.
[0083] Figure 4 This is a structural explosion diagram of a rotating ring in an opening and closing mechanism for an LED street lamp for smart city construction according to the present invention.
[0084] Figure 5 This is a structural state diagram of the opening and closing mechanism of an LED street lamp for smart city construction of the present invention when the LED street lamp is electrically disconnected from another street lamp in a street lamp ring network.
[0085] Figure 6 This is a schematic diagram of the structural state transformation of an opening and closing mechanism for an LED street lamp for smart city construction according to the present invention based on the number of connected street lamps.
[0086] Figure 7 The figure is a schematic diagram of the line connection of an LED street lamp for smart city construction in a street lamp ring network according to the present invention.
[0087] Figure 8 This is a system block diagram of an LED street lamp for smart city construction according to the present invention.
[0088] Fig. 9 The system block diagram of a control system of an LED street lamp of the present invention.
[0089] Fig.10 The present invention is a flowchart of an execution method of a control system for an LED street lamp.
[0090] Fig.11 The present invention is a flowchart of an execution method of an acquisition module in a control system of an LED street lamp.
[0091] Fig.12 The present invention is a flowchart of an execution method of a main control module in a control system of an LED street lamp.
[0092] In the figure:
[0093] 100, control system; 200, street light;
[0094] 110, acquisition module; 120, main control module; 130, communication module;
[0095] 210, energy supply end; 220, energy storage end; 230, light intensity sensor; 240, LED; 250, opening and closing mechanism; 260, controller;
[0096] 251, rotating ring; 252, terminal; 253, motor; 254, insulating seat; 255, bottom shell; 256, top shell;
[0097] 2511, conductive ring one; 2512, conductive ring two; 2513, insulating barrier one;
[0098] 2521. Conductive column one; 2522. Conductive column two; 2523. Insulating barrier two. DETAILED DESCRIPTION
[0099] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the present invention will be further described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0100] Embodiment 1
[0101] In order to address the shortcomings of the existing technology, such as Figure 8 As shown, the present invention provides a control system 100 for an LED street lamp, comprising:
[0102] The acquisition module 110 is used to obtain at least the voltage value V of the energy supply terminal 210 within a time period k. s k , current value I s k , the voltage value V of the energy storage terminal 220 e k , current value I e k The light intensity L of LED 240 k , generate parameter set A for k time periods k , and the parameter set A for the k time period k Preprocessing is performed, and the parameter set A of the k time period after preprocessing is k Perform noise cleaning and extract corresponding parameter features; and obtain at least a real-time parameter set A a (V s a ,I s a ,V e a ,I e a ,L a ), the real-time parameter set includes the real-time voltage value V of the energy supply end 210 s a , real-time current value I s a , energy storage end 220 real-time voltage value V e a , real-time current value I e aWith the real-time light intensity L a .
[0103] Specifically, Fig. 9 As shown, an acquisition module 110 in a control system 100 of an LED street lamp includes:
[0104] A processing unit, used to obtain at least the voltage value V of the energy supply terminal 210 within a time period k s k , current value I s k , the voltage value V of the energy storage terminal 220 e k , current value I e k The light intensity L of LED 240 k , generate parameter set A for k time periods k , and the parameter set A for the k time period k Preprocessing is performed, wherein the preprocessing steps are as follows Fig.11 The method flow shown includes:
[0105] Missing value handling: Check parameter set A for k time periods k Check whether there are null values or missing records in the data, and adopt filling strategies according to the situation, such as using previous and next value interpolation, average value filling, or filling based on time series model prediction;
[0106] Outlier detection: Use statistical methods (such as box plot analysis, Z-score or IQR) to analyze the parameter set A of k time periods. k Identify and process outliers to prevent them from interfering with the analysis results. Outliers can be marked, replaced, or deleted, depending on the cause and the set of preceding and following parameters.
[0107] Timeline calibration: Ensure that the parameter set A for k time periods k Different parameter data in the system are precisely aligned on the time axis. In the actual implementation process, given the different sampling rates of different sensors, time series resampling or interpolation processing is required to ensure data synchronization.
[0108] A calculation unit for the parameter set A of the preprocessed k time periods k Perform noise cleaning and extract corresponding parameter features, where noise cleaning includes:
[0109] Feature extraction: Use the moving average algorithm to extract the feature from the parameter set A of the preprocessed k time periods k The corresponding parameter features, such as maximum value, minimum value, mean value, standard deviation, peak-to-valley difference, and time series characteristics (such as sliding average and lag characteristics) are extracted from the data. The specific calculation of the mean value is as follows:
[0110] Average voltage of the energy supply terminal 210 during the k period The average voltage of the energy storage terminal 220 Calculation:
[0111] (1)
[0112] (1) In the formula, N V is the total number of voltage acquisitions in the k time period, is the voltage measurement value at time i in time period k, N is the total number of time periods, and i is the number of each time period;
[0113] As mentioned above, for the average voltage of the energy storage terminal 220 The feature extraction algorithm is equivalent to the average voltage of 210 at the energy supply end. The feature extraction algorithm for the light intensity L of LED240 k The feature extraction algorithm is also equivalent to the 210V power supply end in the k time period. e k Feature extraction algorithm;
[0114] Effective current of energy supply terminal 210 in time period k The effective current of the energy storage terminal is 220 Calculation:
[0115] (2)
[0116] (2) In the formula, N I is the total number of current acquisitions in the k time period, is the voltage measurement value at time i in time period k, is the RMS value of the current in time period k, N is the total number of moments, and i is the number of each moment;
[0117] As mentioned above, for the effective current of the energy storage terminal 220 The feature extraction algorithm is equivalent to the effective current of 210 at the energy supply end Special diagnosis extraction algorithm.
[0118] Among them, the use of moving average algorithm to clean the parameter set of k time periods can effectively reduce random fluctuations and noise in the data. On the one hand, it makes the data series smoother and reflects the long-term trend, which helps the model focus on the main patterns of the data rather than short-term fluctuations or measurement errors, thereby improving the overall quality and reliability of the data;
[0119] On the other hand, it can ensure that the extracted features are more stable and representative, especially by selecting the most influential feature set, which can ensure that the model learns the most critical information for the prediction target, indirectly reducing the dimension and further improving the training speed and efficiency of the prediction model. In addition, it allows operation and maintenance personnel to better understand the working principle of the prediction model, which helps to quickly locate problems in the subsequent operation and maintenance process.
[0120] And if Fig. 9 As shown, a main control module 120 in a control system 100 of an LED street lamp is used to build a prediction model based on the corresponding parameter characteristics, apply the trained prediction model, and input the real-time parameter set A a (V s a ,I s a ,V e a ,I e a ,L a ) to make predictions and output the corresponding prediction parameter set A p (V s p ,I s p ,V e p ,I e p ,L p ), compare the prediction parameter set A p With real-time parameter set A a , identify the fault type and execute the response strategy based on the fault type.
[0121] It is worth mentioning that by continuously monitoring and collecting detailed parameters of the energy supply end 210, the energy storage end 220 and the LED 240 in the k time period, the system can establish a comprehensive performance parameter set A k , by the performance parameter set A k Preprocessing and noise cleaning can effectively remove errors and interference signals in the monitoring process, ensure the accuracy and reliability of monitoring data, and use the prediction model to predict the real-time parameter set A. a Through analysis, potential street lamp 200 failure trends can be discovered in advance, and early identification and warning of street lamp 200 failures can be achieved;
[0122] And, based on the prediction parameter set A output by the prediction model p With real-time parameter set A aThe comparative analysis can quickly and accurately identify the specific types of faults, including energy supply end 210 faults, energy storage end 220 faults and LED 240 faults, thereby forming an accurate fault location capability, allowing the operation and maintenance team to formulate targeted maintenance plans, saving a lot of inspection work, saving manpower and material resources, and making operation and maintenance operations more efficient.
[0123] Specifically, Fig. 9 As shown, a main control module 120 in a control system 100 of an LED street lamp includes:
[0124] Model building unit, based on a set of parameters A for several groups of k time periods k The corresponding parameter features extracted from the training and optimization of the LSTM model are used to build a prediction model:
[0125] Data processing: parameter set A for several groups of k time periods k The corresponding parameter features extracted are normalized to eliminate the dimension effect;
[0126] Sequence division: The normalized corresponding parameter features are divided into training set, validation set and test set, with the ratios of 70%, 15% and 15% or 60%, 20% and 20% respectively;
[0127] Model building: Use a deep learning framework (such as TensorFlow or PyTorch) to build an LSTM model and set the appropriate number of hidden layers, number of neurons, and learning rate;
[0128] Model training: The model is trained based on the corresponding parameter characteristics of the training set through back propagation and gradient descent optimization algorithms until the validation set loss no longer decreases significantly;
[0129] Model evaluation: Evaluate the model prediction accuracy based on the test set;
[0130] Among them, the mean square error is used as the loss function LossFunction:
[0131]
[0132] Among them, (A p -A a )=(V s p1 -V s a1 )+(V s p2 -V s a2 )+......+(V s pN -V saN )
[0133] or (A p -A a )=(I s p1 -I s a1 )+(I s p2 -I s a2 )+......+(I s pN -I s aN )
[0134] or (A p -A a )=(V e p1 -V e a1 )+(V e p2 -V e a2 )+......+(V e pN -V e aN )
[0135] or (A p -A a )=(I e p1 -I e a1 )+(I e p2 -I e a2 )+......+(I e pN -I e aN )
[0136] or (A p -A a )=(L p1 -L a1 )+(L p2 -L a2 )+......+(L pN -L aN );
[0137] Where N is the total number of parameters in the parameter set.
[0138] It is worth mentioning that before building the prediction model, the parameter set A of the k time period is checked.k Whether there are null values or missing records in the data, use previous and next value interpolation, average value filling or prediction filling based on time series model to fill in null values or missing records, and use statistical methods to calculate the parameter set A of k time period k Identify and process outliers while ensuring that the parameter set A for the k time period k The different parameter data in the model are precisely aligned on the time axis. On the one hand, by filling in missing values and eliminating outliers and noise, it can ensure that the data during the prediction model training is sufficiently accurate to the greatest extent possible, which plays a decisive role in the accuracy of the prediction model. It can also better reflect the trend and periodicity of the data and help the model capture more complex time series relationships.
[0139] On the other hand, more accurate timeline alignment and feature extraction help the model capture the dynamic relationship and potential laws between parameters, especially the interaction between light intensity and energy supply, which is crucial for predicting energy demand, changes in light demand, and system performance evaluation. For example, the peak-to-valley difference can reveal the fluctuation characteristics of energy consumption or light intensity over time, which helps the model predict future trends more accurately.
[0140] In addition, aligned time series data and feature extraction can help quickly identify which specific parameter or system part first exhibits an abnormality, thereby accurately locating the source of the fault and accelerating maintenance response. Furthermore, different fault types will leave specific patterns in the data. Preprocessed data and extracted features can help develop classification models, identify fault types, and provide a basis for taking correct maintenance measures.
[0141] The parameter prediction unit inputs the real-time parameter set A according to the constructed prediction model a (V s a ,I s a ,V e a ,I e a ,L a ) to make predictions and output the corresponding prediction parameter set A p (V s p ,I s p ,V e p ,I e p ,L p ).
[0142] like Fig.12 In the method flow shown, the fault judgment unit specifically performs the following steps:
[0143] Compare real-time parameter set A a and the corresponding prediction parameter set A p , using the threshold method to identify fault types, including:
[0144] Using multi-index comprehensive evaluation, if (A a -A p )>δ A , then the street light fault is judged, where is the preset threshold;
[0145] The specific judgment of street light failure is:
[0146] like or It is determined that the energy supply end 210 is faulty, and a type of response strategy and a type of alarm information are generated;
[0147] like or It is determined that the energy storage terminal 220 is faulty, and a second-class response strategy and a second-class alarm information are generated;
[0148] If L a -L p >δ L , it is judged that LED240 is faulty and three types of alarm information are generated.
[0149] Meanwhile, if and or and If the conditions of and are met at the same time, it is more strongly indicated that there is a fault at the energy supply end 210 or the energy storage end 220, and if L a Much lower than L p , especially when both the energy supply end 210 and the energy storage end 220 are normal, it may indicate that the LED 240 is damaged or aged. a The trend over time shows a sharp decline or fluctuation, even if condition L is not met. a -L p >δ L , should be regarded as a potential fault signal;
[0150] In addition, during the above fault type identification process, if or or With L a -L p ≤δ L , then the corresponding function is judged to be normal.
[0151] like Fig.12 In the method flow shown, the collaborative decision-making unit specifically performs the following steps:
[0152] According to a type of response strategy, the opening and closing mechanism 250 is called to electrically connect the energy storage end 220 of the current faulty street lamp with the energy supply end 210 of any other street lamp in the street lamp ring network, and the communication module 130 is called to send a type of alarm information to the client;
[0153] According to the second type of response strategy, the opening and closing mechanism 250 is called to electrically connect the energy supply end 210 of the current faulty street lamp with the energy storage end 220 of any other street lamp in the street lamp ring network, and the energy storage end 220 of this street lamp is electrically connected to the LED 240 of the faulty street lamp, and the communication module 130 is called to send the second type of alarm information to the client;
[0154] According to the three types of response strategies, the communication module 130 is called to send the three types of alarm information to the client.
[0155] Wherein, the communication module 130 is based on any one or a combination of several protocols such as Zigbee, LoRa, NB-IoT, 4G Cat.1 or PLC;
[0156] Class I alarm information, Class II alarm information, and Class III alarm information can be classified and sent to the client in the form of different event urgency levels to effectively remind management personnel and make corresponding operation and maintenance operations.
[0157] In this embodiment, those skilled in the art will appreciate that, in order to ensure the normal use of the control system 100, it is necessary to connect the control system 100 to an independent control power supply or to connect it to the municipal power grid. At the same time, in order to further save costs, a single control system 100 can be used to control multiple street lamps 200, with a single control system 100 integrated control cabinet as the control center. This not only facilitates subsequent centralized operation and maintenance work, but also effectively prevents the normal operation of the control system 100 from being affected when the energy storage end 220 fails.
[0158] In addition, in this embodiment, there needs to be a line connection relationship between at least two street lamps, and street lamps with good functions adjacent to or closest to the faulty street lamp are preferentially selected for coordination, so as to reduce the power transmission distance and reduce the loss. Street lamps with good functions can be screened based on the prediction model and the support vector machine algorithm, and the greedy algorithm or genetic algorithm can be used for selection and optimization to select the optimal coordination scheme;
[0159] As a preferred embodiment, the street lamp further includes a positioning module to obtain geographical location information of the street lamp, including coordinates, adjacent facilities, layout of sidewalks and roadways, and relative distances between street lamps, etc., which is helpful for subsequent geospatial analysis;
[0160] In addition, those skilled in the art will understand that, in this embodiment, the street light ring network is merely an example of a street light arrangement method of the present invention and does not imply any limiting effect. In addition, branch layout, matrix layout or freestyle layout, etc., will not have any impact on the implementation of the present invention.
[0161] Embodiment 2
[0162] Based on the above embodiment 1, Figure 1 , 2 As shown, the present invention also provides an LED street lamp 200 for smart city construction, wherein a light intensity sensor 230 is fixedly connected to a portion of the street lamp 200 adjacent to an LED 240 through a housing;
[0163] The energy storage end 220, the controller 260 and the opening and closing mechanism 250 are fixedly connected inside the bottom section of the street lamp 200, which is more convenient for subsequent operation and maintenance operations by the operation and maintenance personnel.
[0164] It should be noted that the present invention is an LED street lamp for smart city construction. In addition to solar power generation, the energy supply end 210 also has wind power generation or access to the municipal power grid. Except for access to the municipal power grid, which requires judging the fault of the street lamp based on the energy storage end 220, wind power generation can also adopt the street lamp fault judgment method in the above-mentioned control system 100. Correspondingly, the acquisition module needs to add corresponding sensors.
[0165] The acquisition module includes at least a voltage sensor for real-time acquisition of the voltage V of the energy supply end 210 and the energy storage end 220, and can also acquire the voltage of LED240; it also includes a current sensor and a light sensor, a voltage sensor for real-time acquisition of the current I of the energy supply end 210 and the energy storage end 220, and can also acquire the current of LED240, and the light sensor is used to real-time acquire the light intensity L of LED240.
[0166] The opening and closing mechanism 250 mainly controls the hardware electrical connection between the energy supply end 210, the energy storage end 220 and the LED 240 of at least two groups of street lamps 200, and is called and controlled via the control system 100. Here, a faulty street lamp and another street lamp with normal function are taken as an example. When it is determined that the energy supply end 210 of the street lamp 200 is faulty, the control system 100 controls the opening and closing mechanism 250 to operate to disconnect the electrical connection between the faulty energy supply end 210 and the other ends. Similarly, when it is determined that the energy storage end 220 of the street lamp 200 is faulty, the control system 100 controls the opening and closing mechanism 250 to operate to disconnect the electrical connection between the faulty energy storage end 220 and the other ends. When it is determined that the LED 240 of the street lamp 200 is faulty, the control system 100 controls the opening and closing mechanism 250 to operate to disconnect the electrical connection between the LED 240 and the other ends.
[0167] Among them, if the energy supply end 210 of the current street lamp 200 fails, after the control change of the opening and closing mechanism 250, the energy storage end of the current street lamp 200 is electrically connected with the energy supply end 210 of another street lamp 200, and during the energy supply period (i.e. daytime), the energy storage end 220 of the faulty street lamp is supplied with energy via the energy supply end 210 of another street lamp 200, so that the energy storage end 220 on the faulty street lamp 200 can normally supply power to the LED 240 of this street lamp 200 during the lighting period;
[0168] If the energy storage end 220 of the current street lamp 200 fails, after the control change of the opening and closing mechanism 250, the energy supply end 210 of the current street lamp 200 is electrically connected to the energy storage end 220 of another street lamp 200, and the LED 240 of the current street lamp 200 is electrically connected to the energy storage end 220 of another street lamp 200. During the energy supply period (i.e., daytime), the energy storage end 220 of the other street lamp 200 is assisted in energy supply via the energy supply end 210 of the current street lamp 200, so that the energy storage end 220 of the other street lamp 200 can normally supply power to the LED 240 of the faulty street lamp 200 during the lighting period;
[0169] If the LED 240 of the current street lamp 200 fails, after the control of the opening and closing mechanism 250 is changed, the LED 240 of the current street lamp 200 is disconnected from the electrical connection with other ends to avoid energy waste caused by invalid power consumption after the LED 240 is damaged. At the same time, it can effectively avoid the risk of fire caused by overload and short circuit due to continuous power supply after the LED 240 is damaged.
[0170] It is understandable to those skilled in the art that, during normal operation of the street lamp 200 system, the electrical connection between at least two groups of street lamps 200 is based on the opening and closing mechanism 250 being in the disconnected state, thereby forming electrical isolation, so that the normal operation of the street lamps 200 will not affect each other.
[0171] It is worth mentioning that by controlling the opening and closing mechanism 250 through the control system 100 to actively disconnect the electrical connection between the energy supply end 210, the energy storage end 220 and the LED 240, the circuit fault can be effectively prevented from expanding, and the energy waste caused by the invalid power consumption of the fault end can be avoided. At the same time, the fire risk caused by electrical short circuit, overload and other conditions can be effectively avoided, so as to ensure public safety and the construction safety of the operation and maintenance personnel in the subsequent operation and maintenance process;
[0172] Correspondingly, since the electrical connection between the LED 240 on the faulty street lamp 200 and the faulty end can be disconnected in time, the abnormal current flow caused by the local fault can be prevented to protect other circuit components from being damaged, thereby extending the service life of the entire street lamp system and reducing the operation and maintenance costs;
[0173] In addition, the opening and closing mechanism 250 is used to transform the energy supply relationship between the energy supply end 210, energy storage end 220 or LED 240 of the faulty street lamp 200 and other street lamps 200 with normal functions, so that other street lamps 200 with normal functions can temporarily supply power to the faulty street lamp 200 to ensure the continuity of lighting services, improve the overall stability and maintainability of the street lamp system, and indirectly reduce the impact on night traffic and pedestrian safety.
[0174] Embodiment 3
[0175] Based on the above embodiment 1, considering that there are multiple line connections between at least two street lamps 200 and rapid disconnection is required, this embodiment is proposed:
[0176] Specifically, Figure 3 , 4 As shown, an opening and closing mechanism 250 for an LED street lamp 200 for smart city construction, the quantity relationship between the street lamp 200 and the connection terminal 252 is:
[0177] a n =2n+1;
[0178] When there is a n When the terminal 252 is connected, the rotating ring 251 can be simultaneously attached at one time. A terminal 252;
[0179] An insulating seat 254 is provided between the motor 253 and the rotating ring 251, and the rotor on the motor 253 is fixedly connected to the rotating ring 251 through the insulating seat 254; the motor 253 is electrically connected to the control system 100; the opening and closing mechanism 250 also includes a top shell 256 and a bottom shell 255, the top shell 256 and the bottom shell 255 constitute the outer shell of the opening and closing mechanism 250, and the main body of the motor 253 is fixedly connected to the top shell 256; the conductive ring 2512 is rotatably connected to the bottom shell 255 based on the terminal 252 located in the center.
[0180] It should be noted that the present invention is an LED street lamp 200 for smart city construction. Through the opening and closing mechanism 250, in this embodiment, there is a line connection between at least two street lamps 200, and the connection ends of the energy supply end 210 and the energy storage end 220 on the two street lamps 200 are integrated into the opening and closing mechanism 250 attached to the current street lamp 200, and the connection end of the LED 240 on the current street lamp 200 is also integrated into the opening and closing mechanism 250, and the other street lamp is the same;
[0181] In a normal use state, that is, there is no risk of failure of the street lamp 200, the structural state of the opening and closing mechanism 250 is as follows: Figure 5As shown, at this time, there is no electrical connection between the two street lamps 200, so that the operations of the two street lamps 200 will not interfere with each other. Relatively speaking, the control system 100 can independently and accurately collect performance parameters such as voltage, current and light intensity of each street lamp 200, effectively avoiding measurement errors caused by mutual interference between lines and ensuring the purity of the data.
[0182] In the meantime, there are two sets of power supply terminals 210 and energy storage terminals 220 and a single LED 240 line connection requirements, and the rotation angle of the rotor on the motor 253 is precisely controlled by the control system 100. In this embodiment, there are five wiring terminals 252, such as Figure 4 , 5 In the structure shown, the connection terminals 252 connected to the energy supply terminals 210 and the energy storage terminals 220 of the two street lamps 200 are respectively arranged opposite to each other. The connection terminal of the LED 240 of the current street lamp 200 is connected to the connection terminal 252 located in the center, thereby forming a ring connection circuit, in which the angle distribution of the four connection terminals 252 based on the connection terminal 252 located in the center is 90°;
[0183] In the actual implementation process, the motor 253 drives the rotating ring 251 to rotate. In the normal state of the street lamp 200, that is, Figure 5 In the structural state shown, the starting angle of the rotating ring 251 based on the annular connecting line is 45°;
[0184] If the energy supply end 210 of the current street lamp 200 fails, based on a type of coping strategy in the above-mentioned embodiment 1, the motor 253 drives the rotating ring 251 to rotate 135° counterclockwise or 225° clockwise, so that the energy storage end 220 and LED 240 of the current street lamp 200 are electrically connected with the energy supply end 210 and energy storage end 220 of another street lamp 200;
[0185] If the energy storage end 220 of the current street lamp 200 fails, based on the second type of response strategy in the above-mentioned embodiment 1, the motor 253 drives the rotating ring 251 to rotate 135° clockwise or 225° counterclockwise, so that the energy supply end 210 and LED 240 of the current street lamp 200 are electrically connected with the energy supply end 210 and energy storage end 220 of another street lamp 200.
[0186] As a preferred embodiment, in order to further improve the fault response capability of the street lamp 200, the street lamp 200 needs to be connected to at least two street lamps 200 by lines. Figure 6 As shown, the number of the wiring terminals 252 needs to be at least seven. Similar to the above, one of the wiring terminals 252 connected to the LED 240 on the current street lamp 200 is located at the center of the ring connection line, and the distribution angle is 60°;
[0187] In addition, in this embodiment, in order to facilitate the control of the rotating ring 251, the motor 253 is preferably a stepper motor, wherein based on the above-mentioned implementation mode, the driving angle of the stepper motor can be controlled via the control system 100 based on the classification algorithm, such as the classification algorithm based on fuzzy logic or preset control rules, based on a support vector machine, when the fault is relative to the first type of response strategy, the motor 253 rotates 135° counterclockwise, and when the fault is relative to the second type of response strategy, the motor 253 rotates 135° clockwise. Because the classification dimension is low and the number of parameters is small, it is sufficient to select a control module with control rules therein.
[0188] It can be understood by those skilled in the art that, as a known technology in the art, the line connection relationship between at least two groups of street lamps 200, and the arrangement of component installations, such as the existence of AC-DC converters, rectifiers or DC-DC converters, etc., are not elaborated on here. Moreover, in the actual installation process of the street lamps 200, such as the need for series or parallel connection of the street lamps 200, the positive and negative poles between the connection terminals at each end thereof are distinguished based on the conductive column 1 2521 and the conductive column 2 2522.
[0189] It is worth mentioning that, by forming a ring connection line based on at least five wiring terminals 252 through the rotating ring 251, a motor 253 can be used alone to realize multi-point flexible switching of the energy supply end 210, the energy storage end 220 and the LED 240 between at least two street lamps 200, thereby minimizing the need for additional switching devices and complex circuits, thereby reducing hardware costs;
[0190] In addition, due to the design of the ring connection line, all necessary electrical connections are integrated on one plane, without the need for redundant space layout, which saves the installation space of the line to the greatest extent and is more suitable for the aesthetic and environmental protection requirements of smart city construction;
[0191] In addition, the opening and closing mechanism 250 is controlled based on the control system 100, and the corresponding motor 253 rotation instructions can be quickly executed according to the fault type of the street lamp 200, which greatly simplifies the operation process and effectively reduces the possibility of manual intervention, indirectly improving the response speed and accuracy of the street lamp system.
[0192] In summary, the present invention provides flexible infrastructure support for the construction of smart cities, and is more convenient for future technology iteration and function expansion.
[0193] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for illustrating the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed for protection.
Claims
1. A control system (100) for an LED street lamp, characterized in that: include: An acquisition module (110), the acquisition module comprising a device for acquiring a voltage value V of the energy supply terminal (210) within a time period k s k , current value I s k , the voltage value V of the energy storage terminal (220) e k , current value I e k The light intensity L of the LED (240) k , generate parameter set A for k time periods k , and the parameter set A for the k time period k a processing unit for performing preprocessing; And the parameter set A of the k time periods after preprocessing k Perform noise cleaning and extract the calculation unit of the corresponding parameter features; and obtain at least the real-time parameter set A a (V s a ,I s a ,V e a ,I e a ,L a ); The main control module (120) includes a model building unit for training and tuning the LSTM model based on corresponding parameter features extracted from the parameter sets of the plurality of groups of the time periods to build a prediction model; and a model building unit for applying the prediction model and inputting a real-time parameter set A. a (V s a ,I s a ,V e a ,I e a ,L a ) to make predictions and output the corresponding prediction parameter set A p (V s p ,I s p ,V e p ,I e p ,L p ) parameter prediction unit; and for comparing the real-time parameter set A a With the prediction parameter set A p ,A fault judgment unit that uses the threshold method to identify the fault type and generate the response strategy; The fault judgment unit is used to compare the real-time parameter set A a With the prediction parameter set A p , use the threshold method to identify the fault type and generate a response strategy, including: If V s a -V s p >δ VS or I s a -I s p >δ IS ,generate a type of response strategy and a type of alarm information; If V e a -V e p >δ Ve or I e a -I e p >δ Ie ,generate second-class response strategies and second-class alarm information; If L a -L p >δ L ,generate three types of alarm information; The main control module (120) also includes a collaborative decision-making unit: Used to call the opening and closing mechanism (250) according to the first type of response strategy, so that the energy storage end (220) of the current faulty street lamp is electrically connected to the energy supply end (210) of any other street lamp in the street lamp ring network, and call the communication module (130) to send a type of alarm information to the client; Used to call the opening and closing mechanism (250) according to the second type of response strategy, so that the energy supply end (210) of the current faulty street lamp is electrically connected to the energy storage end (220) of any other street lamp in the street lamp ring network, and the energy storage end (220) of this street lamp is electrically connected to the LED (240) of the faulty street lamp, and call the communication module (130) to send the second type of alarm information to the client; Used to call the communication module (130) according to the three types of response strategies, and send the three types of alarm information to the client; The opening and closing mechanism (250) includes a n terminals (252), and a n One of the terminals (252) electrically connected to the LED (240) on the current street lamp (200) is located in the center, and the other terminals (252) electrically connected to the LED (240) on the street lamp (200) are located in the center. n - a connection terminal (252) is based on a ring array of one connection terminal (252) that is electrically connected to the LED (240) on the current street lamp (200), forming a ring circuit; It also includes a rotating ring (251) and a motor (253), wherein the rotating ring (251) and a n The terminals (252) are all made of conductive materials, and the rotating ring (251) and a n The wiring terminals (252) are movably attached, one of the wiring terminals (252) located in the center is fixedly connected to the rotating ring (251), and the rotating ring (251) is externally connected to the motor (253) through an insulating seat; The relationship between the number of street lamps (200) and the connection terminals (252) is: <h2 style=";text-align:left;direction:ltr">a<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> =2n+1.
2. The control system (100) of an LED street lamp according to claim 1, characterized in that: The parameter set A for the k time period k Preprocessing is performed, including: Missing value handling: Check parameter set A for k time periods k Check whether there are null values or missing records in the data, and use previous and next value interpolation, average value filling, or prediction filling based on time series model to fill null values or missing records; Outlier detection: Use statistical methods to detect the parameter set A in k time periods. k Identify and process outliers; Timeline calibration: Ensure that the parameter set A for k time periods k The different parameter data within are precisely aligned on the time axis.
3. The control system (100) of an LED street lamp according to claim 1, characterized in that: The calculation unit is used to calculate the parameter set A of the k time periods after preprocessing. k Perform noise cleaning and extract corresponding parameter features, including: The moving average algorithm is used to obtain the parameter set A of the preprocessed k time periods. k Extract the maximum, minimum, mean, standard deviation, or peak-to-valley difference from the data.
4. An LED street lamp (200) for smart city construction, comprising an energy supply end (210), an energy storage end (220) and an LED (240), characterized in that: It also includes a controller (260) having a built-in control system (100) for an LED street lamp as claimed in any one of claims 1 to 3, and a voltage collector, a current collector, a light intensity sensor (230) and an opening and closing mechanism (250) electrically connected to the controller (260), wherein: A voltage collector, used to collect real-time voltage parameters of the energy supply end (210), the energy storage end (220) or the LED (240); A current collector, used to collect real-time current parameters of the energy supply end (210), the energy storage end (220) or the LED (240); A light intensity sensor (230), used for collecting the real-time light intensity of the LED (240); The opening and closing mechanism (250) is used to control the mutual electrical connection between the energy supply end (210), the energy storage end (220) and the LED (240) in at least two street lamps.
5. The LED street lamp (200) for smart city construction according to claim 4, characterized in that: The terminal (252) includes a conductive column 1 (2521) and a conductive column 2 (2522), and an insulating spacer 2 (2523) is provided between the conductive column 1 (2521) and the conductive column 2 (2522); the rotating ring (251) includes a conductive ring 1 (2511) and a conductive ring 2 (2512), and an insulating spacer 1 (2513) is provided between the conductive ring 1 (2511) and the conductive ring 2 (2512); The rotating ring (251) is based on the conductive ring 1 (2511) and a n The conductive pillar 2 (2522) in the terminal (252) forms a movable fit, and the rotating ring (251) is based on the conductive ring 2 (2512) on it and a n A conductive post (2521) in a terminal (252) forms a movable fit.
Citation Information
Patent Citations
Street lamp predictive maintenance system and method based on deep learning
CN110766590A