Energy-saving control system of LED wall washer with built-in driver

Abstract

The invention discloses an energy-saving control system of an LED wall washer with a built-in driver, and relates to the field of energy-saving control of the LED wall washer, and the energy-saving control system comprises environment monitoring equipment which is used for monitoring data of an external environment of the LED wall washer; the data acquisition module is used for receiving and recording monitoring data of the environment monitoring equipment outside the LED wall washer; the data processing module is used for smoothing and denoising the received external environment data of the LED wall washer based on a first-order Kalman filtering algorithm; the'light weakness' detection module is used for judging and predicting whether the'light weakness' problem is generated in the LED wall washing lamp or not based on a machine learning algorithm model; and the internal voltage stabilizing equipment is used for balancing the internal voltage difference of the LED wall washing lamp and avoiding the problem of'weak light '. The LED wall washer has the advantages that the internal pressure difference of the LED wall washer is accurately and automatically controlled and adjusted, the phenomenon of light weakness is reduced, and the service life of the LED wall washer is prolonged.

Description

Technical Field

[0001] The invention relates to the field of energy-saving control of LED wall washer lamps, and in particular to an energy-saving control system of LED wall washer lamps with a built-in driver. Background Art

[0002] The function of LED wall washer is to make the wall look like it has been washed by water through light. It is mainly used for architectural decorative lighting, green landscape lighting, atmosphere lighting and outlining the outline of large buildings. When it rains, snows or is in a high pressure and high humidity environment, as the LED wall washer is turned off, the temperature inside the LED wall washer will drop sharply. Due to thermal expansion and contraction, the pressure inside the lamp shell will decrease rapidly, and even a partial vacuum will be generated. At this time, due to the pressure difference between the inside and outside of the lamp shell, the outside air or moisture will quietly mix into a small amount of air or moisture inside the lamp shell and condense into water vapor. Once water vapor is generated in the sealed lamp body, it is not easy to be discharged. With the accumulation of water vapor, electronic components will be corroded in the lamp body and the light intensity will be affected, which will seriously affect the life and power of the LED wall washer, resulting in the phenomenon of "weak light". Therefore, it is very necessary to reduce the "weak light" phenomenon of LED wall washer, increase the life of LED wall washer and reduce energy loss.

[0003] The existing technology for reducing the "weak light" phenomenon of LED wall washers mainly focuses on the design of hardware devices. It is to increase the sealing and balanced pressure difference of LED wall washers from the hardware level, so as to reduce the "weak light" phenomenon of LED wall washers. However, this method has a high sealing performance for LED wall washers. With the passage of time and the corrosion of complex environments, the sealing performance of LED wall washers will gradually deteriorate, shortening the life of LED wall washers. Secondly, the balanced pressure difference method of LED wall washers at the hardware level makes it difficult to maintain precise control of the internal pressure difference balance of LED wall washers. Unstable internal pressure difference can easily lead to loss of the lamp shell of LED wall washers. Long-term complex environments will also shorten the life of LED wall washers and produce the "weak light" phenomenon. Summary of the invention

[0004] In order to solve the above technical problems, an energy-saving control system of an LED wall washer lamp with a built-in drive is provided. This technical solution solves the problems proposed in the above background technology. However, this method has a high sealing performance for the LED wall washer lamp. With the passage of time and the corrosion of the complex environment, the sealing performance of the LED wall washer lamp will gradually deteriorate, shortening the life of the LED wall washer lamp. Secondly, the balanced pressure difference method of the LED wall washer lamp at the hardware level makes it difficult to maintain precise control of the internal pressure difference balance of the LED wall washer lamp. The unstable internal pressure difference can easily lead to the loss of the lamp shell of the LED wall washer lamp. The long-term complex environment will also shorten the life of the LED wall washer lamp and cause the "weak light" phenomenon.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] An energy-saving control system for an LED wall washer lamp with a built-in driver, comprising:

[0007] An environmental monitoring device, the environmental monitoring device is installed outside the LED wall washer, and the environmental monitoring device is used to monitor the data of the external environment of the LED wall washer;

[0008] A data acquisition module, the data acquisition module is used to receive and record monitoring data of the environmental monitoring device outside the LED wall washer;

[0009] A data processing module, wherein the data processing module is used to smooth and denoise the received external environment data of the LED wall washer lamp based on a first-order Kalman filter algorithm;

[0010] A "weak light" detection module, which is used to judge and predict whether a "weak light" problem will occur inside the LED wall washer based on a machine learning algorithm model;

[0011] An internal voltage stabilizing device is used to balance the internal voltage difference of the LED wall washer lamp to avoid the problem of "weak light".

[0012] Preferably, the receiving and recording of monitoring data of the environmental monitoring device outside the LED wall washer specifically includes:

[0013] According to the monitoring data and communication protocol type of the monitoring equipment, set the corresponding data packet format;

[0014] A built-in driver chip is provided to receive and record monitoring data of the environmental monitoring device outside the LED wall washer;

[0015] The received monitoring data of the external LED wall washer lamp is displayed in real time through the data display device.

[0016] Preferably, the data processing module specifically includes:

[0017] A noise variance determination unit, the noise variance determination unit being used to determine a process noise covariance Q and a measurement noise covariance R of the monitoring data according to a product description of the environmental monitoring device;

[0018] A data updating unit, the data updating unit is used to predict the environmental monitoring data at the t-th moment according to the environmental monitoring data at the t-1th moment, calculate the estimated deviation value and the Kalman gain value at the t-1th moment; calculate the updated value of the estimated deviation at the tth moment according to the estimated deviation value and the Kalman gain value; obtain the state estimation value at the tth moment according to the Kalman gain value and the observed value at the tth moment;

[0019] A model building unit, wherein the model building unit is used to update and process the observation data at the next moment according to the updated value of the predicted deviation at the tth moment and the state estimation value at the tth moment, and complete the establishment of the first-order Kalman filter algorithm model;

[0020] A data filtering unit is used to smooth and denoise the monitoring data of the environmental monitoring device received in real time based on the updated first-order Kalman algorithm model.

[0021] Preferably, the expression for calculating the estimated deviation value and Kalman gain at time t-1 is:

[0022]

[0023] Where P(t|t-1) is the expected deviation value of the monitoring data at the t-1th moment, P(t-1) is the deviation value of the monitoring data at the t-1th moment, Q is the process noise covariance of the monitoring data, K is the value of the Kalman gain at the t-1th moment, and R is the measurement noise covariance;

[0024] The updated value expression of the estimated deviation at the tth moment is:

[0025] P(t)=(1-K)P(t|t-1)

[0026] Where P(t) is the updated value of the estimated deviation of the monitoring data at time t;

[0027] The expression for obtaining the estimated state value at time t is:

[0028] x(t)=x(t-1)+K[yx(t-1)]

[0029] Where x(t) is the estimated value of the monitoring data state at the tth moment, x(t-1) is the state value of the monitoring data at the t-1th moment, and y is the observed value of the monitoring data at the tth moment.

[0030] Preferably, the “weak light” detection module specifically includes:

[0031] A data standardization unit, which is used to standardize the smoothed and denoised data and digitally replace the "light weakness" level based on big data analysis;

[0032] A data acquisition unit, the data acquisition unit is used to carry out a "weak light" simulation experiment, and acquire relevant experimental data and historical data of environmental factors and "weak light" levels based on big data;

[0033] A sample establishment unit, the sample establishment unit is used to establish a training set and a test set for judging the "light weakness" level according to the obtained experimental data and historical data of the environmental factors and the "light weakness" level;

[0034] A linear relationship unit, wherein the linear relationship unit is used to establish a linear regression expression for determining the level of "light weakness" based on the influencing factors and "light weakness";

[0035] A "weak light" judgment unit, wherein the "weak light" judgment unit is used to establish a machine learning model for judging the "weak light" level based on the least squares method, and to judge and predict whether a "weak light" problem will occur inside the LED wall washer.

[0036] Preferably, the linear regression expression for determining the "light weakness" level is:

[0037] y(x)=θ0+θ1x1+θ2x2+θ3x3

[0038] Wherein, y(x) is the calculated value of the "weak light" level generated inside the LED wall washer, θ0 is the equilibrium constant term, x1 is the temperature factor affecting the "weak light" level, x2 is the humidity factor affecting the "weak light" level, x3 is the air pressure factor affecting the "weak light" level, θ1 is the parameter affecting the temperature factor of the "weak light" level, θ2 is the parameter affecting the humidity factor of the "weak light" level, and θ3 is the parameter affecting the air pressure factor of the "weak light" level;

[0039] The least squares expression is:

[0040]

[0041] Where, Y is the sum of the squares of the errors between the calculated value and the actual value of the “weak light” level generated by the LED wall washer, m is the number of sample groups in the training set for judging the “weak light” level, y(x i ) is the calculated value of the “light weakness” level of the i-th group of training data, y i is the actual value of the “light weakness” level of the i-th group of training set data.

[0042] Preferably, the internal voltage stabilizing device is used to balance the internal pressure difference of the LED wall washer lamp to avoid the problem of "weak light" and specifically includes:

[0043] Based on big data analysis, the energy loss of LED wall washers due to the "weak light" level is obtained, and the energy saving level interval [0-a] is set;

[0044] Based on the judgment and prediction results of the "weak light" detection, determine whether the "weak light" detection exceeds the energy-saving level range. If so, it means that the current "weak light" level requires more energy to maintain the expected brightness and needs to be adjusted. If not, it means that the LED wall washer is operating normally;

[0045] According to the pulse signal generated by the judgment and prediction results of "weak light" detection, the internal voltage stabilization equipment is controlled to balance the internal pressure difference of the LED wall washer lamp;

[0046] The internal pressure stabilizing device is a controllable gas volume device that releases and absorbs dry gas, and balances the internal pressure difference of the LED wall washer lamp by taking in and out air inside the LED wall washer lamp.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] Compared with the existing technical method of reducing the "weak light" phenomenon by increasing the sealing and balancing the pressure difference of LED wall washers from the hardware level, this solution monitors the external environment of LED wall washers by setting up environmental monitoring equipment, and uses the first-order Kalman filter algorithm model to smooth and denoise the monitoring data to ensure the stability and reliability of the monitoring data, thereby ensuring that the subsequent machine learning algorithm model can accurately judge and predict whether there will be a "weak light" problem inside the LED wall washers. Finally, through the pulse signal generated by the judgment result of the machine learning algorithm model, the internal voltage stabilizing device is controlled to balance the internal pressure difference of the LED wall washers, thereby realizing precise automatic control and adjustment of the internal pressure difference of the LED wall washers, reducing the occurrence of the "weak light" phenomenon and increasing the service life of the LED wall washers. At the same time, the adaptability of the LED wall washers to complex environments is enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flow chart of an energy-saving control system for an LED wall washer lamp with a built-in driver according to the present invention;

[0050] Figure 2 This is a flow chart of the present invention, which is based on the first-order Kalman filter algorithm, for smoothing and denoising the received external environment data of the LED wall washer lamp;

[0051] Figure 3 This is a flow chart of the present invention for judging and predicting whether a "weak light" problem will occur inside an LED wall washer based on a machine learning algorithm model. DETAILED DESCRIPTION

[0052] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0053] Reference Figure 1 As shown, an energy-saving control system of an LED wall washer lamp with a built-in driver includes:

[0054] An environmental monitoring device, the environmental monitoring device is installed outside the LED wall washer, and the environmental monitoring device is used to monitor the data of the external environment of the LED wall washer;

[0055] A data acquisition module, the data acquisition module is used to receive and record monitoring data of the environmental monitoring device outside the LED wall washer;

[0056] A data processing module, wherein the data processing module is used to smooth and denoise the received external environment data of the LED wall washer lamp based on a first-order Kalman filter algorithm;

[0057] A "weak light" detection module, which is used to judge and predict whether a "weak light" problem will occur inside the LED wall washer based on a machine learning algorithm model;

[0058] An internal voltage stabilizing device is used to balance the internal voltage difference of the LED wall washer lamp to avoid the problem of "weak light".

[0059] Reference Figure 1 As shown, the data processing module specifically includes:

[0060] A noise variance determination unit, the noise variance determination unit being used to determine a process noise covariance Q and a measurement noise covariance R of the monitoring data according to a product description of the environmental monitoring device;

[0061] A data updating unit, the data updating unit is used to predict the environmental monitoring data at the t-th moment according to the environmental monitoring data at the t-1th moment, calculate the estimated deviation value and the Kalman gain value at the t-1th moment; calculate the updated value of the estimated deviation at the tth moment according to the estimated deviation value and the Kalman gain value; obtain the state estimation value at the tth moment according to the Kalman gain value and the observed value at the tth moment;

[0062] A model building unit, wherein the model building unit is used to update and process the observation data at the next moment according to the updated value of the predicted deviation at the tth moment and the state estimation value at the tth moment, and complete the establishment of the first-order Kalman filter algorithm model;

[0063] A data filtering unit is used to smooth and denoise the monitoring data of the environmental monitoring device received in real time based on the updated first-order Kalman algorithm model.

[0064] Reference Figure 1 As shown, the “weak light” detection module specifically includes:

[0065] A data standardization unit, which is used to standardize the smoothed and denoised data and digitally replace the "light weakness" level based on big data analysis;

[0066] A data acquisition unit, the data acquisition unit is used to carry out a "weak light" simulation experiment, and acquire relevant experimental data and historical data of environmental factors and "weak light" levels based on big data;

[0067] A sample establishment unit, the sample establishment unit is used to establish a training set and a test set for judging the "light weakness" level according to the obtained experimental data and historical data of the environmental factors and the "light weakness" level;

[0068] A linear relationship unit, wherein the linear relationship unit is used to establish a linear regression expression for determining the level of "light weakness" based on the influencing factors and "light weakness";

[0069] A "weak light" judgment unit, wherein the "weak light" judgment unit is used to establish a machine learning model for judging the "weak light" level based on the least squares method, and to judge and predict whether a "weak light" problem will occur inside the LED wall washer.

[0070] It can be explained that this solution monitors the environment of the LED wall washer lamp housing by setting up environmental monitoring equipment, and uses the first-order Kalman filter algorithm model to smooth and denoise the monitoring data to ensure the stability and reliability of the monitoring data, thereby ensuring that the subsequent machine learning algorithm model can accurately judge and predict whether there will be a "weak light" problem inside the LED wall washer lamp. Finally, through the pulse signal generated by the judgment result of the machine learning algorithm model, the internal voltage stabilizing device is controlled to balance the internal pressure difference of the LED wall washer lamp, thereby realizing precise automatic control and adjustment of the internal pressure difference of the LED wall washer lamp through the software program compiled by the intelligent algorithm.

[0071] Reference Figure 2 As shown, the smoothing and denoising of the received external environment data of the LED wall washer lamp based on the first-order Kalman filter algorithm specifically includes:

[0072] According to the product description of the environmental monitoring equipment, determine the process noise covariance Q and measurement noise covariance R of the monitoring data;

[0073] Predict the environmental monitoring data at time t based on the environmental monitoring data at time t-1, and calculate the estimated deviation value and Kalman gain value at time t-1;

[0074] Calculate the updated value of the estimated deviation at time t according to the estimated deviation value and the value of the Kalman gain;

[0075] According to the value of Kalman gain, combined with the observation value at time t, the state estimation value at time t is obtained;

[0076] According to the estimated deviation update value at the tth moment and the state estimation value at the tth moment, the observation data at the next moment is updated and processed according to the following steps to complete the establishment of the first-order Kalman filter algorithm model;

[0077] Based on the updated first-order Kalman algorithm model, the monitoring data of the environmental monitoring equipment received in real time is smoothed and denoised;

[0078] The expression for calculating the estimated deviation value and Kalman gain at time t-1 is:

[0079]

[0080] Where P(t|t-1) is the expected deviation value of the monitoring data at the t-1th moment, P(t-1) is the deviation value of the monitoring data at the t-1th moment, Q is the process noise covariance of the monitoring data, K is the value of the Kalman gain at the t-1th moment, and R is the measurement noise covariance;

[0081] The updated value expression of the estimated deviation at the tth moment is:

[0082] P(t)=(1-K)P(t|t-1)

[0083] Where P(t) is the updated value of the estimated deviation of the monitoring data at time t;

[0084] The expression for obtaining the estimated state value at time t is:

[0085] x(t)=x(t-1)+K[yx(t-1)]

[0086] Where x(t) is the estimated value of the monitoring data state at the tth moment, x(t-1) is the state value of the monitoring data at the t-1th moment, and y is the observed value of the monitoring data at the tth moment.

[0087] It can be explained that in the face of different environments, especially complex outdoor environments, due to the vibration or electromagnetic signal interference of the LED wall washer, abnormal excitation signals are generated, and the peak values ​​of these abnormal excitation signals will have an important impact on the subsequent judgment and prediction of whether there will be a "weak light" problem inside the LED wall washer based on the machine learning algorithm. Therefore, it is necessary to add a real-time filtering algorithm to smooth and reduce the noise of the received environmental monitoring data through the filtering algorithm model to ensure the stability and reliability of the data. The first-order Kalman filter algorithm is a calculation method that predicts the next state through the current state and observation values. It is an algorithm for optimal estimation of the system state and has good real-time and smoothing and noise reduction capabilities. For the smoothing and noise reduction of monitoring data, this scheme uses the first-order Kalman filter algorithm, combined with the actual situation of the monitoring data, to simplify the basic formula of the Kalman filter algorithm and the value of the related parameter matrix, thereby solving the complexity of the algorithm.

[0088] Reference Figure 3 As shown, the method of judging and predicting whether a "weak light" problem will occur inside an LED wall washer based on a machine learning algorithm model specifically includes:

[0089] The smoothed and denoised data is standardized, and the “light weakness” level is replaced digitally based on big data analysis;

[0090] Conduct "weak light" simulation experiments and obtain relevant experimental data and historical data on environmental factors and "weak light" levels based on big data;

[0091] Based on the experimental data and historical data of environmental factors and “light weakness” levels, a training set and a test set for judging “light weakness” levels are established;

[0092] Based on the influencing factors and "light weakness", a linear regression expression for judging the "light weakness" level is established;

[0093] Based on the least squares method, a machine learning model for judging the "weak light" level is established to judge and predict whether there will be a "weak light" problem inside the LED wall washer;

[0094] The linear regression expression for judging the “weak light” level is:

[0095] y(x)=θ0+θ1x1+θ2x2+θ3x3

[0096] Wherein, y(x) is the calculated value of the "weak light" level generated inside the LED wall washer, θ0 is the equilibrium constant term, x1 is the temperature factor affecting the "weak light" level, x2 is the humidity factor affecting the "weak light" level, x3 is the air pressure factor affecting the "weak light" level, θ1 is the parameter affecting the temperature factor of the "weak light" level, θ2 is the parameter affecting the humidity factor of the "weak light" level, and θ3 is the parameter affecting the air pressure factor of the "weak light" level;

[0097] The least squares expression is:

[0098]

[0099] Where, Y is the sum of the squares of the errors between the calculated value and the actual value of the “weak light” level generated by the LED wall washer, m is the number of sample groups in the training set for judging the “weak light” level, y(x i ) is the calculated value of the “light weakness” level of the i-th group of training data, y i is the actual value of the “light weakness” level of the i-th group of training set data.

[0100] It can be explained that the "weak light" phenomenon of LED wall washer lamps is mainly due to the fact that when it rains, snows, or is in a high-pressure, high-humidity environment, as the LED wall washer lamps are turned off, the temperature inside the LED wall washer lamps will drop sharply. Due to thermal expansion and contraction, the pressure inside the lamp housing will decrease rapidly, and even a partial vacuum will be generated. At this time, due to the pressure difference between the inside and outside of the lamp housing, a small amount of air or moisture from the outside will quietly mix into the lamp housing and condense into water vapor. Once water vapor is generated in the sealed lamp body, it is not easy to be discharged. As water vapor accumulates, it will corrode electronic components in the lamp body and affect the light intensity, seriously affecting the life and power usage of the LED wall washer lamp, thus causing the "weak light" phenomenon. Therefore, when establishing a machine learning algorithm model, the main influencing factors that need to be considered are temperature, humidity and pressure. Through big data analysis and the test results of the "weak light" simulation experiment, the relationship between temperature, humidity and pressure and the "weak light" level is established, so as to use the machine learning algorithm model to judge and predict whether the "weak light" problem will occur inside the LED wall washer lamp. Secondly, since the adjustment of the internal pressure difference of the LED wall washer lamp requires good real-time performance, it is ensured that when the problem pressure difference occurs, the internal voltage stabilizing device can be driven in time to balance the internal pressure difference of the LED wall washer lamp. Therefore, by combining the trained machine learning algorithm model with the first-order Kalman filter algorithm model, the internal pressure difference of the LED wall washer lamp can be effectively adjusted to ensure the timeliness of handling the "weak light" problem.

[0101] To sum up, the advantages of the present invention are: compared with the existing technical method of reducing the "weak light" phenomenon by increasing the sealing and balancing the pressure difference of the LED wall washer lamp from the hardware level, this solution monitors the outer shell environment of the LED wall washer lamp by setting up environmental monitoring equipment, and uses the first-order Kalman filter algorithm model to smooth and denoise the monitoring data to ensure the stability and reliability of the monitoring data, thereby ensuring that the subsequent machine learning algorithm model can accurately judge and predict whether the LED wall washer lamp will have a "weak light" problem. Finally, through the pulse signal generated by the judgment result of the machine learning algorithm model, the internal voltage stabilizing device is controlled to balance the internal pressure difference of the LED wall washer lamp, thereby realizing precise automatic control and adjustment of the internal pressure difference of the LED wall washer lamp, reducing the occurrence of the "weak light" phenomenon and increasing the service life of the LED wall washer lamp. At the same time, the adaptability of the LED wall washer lamp to complex environments is enhanced.

[0102] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. An energy-saving control system for an LED wall washer with a built-in driver, characterized in that: include: An environmental monitoring device, the environmental monitoring device is installed outside the LED wall washer, and the environmental monitoring device is used to monitor the data of the external environment of the LED wall washer; A data acquisition module, the data acquisition module is used to receive and record monitoring data of the environmental monitoring device outside the LED wall washer; A data processing module, wherein the data processing module is used to smooth and denoise the received external environment data of the LED wall washer lamp based on a first-order Kalman filter algorithm; A "weak light" detection module, which is used to judge and predict whether a "weak light" problem will occur inside the LED wall washer based on a machine learning algorithm model; An internal voltage stabilizing device is used to balance the internal voltage difference of the LED wall washer lamp to avoid the problem of "weak light".

2. The energy-saving control system of an LED wall washer with a built-in driver according to claim 1, characterized in that: The receiving and recording of monitoring data of the environmental monitoring device outside the LED wall washer specifically includes: According to the monitoring data and communication protocol type of the monitoring equipment, set the corresponding data packet format; A built-in driver chip is provided to receive and record monitoring data of the environmental monitoring device outside the LED wall washer; The received monitoring data of the external LED wall washer lamp is displayed in real time through the data display device.

3. The energy-saving control system of an LED wall washer with a built-in driver according to claim 2, characterized in that: The data processing module specifically includes: A noise variance determination unit, the noise variance determination unit being used to determine a process noise covariance Q and a measurement noise covariance R of the monitoring data according to a product description of the environmental monitoring device; A data updating unit, the data updating unit is used to predict the environmental monitoring data at the t-th moment according to the environmental monitoring data at the t-1th moment, calculate the estimated deviation value and the Kalman gain value at the t-1th moment; calculate the updated value of the estimated deviation at the tth moment according to the estimated deviation value and the Kalman gain value; obtain the state estimation value at the tth moment according to the Kalman gain value and the observed value at the tth moment; A model building unit, wherein the model building unit is used to update and process the observation data at the next moment according to the updated value of the predicted deviation at the tth moment and the state estimation value at the tth moment, and complete the establishment of the first-order Kalman filter algorithm model; A data filtering unit is used to smooth and denoise the monitoring data of the environmental monitoring device received in real time based on the updated first-order Kalman algorithm model.

4. The energy-saving control system of the LED wall washer with built-in driver according to claim 3, characterized in that: The expression for calculating the estimated deviation value and Kalman gain at time t-1 is: Where P(t|t-1) is the expected deviation value of the monitoring data at the t-1th moment, P(t-1) is the deviation value of the monitoring data at the t-1th moment, Q is the process noise covariance of the monitoring data, K is the value of the Kalman gain at the t-1th moment, and R is the measurement noise covariance; The updated value expression of the estimated deviation at the tth moment is: P(t)=(1-K)P(t|t-1) Where P(t) is the updated value of the estimated deviation of the monitoring data at time t; The expression for obtaining the estimated state value at time t is: x(t)=x(t-1)+K[yx(t-1)] Where x(t) is the estimated value of the monitoring data state at the tth moment, x(t-1) is the state value of the monitoring data at the t-1th moment, and y is the observed value of the monitoring data at the tth moment.

5. The energy-saving control system of an LED wall washer with a built-in driver according to claim 4, characterized in that: The "weak light" detection module specifically includes: A data standardization unit, which is used to standardize the smoothed and denoised data and digitally replace the "light weakness" level based on big data analysis; A data acquisition unit, which is used to carry out a "weak light" simulation experiment and acquire relevant experimental data and historical data on environmental factors and "weak light" levels based on big data; A sample establishment unit, the sample establishment unit is used to establish a training set and a test set for judging the "weak light" level according to the obtained experimental data and historical data of the environmental factors and the "weak light" level; A linear relationship unit, wherein the linear relationship unit is used to establish a linear regression expression for determining the level of "light weakness" based on the influencing factors and "light weakness"; A "weak light" judgment unit, the "weak light" judgment unit is used to establish a machine learning model for judging the "weak light" level based on the least squares method, to judge and predict whether a "weak light" problem will occur inside the LED wall washer.

6. The energy-saving control system of an LED wall washer with a built-in driver according to claim 5, characterized in that: The linear regression expression for judging the "weak light" level is: y(x)=θ0+θ1x1+θ2x2+θ3x3 In the formula, y(x) is the calculated value of the "weak light" level generated inside the LED wall washer, θ0 is the equilibrium constant term, x1 is the temperature factor affecting the "weak light" level, x2 is the humidity factor affecting the "weak light" level, x3 is the air pressure factor affecting the "weak light" level, θ1 is the parameter affecting the temperature factor of the "weak light" level, θ2 is the parameter affecting the humidity factor of the "weak light" level, and θ3 is the parameter affecting the air pressure factor of the "weak light" level; The least squares expression is: Where, Y is the sum of the squares of the errors between the calculated value and the actual value of the "weak light" level generated by the LED wall washer, m is the number of sample groups in the training set for judging the "weak light" level, y(x i ) is the calculated value of the "light weakness" level of the i-th group of training data, y i is the actual value of the "light weakness" level of the i-th group of training set data.

7. The energy-saving control system of an LED wall washer with a built-in driver according to claim 6, characterized in that: The internal voltage stabilizing device is used to balance the internal voltage difference of the LED wall washer to avoid the problem of "weak light". Specifically, it includes: Based on big data analysis, the energy loss of LED wall washers due to "weak light" level is obtained, and the energy saving level interval [0-a] is set; Based on the judgment and prediction results of "weak light" detection, determine whether the "weak light" detection exceeds the energy-saving level range. If so, it means that the current "weak light" level requires more energy to maintain the expected brightness and needs to be adjusted. If not, it means that the LED wall washer is operating normally; According to the pulse signal generated by the judgment and prediction results of "weak light" detection, the internal voltage stabilization equipment is controlled to balance the internal pressure difference of the LED wall washer; The internal pressure stabilizing device is a controllable gas volume device that releases and absorbs dry gas, and balances the internal pressure difference of the LED wall washer lamp by taking in and out air inside the LED wall washer lamp.

Images