Ship cabin illumination brightness automatic adjusting method based on fuzzy neural network

Abstract

The invention discloses a ship cabin illumination brightness automatic adjustment method based on a fuzzy neural network. The method comprises the steps of ship cabin brightness data acquisition and cleaning estimation; the method comprises the following steps of: cleaning brightness data acquired by a plurality of sensors distributed in an indoor environment and an outdoor environment of a ship cabin by using a self-adaptive weighted fusion algorithm so as to obtain brightness estimated value data of the ship cabin; constructing a ship cabin illumination brightness automatic adjustment model based on the fuzzy neural network; obtaining horizontal and vertical illuminance of the point light source according to the ship cabin brightness estimation value data; inputting the obtained illumination into an automatic adjustment model, and performing fuzzification processing on parameters by the model to generate fuzzy linguistic variables; and performing fuzzy reasoning according to the generated fuzzy linguistic variable and the fuzzy rule base to obtain fuzzy output, thereby realizing adjustment of the contrast. Experimental results prove that the method can complete automatic adjustment of the illumination brightness of the ship cabin and ensure the comfort of the illumination brightness in the cabin.

Description

Technical Field

[0001] The invention relates to the technical field of ship cabin lighting control, and in particular to a ship cabin lighting brightness automatic adjustment method based on fuzzy neural network. Background Art

[0002] With the rapid development of modern navigation technology, ships have become the main mode of transportation connecting countries around the world, and the comfort and safety of their internal environment have become the focus of the shipping industry. As a space where crew members and passengers stay for a long time, the lighting system of the ship cabin is not only a tool to provide basic lighting, but also one of the key factors affecting work efficiency, living experience and navigation safety.

[0003] With the rapid development of intelligent control technology, the automatic adjustment system of ship cabin lighting brightness has emerged, marking that the internal environment management of ships has entered a new intelligent era. The automatic adjustment system of ship cabin lighting brightness integrates advanced sensor technology, data analysis algorithms and energy-efficient LED lighting equipment. It can intelligently adjust the lighting brightness according to the actual lighting conditions in the cabin, time changes, personnel activities and functional requirements of specific areas, and realize personalized, comfortable and efficient lighting management.

[0004] In the existing technology, Liang Bo et al. proposed a lighting brightness adjustment method using dynamic optimization and intelligent control in "Dynamic Optimization and Intelligent Control of Highway Tunnel Lighting Considering Visibility Influence"; by analyzing the field data under different time and space environments, the visibility change law is obtained; an on-demand lighting and dynamic optimization model based on visualization, light intensity, etc. is constructed; the lighting scene is segmented and combined with the fuzzy radial basis neural network algorithm to establish an intelligent control model to achieve lighting brightness adjustment. Reasonable segmentation of lighting scenes is the key to establishing an intelligent control model. If the division is not fine or reasonable, the adjustment performance of the model in a specific scene will be reduced. Chen Z et al. proposed a lighting brightness adjustment method using support vector machines in "An intelligent lighting control system for urban street lamps using internet of things"; geographic information technology is added to the intelligent lighting system for positioning, and a set of intelligent lighting adjustment systems is constructed in combination with the support vector machine algorithm to control and monitor the lighting lamps in real time and complete the lighting brightness adjustment; the training and prediction effects of the support vector machine algorithm depend on the quality of the input data. Inaccurate data or noise will directly affect the adjustment accuracy of the lighting brightness. Summary of the invention

[0005] 1. Technical problems to be solved:

[0006] In view of the above technical problems, the present invention provides a method for automatic adjustment of ship cabin lighting brightness based on fuzzy neural network. The scheme uses an adaptive weighted algorithm to clean the collected ship cabin illumination data, determines the lighting brightness at different angles according to the inverse square law of distance and the cosine law, and adjusts variables in combination with fuzzy neural network to correct output errors, thereby completing automatic adjustment of ship cabin lighting brightness.

[0007] 2. Technical solution:

[0008] A method for automatically adjusting the brightness of ship cabin lighting based on fuzzy neural network, characterized in that it includes:

[0009] Step 1: Acquisition and cleaning of ship cabin light data; the light data collected by multiple sensors distributed in the indoor and outdoor environments of the ship cabin are cleaned by using an adaptive weighted fusion algorithm to obtain the estimated value data of the ship cabin brightness; the light data includes indoor and outdoor brightness, point light source and its light intensity data;

[0010] Step 2: construct a ship cabin lighting brightness automatic adjustment model based on fuzzy neural network; the ship cabin lighting brightness automatic adjustment model based on fuzzy neural network performs fuzzification processing on the input light parameters to generate fuzzy language variables; establishes a fuzzy rule base according to expert knowledge or experience; performs fuzzy reasoning according to the generated fuzzy language variables and fuzzy rule base to obtain fuzzy output; the fuzzy output is converted into specific parameters as the output of the model;

[0011] Step 3: According to the estimated value data of the ship cabin brightness, two light parameters, namely the illumination brightness generated by the point light source on the horizontal plane and the illumination on the vertical plane, are obtained; the two obtained parameters are used as the input of the fuzzy neural network, and the fuzzy neural network outputs the parameter adjustment result.

[0012] Furthermore, in step 1, an adaptive weighted algorithm is used to clean the acquired data, and multiple sensors distributed in the indoor and outdoor environments of the ship cabin collect light data, and the collected light data is weighted and fused. Under the condition of ensuring the minimum total variance, an adaptive method is used to find the corresponding optimal weighting factor, so that the fused measurement value is the optimal value, which is the relatively accurate ship cabin brightness estimation value after cleaning; the cleaned value is expressed as follows:

[0013]

[0014] In the above formula, X,(t) represents the measured value after cleaning, including the cleaned value of brightness data, the cleaned value of point light source data and the cleaned value of light intensity data; Q i represents the corresponding optimal weighting factor time series data; X irepresents the collected data; m represents the collection time.

[0015] Furthermore, in step 3, two light parameters, namely, the illumination brightness generated on the horizontal plane and the illumination brightness on the vertical plane, are obtained according to the estimated value data of the ship cabin brightness. The specific calculation process is as follows:

[0016] The illumination E generated by the point light source S illuminating the horizontal surface h as follows:

[0017]

[0018] Vertical illumination E of point light source S v The expression is:

[0019]

[0020] In the above formula: h is the height from the light source to the horizontal plane; I θ represents light intensity; θ represents light source irradiation.

[0021] Furthermore, in step 3, the illumination parameter E h and E v As the input of the fuzzy neural network, the input illumination parameter E h and E v Perform fuzzification processing and convert it into fuzzy language variables as follows:

[0022]

[0023] In the above formula: μ A (E h ,E v ) is the fuzzy membership function corresponding to the converted fuzzy linguistic variable; c is the center of the fuzzy set;

[0024] In the established fuzzy rule base, the illumination parameters E h and E v The minimum and maximum thresholds of h or / and E v If the value of E is lower than the preset corresponding minimum threshold, the control parameter indicating "increase brightness" is output; h or / and E v If the value is higher than the corresponding preset maximum threshold, a control parameter indicating "reducing brightness" is output;

[0025] Fuzzy reasoning is performed based on the input fuzzy language variables and fuzzy rule base to obtain fuzzy output, and the fuzzy output is converted into a specific brightness adjustment value:

[0026]

[0027] In the above formula, y is the output value after defuzzification; μ i (E h ,E v ) is the membership degree of the i-th fuzzy set; y i is the output value corresponding to the i-th fuzzy set;

[0028] According to the defuzzified output value y, the lighting brightness of the ship cabin is adjusted.

[0029] 3. Beneficial effects:

[0030] This method is a ship cabin lighting brightness automatic adjustment method based on fuzzy neural network. After obtaining the cleaned brightness data using an adaptive weighted algorithm in the data acquisition stage, the lighting brightness of the point light sources inside and outside the cabin at each angle is obtained by the cosine law. The obtained brightness parameters are input into the fuzzy neural network for fuzzification processing and converted into fuzzy language variables. A fuzzy rule base is established based on expert knowledge or experience, and fuzzy reasoning is performed based on the input fuzzy language variables and the fuzzy rule base to obtain fuzzy output. The fuzzy output is converted into a specific brightness adjustment value, and the lighting brightness of the ship cabin is adjusted by the defuzzified output value. Experimental results show that this method can complete the automatic adjustment of the ship cabin lighting brightness and ensure the comfort of the cabin lighting brightness. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is an overall flow chart of a method for automatically adjusting the brightness of ship cabin lighting based on a fuzzy neural network of the present invention;

[0032] Figure 2 The figure is a comparison result of the power consumption of the prior art used in the comparative example and the present method;

[0033] Figure 3 A comparison chart of the illumination and color rendering index of the prior art used in the comparative example and the present method;

[0034] Figure 4 It is a comparison chart of the color rendering index of different cabins used in the prior art and the present method in the comparative example;

[0035] Figure 5 This is the prior art used in the comparative example and the working area lighting brightness adjustment state of this method. DETAILED DESCRIPTION

[0036] The present invention will be described in detail below with reference to the accompanying drawings.

[0037] As attached Figure 1 As shown: A method for automatically adjusting the brightness of ship cabin lighting based on fuzzy neural network, characterized in that it includes:

[0038] Step 1: Acquisition and cleaning of ship cabin light data; the light data collected by multiple sensors distributed in the indoor and outdoor environments of the ship cabin are cleaned by using an adaptive weighted fusion algorithm to obtain the estimated value data of the ship cabin brightness; the light data includes indoor and outdoor brightness, point light source and its light intensity data;

[0039] Step 2: construct a ship cabin lighting brightness automatic adjustment model based on fuzzy neural network; the ship cabin lighting brightness automatic adjustment model based on fuzzy neural network performs fuzzification processing on the input light parameters to generate fuzzy language variables; establishes a fuzzy rule base according to expert knowledge or experience; performs fuzzy reasoning according to the generated fuzzy language variables and fuzzy rule base to obtain fuzzy output; the fuzzy output is converted into specific parameters as the output of the model;

[0040] Step 3: According to the estimated value data of the ship cabin brightness, two light parameters, namely the illumination generated by the horizontal plane and the illumination generated by the point light source in the vertical plane, are obtained; the two obtained parameters are used as the input of the fuzzy neural network, and the fuzzy neural network outputs the parameter adjustment result.

[0041] Furthermore, in step 1, an adaptive weighted algorithm is used to clean the acquired data, and multiple sensors distributed in the indoor and outdoor environments of the ship cabin collect light data, and the collected light data is weighted and fused. Under the condition of ensuring the minimum total variance, an adaptive method is used to find the corresponding optimal weighting factor, so that the fused measurement value is the optimal value, which is the relatively accurate ship cabin brightness estimation value after cleaning; the cleaned value is expressed as follows:

[0042]

[0043] In the above formula, X,(t) represents the measured value after cleaning, including the cleaned value of brightness data, the cleaned value of point light source data and the cleaned value of light intensity data; Q i represents the corresponding optimal weighting factor time series data; X i represents the collected data; m represents the collection time.

[0044] Furthermore, in step 3, two light parameters, namely, the illumination generated on the horizontal plane and the illumination on the vertical plane, are obtained according to the estimated value data of the ship cabin brightness. The specific calculation process is as follows:

[0045] The illumination E generated by the point light source S illuminating the horizontal surface h as follows:

[0046]

[0047] Vertical illumination E of point light source S v The expression is:

[0048]

[0049] In the above formula: h is the height from the light source to the horizontal plane; I θ represents the light intensity; θ represents the angle at which the light source shines on the horizontal plane; and R is the distance from the light source to the illuminated point.

[0050] Furthermore, in step 3, the illumination parameter E h and E v As the input of the fuzzy neural network, the input illumination parameter E h and E v Perform fuzzification processing and convert it into fuzzy language variables as follows:

[0051]

[0052] In the above formula: μ A (E h ,E v ) is the fuzzy membership function corresponding to the converted fuzzy linguistic variable; c is the center of the fuzzy set;

[0053] In the established fuzzy rule base, the illumination parameters E h and E v The minimum and maximum thresholds of h or / and E v If the value of E is lower than the preset corresponding minimum threshold, the control parameter indicating "increase brightness" is output; h or / and E v If the value is higher than the corresponding preset maximum threshold, a control parameter indicating "reducing brightness" is output;

[0054] Fuzzy reasoning is performed based on the input fuzzy language variables and fuzzy rule base to obtain fuzzy output, and the fuzzy output is converted into a specific brightness adjustment value:

[0055]

[0056] In the above formula, y is the output value after defuzzification; μ i (E h ,E v ) is the membership degree of the i-th fuzzy set; y i is the output value corresponding to the i-th fuzzy set;

[0057] According to the defuzzified output value y, the illumination of the ship cabin is adjusted.

[0058] Comparative Example:

[0059] In order to verify the automatic adjustment performance of this method, this comparative example selects the genetic simulated annealing algorithm and the improved sparrow search algorithm to conduct comparative experiments with this method. The experimental locations are 10 different areas such as the cockpit, engine room, crew dormitory, passenger cabin, storage room, and rest area, among which No. 1-4 is the working area and No. 5-10 is the living and resting area. Table 1 shows the relevant parameters of the ship in the experiment.

[0060] Table 1 Ship related experimental parameters

[0061]

[0062]

[0063] from Figure 2 It can be seen that as time goes by, although the power consumption of this method has increased, it has always remained at a lower level compared to the other two existing methods. This shows that this method can effectively control energy consumption and avoid unnecessary energy waste when adjusting the brightness of ship cabin lighting. Compared with this method, the power consumption of the genetic simulated annealing method and the improved sparrow search method is significantly higher and fluctuates greatly. It proves that when adjusting the lighting brightness, both require more energy support, which leads to increased energy consumption. In summary, this method can realize the automatic adjustment of the ship cabin lighting brightness, which not only guarantees the lighting needs in the cabin, but also avoids energy waste caused by excessive lighting.

[0064] In order to further demonstrate the lighting effect of this method, the illumination and color rendering are used as evaluation indicators to ensure the safety and comfort of each area, such as Figure 3 , 4 Shown is a comparison of illuminance and color rendering index for the three methods.

[0065] from Figure 3 It can be seen that in the comparative example, the illumination value of this method is maintained in the range of 400-500lx in cabins 1-4, which can effectively provide a stable and suitable lighting environment for the staff and enhance the safety of the working area; cabins 5-10 are below 300lx, which ensures the basic lighting brightness and reduces resource waste. The regional illumination effects of the genetic simulated annealing method and the improved sparrow search method are poor, and the values ​​of both the working area and the living area are between 100-350lx. Although the living and resting areas have better lighting effects, the illumination of the working area is poor, which will cause visual fatigue of the driver and increase uncertainty and risk during driving. In short, this method has better performance in automatically adjusting the lighting brightness of the ship's cabins, can meet the requirements of each area, and ensure the safety of the working area and the comfort of the living and resting areas.

[0066] pass Figure 4 It can be seen that the color rendering index of this method is greater than 90, with high color rendering, and can highly restore the color of objects in the cabin, making the colors look bright and natural, helping the driver to make more accurate judgments when performing tasks that require high color recognition. Compared with the other two methods, the color rendering index of the genetic simulated annealing method is relatively low, and its performance in color restoration is average. Although it will not cause serious color distortion, it will affect the accurate judgment of the color of objects in work scenes that require high color recognition. Compared with the genetic simulated annealing method, the improved sparrow search method has a higher color rendering index in cabins 3, 8, and 9, but lower in other cabins. In summary, this method can realize automatic adjustment of the lighting brightness of ship cabins, and the color restoration of objects in the cabin is high, which helps to improve the overall visual comfort and work efficiency of the cabin.

[0067] The operation of some equipment in the cabin working area is related to the safety of the entire ship. The lighting brightness is a key issue. For this reason, the equipment operation room in the working area is selected and verified by three methods, such as Figure 5 shown.

[0068] pass Figure 5 It can be seen that the entire control area of ​​this method (part b in the figure) is evenly and moderately illuminated, and the details of the equipment are clearly visible. It is proved that this method can improve the operator's work efficiency, reduce the risk of misoperation due to unclear vision, and ensure the safety of ship driving. Although the genetic simulated annealing method and the improved sparrow search method improve the lighting conditions, the brightness distribution is not uniform enough, or there is too bright or too dark in some areas, which affects the operator's vision and judgment, and there is a risk of misoperation. In summary, this method can better adjust the brightness of the ship's cabin lighting, avoid misoperation, and ensure the safety of ship driving.

[0069] Although the present invention has been disclosed as above in terms of preferred embodiments, they are not intended to limit the present invention. Anyone skilled in the art can make various changes or modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection defined by the claims of this application.

Claims

1. A method for automatically adjusting the brightness of ship cabin lighting based on fuzzy neural network, characterized in that: include: Step 1: Acquisition and cleaning estimation of ship cabin light data; The light data collected by multiple sensors distributed in the indoor and outdoor environment of the ship cabin are cleaned by using an adaptive weighted fusion algorithm to obtain the estimated value data of the ship cabin brightness; the light data includes indoor and outdoor brightness, point light source and its light intensity data; Step 2: construct a ship cabin lighting brightness automatic adjustment model based on fuzzy neural network; the ship cabin lighting brightness automatic adjustment model based on fuzzy neural network performs fuzzification processing on the input light parameters to generate fuzzy language variables; establishes a fuzzy rule base according to expert knowledge or experience; performs fuzzy reasoning according to the generated fuzzy language variables and fuzzy rule base to obtain fuzzy output; the fuzzy output is converted into specific parameters as the output of the model; Step 3: According to the estimated value data of the ship cabin brightness, two light parameters, namely the illumination generated by the horizontal plane and the illumination generated by the point light source in the vertical plane, are obtained; the two obtained parameters are used as the input of the fuzzy neural network, and the fuzzy neural network outputs the parameter adjustment result.

2. The method for automatically adjusting the lighting brightness of a ship cabin based on a fuzzy neural network according to claim 1 is characterized in that: In step 1, an adaptive weighted algorithm is used to clean the acquired data. Multiple sensors distributed in the indoor and outdoor environments of the ship's cabin collect light data, and the collected light data is weighted and fused. Under the condition of ensuring the minimum total variance, an adaptive method is used to find the corresponding optimal weighting factor, so that the fused measurement value is the optimal value. The optimal value is the relatively accurate ship cabin brightness estimation value after cleaning; the cleaned value is expressed as follows: In the above formula, X,(t) represents the measured value after cleaning, including the cleaned value of brightness data, the cleaned value of point light source data and the cleaned value of light intensity data; Q i represents the corresponding optimal weighting factor time series data; X i represents the collected data; m represents the collection time.

3. The method for automatically adjusting the lighting brightness of a ship cabin based on a fuzzy neural network according to claim 1 is characterized in that: In step 3, two light parameters, the illumination generated on the horizontal plane and the illumination on the vertical plane, are obtained based on the estimated value data of the ship cabin brightness. The specific calculation process is as follows: The illumination E generated by the point light source S illuminating the horizontal surface h as follows: Vertical illumination E of point light source S v The expression is: In the above formula: h is the height from the light source to the horizontal plane; I θ represents the light intensity; θ represents the angle at which the light source shines on the horizontal plane; and R is the distance from the light source to the illuminated point.

4. The method for automatically adjusting the lighting brightness of a ship cabin based on a fuzzy neural network according to claim 3 is characterized in that: In step 3, the illumination parameter E h and E v As the input of the fuzzy neural network, the input illumination parameter E h and E v Perform fuzzification processing and convert it into fuzzy language variables as follows: In the above formula: μ A (E h ,E v ) is the fuzzy membership function corresponding to the converted fuzzy linguistic variable; c is the center of the fuzzy set; the illumination parameters E are set in the established fuzzy rule base. h and E v The minimum and maximum thresholds of h or / and E v If the value of E is lower than the preset corresponding minimum threshold, the control parameter indicating "increase brightness" is output; h or / and E v If the value is higher than the corresponding preset maximum threshold, the control parameter indicating "reduce brightness" is output; Fuzzy reasoning is performed based on the input fuzzy language variables and fuzzy rule base to obtain fuzzy output, and the fuzzy output is converted into a specific brightness adjustment value: In the above formula, y is the output value after defuzzification; μ i (E h ,E v ) is the membership degree of the i-th fuzzy set; y i is the output value corresponding to the i-th fuzzy set; According to the defuzzified output value y, the lighting brightness of the ship cabin is adjusted.

Images