Multi-sensor fusion intelligent tunnel dimming system adopting improved DS algorithm

Abstract

The invention discloses a multi-sensor fusion intelligent tunnel dimming system adopting an improved DS algorithm, which belongs to the technical field of tunnel lighting control, and comprises an information fusion module for approximately converting a reliability function into a probability function by adopting an inverse DP probability conversion method and inverse conversion, carrying out data fusion through a D-S evidence theory synthesis rule, and outputting a data fusion result; tunnel environment information is obtained; a brightness calculation module; and an illumination adjusting module. According to the light intensity change inside and outside the tunnel, information such as light intensity, traffic flow, vehicle speed and vehicle distance in the tunnel is detected in real time, the light brightness is automatically adjusted according to the information, and it is ensured that the proper illumination level is kept in the tunnel all the time.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel lighting control, and in particular to a multi-sensor fusion intelligent tunnel dimming system using an improved DS algorithm. Background Art

[0002] Traditional tunnel dimming systems and intelligent tunnel lighting dimming systems make stepwise control of lighting according to the maximum traffic volume and driving speed. Specifically, when there is no vehicle passing through the tunnel, a dimming voltage is superimposed on the control loop of the original dimmable lighting fixtures to control the brightness of the lighting fixtures, reducing the brightness to the set low limit level; when there is a vehicle passing through the tunnel, the system gradually restores the brightness to the original set brightness level, realizing the intelligent control of "lights on when the vehicle comes, lights off when the vehicle passes", thereby reducing lighting energy consumption. However, traditional intelligent dimming generally has problems such as rigidity, ineffective dimming, and untimely dimming response, resulting in low actual energy-saving efficiency. Summary of the Invention

[0003] The object of the present invention is to provide a multi-sensor fusion intelligent tunnel dimming system using an improved DS algorithm to solve the problems of rigidity, ineffective dimming, untimely dimming response, and low actual energy-saving efficiency in traditional intelligent dimming based on the energy-saving goal of tunnel dimming using multi-sensor fusion technology.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] A multi-sensor fusion intelligent tunnel dimming system using an improved DS algorithm includes the following:

[0006] An information fusion module: used to receive the traffic flow information, vehicle speed information, illuminance information inside the tunnel, and luminance information outside the tunnel from each sensor, adopt the inverse DP probability conversion method to approximately convert the belief function into a probability function in reverse, and perform data fusion through the D-S evidence theory synthesis rule to obtain the tunnel environment information;

[0007] A brightness calculation module: used to calculate the required lighting brightness information according to the fused tunnel environment information;

[0008] A lighting adjustment module: used to send instructions to the lighting equipment according to the required lighting brightness information for brightness adjustment.

[0009] In the information fusion module, the traffic flow information, vehicle speed information, illuminance information inside the tunnel, and luminance information outside the tunnel are all sorted into 6 aspects of information according to certain rules and stored in the sample information database; X 1Indicates the average vehicle speed information, which is divided into 6 levels according to the actual complexity of control. Take C11 as the first category of vehicle speed of 120 - 100 km / h, C12 represents the second category of vehicle speed of 99 - 80 km / h, C13 is the third category of vehicle speed of 79 - 60 km / h, C14 is the fourth category of 59 - 40 km / h, C15 is the fifth category of 39 - 20 km / h, and C16 is the sixth category of 19 - 0 km / h; X 2 Indicates the brightness information outside the tunnel, which is divided into 6 categories. Take C21 as the first category of brightness of 7000 - 10000, C22 represents the second category of brightness of 6000 - 6999, C23 is the third category of brightness of 5000 - 5999, C24 is the fourth category of brightness of 4000 - 4999, C25 is the fifth category of brightness of 3000 - 3999, and C26 is the sixth category of brightness of 2000 - 2999; X 3 Indicates the traffic flow information, which is divided into 6 categories. Take C31 as the first category of traffic flow of 0 - 6 vehicles / min, take C32 as the second category of traffic flow of 7 - 12 vehicles / min, C33 is the third category of traffic flow of 13 - 18 vehicles / min, C34 is the fourth category of traffic flow of 19 - 24 vehicles / min, C35 is the fifth category of traffic flow of 25 - 30 vehicles / min, and C36 is the sixth category of traffic flow > 30 vehicles / min; X 4 Indicates the illuminance information inside the tunnel. Take C41 as the first category of illuminance of 6000 - 7000 cd, take C42 as the second category of illuminance of 5000 - 6000 cd, C43 is the third category of illuminance of 4000 - 5000 cd, C44 is the fourth category of illuminance of 3000 - 4000 cd, C45 is the fifth category of illuminance of 2000 - 3000 cd, and C46 is the sixth category of illuminance of 0 - 2000 cd.

[0010] The fusion process of the information fusion module is as follows:

[0011] Step 1: Determine whether there is a conflict in the evidence sources of the received sensor information. If there is no conflict, directly use the Dempster combination rule to fuse and obtain the tunnel environment information. If there is a conflict, execute the following Step 2;

[0012] Step 2: For the n received evidences, calculate the distance and similarity between each pair of evidences respectively, and calculate the support degree and credibility of each evidence respectively, and execute the following Step 3;

[0013] Step 3: Calculate the average evidence of the evidence sources, replace the conflicting evidence, and inherit the corresponding weights, and execute the following Step 4;

[0014] Step 4: Modify the evidence, perform weighted averaging on the modified evidence, and execute the following Step 5;

[0015] Step 5: Use the Dempster combination rule to fuse and obtain the tunnel environment information.

[0016] In the information fusion module, the framework θ is a complete identification framework containing n mutually distinct propositions, P(θ) is the set of all subsets, and m 1 and m 2 are two BPAs on the identification framework. The distance between the evidence bodies m 1 and m 2 is expressed as

[0017] In the information fusion module, the similarity measure between the evidence bodies m 1 and m 2 is Sim(m i , m j ) = 1 - d BPA (m i , m j ); i, j = 1, 2,......n.

[0018] In the information fusion module, the support degree Sup(m i ) of the evidence body m i is

[0019] In the information fusion module, the credibility of the evidence body m i is

[0020] As described above, according to the change of light intensity inside and outside the tunnel, the light intensity, traffic flow, vehicle speed, vehicle distance, etc. inside the tunnel are detected in real time, and the light brightness is automatically adjusted according to this information to ensure that the appropriate lighting level is always maintained inside the tunnel.

[0021] Based on the foregoing solution, in an improved solution, the dimming system further includes a segmented setting module: used to divide the tunnel into different numbers of segments according to the range of the brightness information outside the tunnel, and classify the sensors and lighting devices included in each segment accordingly, so as to complete the brightness adjustment. In this way, according to the segmentation of the brightness outside the tunnel, when the weather is dark and the light brightness is weak, the segments can be increased near the entrance of the tunnel to improve the dimming accuracy. When the weather is good and the light brightness is high, the segments can be reduced far inside the tunnel to seek a balance between the loss of fusion calculation and ensuring the illuminance.

[0022] Due to the adoption of the above technical solution, the present invention has the following beneficial effects:

[0023] 1. The present invention improves the lighting efficiency: by automatically adjusting the light brightness, the system can ensure that the appropriate lighting level is always maintained inside the tunnel according to the change of light intensity inside and outside the tunnel, thereby improving the lighting efficiency.

[0024] 2. Enhance driving safety: The system can detect information such as the light intensity, traffic flow, vehicle speed, and vehicle distance in the tunnel in real time, and automatically adjust the light brightness according to this information, thereby enhancing driving safety.

[0025] 3. Reduce energy consumption: By intelligently adjusting the light brightness, the system can significantly reduce energy consumption and achieve the goal of energy conservation and emission reduction.

[0026] 4. Improve system robustness: The application of improved algorithms enables the system to process uncertain information and conflicting evidence, thereby improving the robustness and accuracy of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is the system block diagram of the present invention.

[0028] Figure 2 is the fusion flow chart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0029] The following further describes the specific implementation of the invention with reference to the accompanying drawings.

[0030] Embodiment 1

[0031] A multi-sensor fusion intelligent tunnel dimming system using an improved DS algorithm of the present invention includes modules such as environmental perception and fusion strategy, and planned dimming control.

[0032] The system mainly consists of the following parts:

[0033] Sensor network: including light intensity sensors, traffic flow sensors, vehicle speed sensors, vehicle distance sensors, meteorological sensors, etc., for real-time collection of various information inside and outside the tunnel.

[0034] Central control system: responsible for receiving sensor data, performing multi-sensor information fusion, and calculating the required light brightness.

[0035] Dimming controller: adjusts the light brightness of the lighting lamps in the tunnel according to the instructions of the central control system.

[0036] Communication network: used for data transmission between the sensors and the central control system.

[0037] The system realizes the following functions:

[0038] Real-time data collection: The sensor network real-time collects information such as the light intensity, traffic flow, and meteorological conditions inside and outside the tunnel, and transmits this data to the central control system through the communication network.

[0039] Multi-sensor information fusion: Central control system - adopts the inverse DP probability conversion method, based on the DP synthesis rule, and inversely converts the belief function to approximately convert it into a probability function. Finally, data fusion is carried out through the D-S evidence theory synthesis rule to obtain more accurate and comprehensive environmental information.

[0040] Calculation of light brightness: According to the fused environmental information, the central control system automatically calculates the required light brightness.

[0041] Adjustment of lighting equipment: The central control system sends instructions to the lighting equipment through the communication network to adjust the light brightness in the tunnel.

[0042] During the tunnel dimming process, the driving environment in the tunnel is constantly changing, such as average vehicle speed, outside-tunnel brightness, hourly traffic flow, and in-tunnel illuminance. Each sensor extracts the required target features from the obtained target information, calculates the membership degree of the reference target according to the principle of maximum membership degree, and sends it to the fusion center for fusion.

[0043] First, extract 4 different types of information from the factor information as samples for building the model. Each sample point information needs to be sorted into 6 aspects of information according to certain rules and stored in the sample information database. It is divided into 24 categories in total, represented by C,(i = 1, 2, 3, 4), and each group is represented by [X 1 ,X 2 ,X 3 ,…,Xn}.

[0044] 1) X 1 represents the average vehicle speed information, which is divided into 6 levels according to the actual complexity of the control. If C 11 is Class I vehicle speed of 120 - 100 km / h, and the corresponding C 12 represents Class II vehicle speed of 99 - 80 km / h; C 13 is the third class of vehicle speed of 79 - 60 km / h; …… C 16 is the sixth class of 19 - 0 km / h.

[0045] 2) X 2 represents the outside-tunnel brightness information, which is divided into 6 categories. If C 21 is Class I brightness of 7000 - 10000, and the corresponding C 22 represents Class II brightness of 6000 - 6999; C 23 is the third class of brightness of 505999; …… C 26 is Class VI brightness of 999 - 2000.

[0046] 3) X 3 represents the traffic flow information, which is divided into 6 categories. If C 31 is Class I traffic flow of 0 - 6 vehicles / min. If C32 For Class II, the traffic flow is 7 - 12 vehicles / min, …… C 36 For Class VI, the traffic flow > 30 vehicles / min.

[0047] 4) X 4 Indicates the illuminance information inside the tunnel. If C is taken 41 For Class I, the illuminance inside the tunnel is 6000 - 7000 cd. If C is taken 42 For Class II, the illuminance is 5000 - 6000 cd, …… C 46 For Class VI, the illuminance is 0 - 2000 cd.

[0048] Such as Figure 2 As shown, the specific steps for the system fusion algorithm are as follows:

[0049] ① First, determine whether there is a conflict in the evidence sources. If not, directly fuse using the Dempster combination rule. Otherwise, if there is a conflict, execute ②;

[0050] ② For the n evidences of the system, calculate the distance and similarity between each pair of evidences according to Equations (1) and (2) respectively;

[0051] ③ Calculate the support degree and credibility of each evidence according to Equations (3) and (4) respectively;

[0052] ④ Calculate the average evidence of the evidence sources to replace the conflicting evidence and inherit the corresponding weights;

[0053] ⑤ Perform weighted averaging on the corrected evidence and then fuse using the Dempster combination rule.

[0054] Let the frame θ be a complete identification frame containing n mutually distinct propositions, P(θ) be the set of all subsets, and m 1 and m 2 be two BPAs on the identification frame. The distance between m 1 and m 2 is expressed as:

[0055]

[0056] D is a 2 N × 2 N matrix, and the elements in the matrix are:

[0057]

[0058] d BPA (m 1 , m 2 ) is calculated specifically according to Equation (1).

[0059]

[0060] In formula (1), Ilmll 2 =(m,m); (m 1 ,m 2 ) is the inner product of two vectors, that is

[0061]

[0062] d BPA The distance function can effectively represent the comprehensive influence of the focal elements and the basic probability assignment between two pieces of evidence, reflecting the difference between the evidences.

[0063] Suppose the system collects n pieces of evidence. Using formula (1), the evidence distance between two evidence bodies m 1 and m 2 is obtained and expressed as a distance matrix:

[0064]

[0065] The similarity measure between evidence bodies m 1 and m 2 is:

[0066] Sim(m i ,m j ) = 1 - d BPA (m;,m,); i,j = 1,2,......n; (2)

[0067] The result is represented by a similarity matrix as:

[0068]

[0069] The greater the distance between two pieces of evidence, the smaller their similarity. The support degree Sup(m) of evidence body m; is:

[0070]

[0071] After obtaining the support degree of an evidence m i , the credibility of evidence m i can be obtained:

[0072]

[0073] It is easy to know that That is, the credibility can be used as the weight of each evidence: W = (W 1 ,W 2 ,…Wn), where

[0074] As mentioned above, the purpose of information fusion is to reduce uncertainty and obtain a firm and reliable decision-making result. As a widely used uncertain reasoning method, the D-S theory achieves the purpose of decision-making through approximate reasoning under uncertain conditions. However, the quality of the highly uncertain problem after fusion may have an adverse impact on decision-making to a certain extent. The D-S theory cannot solve the problem when there is serious or complete conflict of evidence, and when there are composite focal elements, an appropriate probability conversion method needs to be adopted to reduce the uncertainty of the information source before performing D-S synthesis. Therefore, when conflicts occur in the information obtained by the sensor, the final judgment cannot be given. In order to avoid the loss of effective information and make full use of conflicting evidence; at the same time, to solve the robustness and one-vote veto phenomenon in the combination of conflicting evidence, a distance function is introduced to measure the degree of support and average evidence among the various evidences in the system. The advantages of this system are as follows:

[0075] Improve lighting efficiency: By automatically adjusting the light brightness, the system can ensure that the lighting level in the tunnel is always appropriate according to the change of light intensity inside and outside the tunnel, thus improving the lighting efficiency.

[0076] Enhance driving safety: The system can detect information such as light intensity, traffic flow, vehicle speed, and vehicle distance in the tunnel in real time, and automatically adjust the light brightness according to this information, thus enhancing driving safety.

[0077] Reduce energy consumption: By intelligently adjusting the light brightness, the system can significantly reduce energy consumption and achieve the goal of energy conservation and emission reduction.

[0078] Improve system robustness: The application of the improved algorithm enables the system to process uncertain information and conflicting evidence, thus improving the robustness and accuracy of the system.

[0079] Embodiment 2

[0080] The foregoing Embodiment 1 can be designed for overall tunnel dimming adjustment operation or segmented tunnel dimming adjustment operation. This Embodiment 2 will be further improved to perform dynamic segmented adjustment according to the brightness of the external environment of the tunnel (the tunnel is divided into multiple segments and the segmented tunnel dimming adjustment operation is performed). The dimming system further includes a segmented setting module: used to adjust the number of segments (divide the tunnel into different numbers of segments) and / or adjust the segment length (adjust the length of each segment) according to the range of the brightness information outside the tunnel. Specifically, it can only adjust the number of segments, only adjust the segment length, or simultaneously adjust the number of segments and the segment length, and classify the sensors and lighting devices included in each segment accordingly, thereby completing the tunnel dimming. In this way, according to the segmentation of the brightness outside the tunnel, when the weather is dark and the light brightness is weak, the number of segments can be increased (for example, divided into the first, second, third, fourth, and last segments) and / or the length of each segment can be adjusted (for example, reducing the lengths of the first and last segments, while the second and fourth segments remain unchanged or are appropriately reduced, and the third segment is appropriately increased) to improve the dimming accuracy. When the weather is good and the light brightness is high, the number of segments can be reduced (for example, divided into the first, second, and last segments) to seek a balance between the energy consumption loss of operations such as the fusion calculation of the tunnel dimming and ensuring the illuminance and accurate dimming.

[0081] Among them, if the external light environments of each tunnel segment are quite different, a separate segmented tunnel dimming adjustment operation is required (specifically including the processing processes of the information fusion module, brightness calculation module, and lighting adjustment module). Of course, in the case where the detection data of the external light environments of the tunnel segments are basically the same, the same segmented tunnel dimming adjustment operation can also be adopted (for example, the detection data of the external light environments of the first and last segments are basically the same, and the same tunnel dimming adjustment operation is adopted).

[0082] Based on the foregoing examples, in a preferred example, on the basis of ensuring that the illuminance meets the requirements (and not excessively), a comparative analysis is made of the overall energy consumption of each segmented tunnel dimming adjustment operation and its lighting, whether the overall energy consumption of more and finer segmented tunnel dimming adjustment operations and their lighting (including lighting energy consumption and tunnel dimming adjustment operation energy consumption) is lower, or whether the overall energy consumption of fewer and coarser segmented tunnel dimming adjustment operations and their lighting is lower, and then the segmented tunnel dimming adjustment operation strategy with lower overall energy consumption is selected to achieve the purpose of lower actual energy consumption, that is, actual energy saving.

[0083] It should be noted that the examples of the above embodiments can be preferably selected one or more in combination according to actual needs. The drawings of multiple examples using a set of combined technical features will not be elaborated one by one here.

[0084] The above description is a detailed explanation and illustration of the preferred and feasible embodiments of the present invention, but these descriptions are not intended to limit the scope of protection required by the present invention. Any equivalent changes or modifications made under the technical teachings disclosed by the present invention shall fall within the scope of patent protection covered by the present invention.

Claims

1. A multi-sensor fusion intelligent tunnel dimming system using an improved DS algorithm, characterized in that: Includes the following: Information fusion module: used to receive the traffic flow information, speed information, illumination information inside the tunnel and brightness information outside the tunnel from each sensor, adopt the inverse DP probability conversion method, and convert the credibility function into a probability function through inverse conversion. Then, data fusion is performed through the DS evidence theory synthesis rule to obtain the tunnel environment information. Brightness calculation module: used to calculate the required light brightness information based on the fused tunnel environment information; Lighting adjustment module: used to send instructions to lighting equipment to adjust the brightness according to the required light brightness information.

2. According to claim 1, a multi-sensor fusion intelligent tunnel dimming system using an improved DS algorithm is characterized in that: In the information fusion module, traffic flow information, vehicle speed information, illumination information in the tunnel and brightness information outside the tunnel are organized into 6 aspects of information according to certain rules and stored in the sample information database; X1 represents the average vehicle speed information, which is divided into 6 levels according to the actual complexity of control, with C11 representing the speed of Class I vehicle at 120-100 km / h, C12 representing the speed of Class II vehicle at 99-80 km / h, and C13 representing the speed of Class III vehicle at 79-60 km / h. hours, C14 is Class IV with a brightness of 59-40 km / h, C15 is Class V with a brightness of 39-20 km / h, and C16 is Class VI with a brightness of 19-0 km / h; X2 represents the brightness information outside the tunnel, which is divided into 6 categories, C21 represents Class I with a brightness of 7000-10000, C22 represents Class II with a brightness of 6000-6999, C23 represents Class III with a brightness of 5000-5999, and C24 represents Class IV with a brightness of 4000-4999 , C25 is V-class brightness 3000-3999, C26 is VI-class brightness 2000-2999; X3 represents traffic flow information, which is divided into 6 categories, C31 is I-class traffic flow 0-6 vehicles / min, C32 is II-class traffic flow 7-12 vehicles / min, C33 is III-class traffic flow 13-18 vehicles / min, C34 is IV-class traffic flow 19-24 vehicles / min, C35 is V-class traffic flow 25-30 vehicles / min, C 36 represents Class VI traffic flow >30 vehicles / min; X4 represents the illumination information in the tunnel, C41 represents Class I illumination of 6000-7000cd, C42 represents Class II illumination of 5000-6000cd, C43 represents Class III illumination of 4000-5000cd, C44 represents Class IV illumination of 3000-4000cd, C45 represents Class V illumination of 2000-3000cd, and C46 represents Class VI illumination of 0-2000cd.

3. According to claim 1, a multi-sensor fusion intelligent tunnel dimming system using an improved DS algorithm is characterized in that: The fusion process of the information fusion module is as follows: Step 1: Determine whether the evidence sources of the received sensor information are conflicting. If there is no conflict, directly use the Dempster combination rule to fuse the tunnel environment information. If there is a conflict, execute the following step 2; Step 2: For the n pieces of evidence received, calculate the distance and similarity between each pair of evidence, and calculate the support and credibility of each piece of evidence, and execute the following step 3; Step 3: Calculate the average evidence of the evidence source, replace the conflicting evidence, and inherit the corresponding weights, and execute the following step 4; Step 4: Correct the evidence, perform weighted averaging on the corrected evidence, and execute the following step 5; Step 5: Use Dempster combination rule to fuse and obtain tunnel environment information.

4. According to claim 3, a multi-sensor fusion intelligent tunnel dimming system using an improved DS algorithm is characterized in that: In the information fusion module, frame θ is a complete identification frame containing n different propositions, P(θ) is the set of all subsets, m1 and m2 are two BPAs on the identification frame, and the distance between evidence bodies m1 and m2 is expressed as 5. The multi-sensor fusion intelligent tunnel dimming system using the improved DS algorithm according to claim 3 is characterized by: In the information fusion module, the similarity measure of evidence bodies m1 and m2 is Sim(m i ,m j )=1-d BPA (m i ,m j ); i,j=1,2,......n.

6. The multi-sensor fusion intelligent tunnel dimming system using the improved DS algorithm according to claim 3, characterized in that: In the information fusion module, the evidence body m i The support Sup(m i )for 7. The multi-sensor fusion intelligent tunnel dimming system using the improved DS algorithm according to claim 3, characterized in that: In the information fusion module, the evidence body m i The credibility is 8. The multi-sensor fusion intelligent tunnel dimming system using the improved DS algorithm according to claim 1, characterized in that: It also includes a segment setting module: used to adjust the number of segments and / or the length of the segments according to the range of brightness information outside the tunnel, and to classify the sensors and lighting equipment included in each segment accordingly, thereby completing the tunnel dimming.

9. The multi-sensor fusion intelligent tunnel dimming system using the improved DS algorithm according to claim 8, characterized in that: The segmented setting module also includes the following contents: comparing and analyzing the dimming adjustment operations of each segmented tunnel and the overall energy consumption of its lighting, and selecting the segmented tunnel dimming adjustment operation with low overall energy consumption.

Images