AI-based infrared sensor ship lock defect monitoring system and method thereof

Abstract

The invention discloses an AI-based infrared sensor ship lock defect monitoring system and method, and belongs to the field of ship lock monitoring. Intensive infrared sensors which are distributed in a staggered mode are installed on the opposite sides of gates on the two sides of the ship lock, so that a monitoring array is obtained, and temperature data monitored by monitoring points and distance data of the gates on the opposite sides are collected; the temperature variation and the distance variation of the monitoring array are normalized to obtain the first risk value of the gate, and due to the fact that the first risk value is comprehensively obtained by the temperature variation and the distance variation of the monitoring array, the first risk value can accurately reflect whether the gate body has defects or not. Besides, movement of a communicating valve plate on a station department is divided into three stages, each movement stage is divided into a plurality of opening sub-areas, the risk value of each stage is obtained according to the vertical variation of the valve plate moving to each opening sub-area in each stage, and meanwhile the risk values of all the stages are combined to obtain a second risk value.

Description

Technical Field

[0001] The present invention relates to ship lock monitoring, and in particular to an AI-based infrared sensor ship lock defect monitoring system and method. Background Art

[0002] A ship lock is a box-shaped hydraulic structure designed to ensure smooth passage of ships and mitigate concentrated water level differences in a waterway. This box-shaped navigation structure uses gates at both ends to raise and lower the water level, allowing ships to overcome concentrated water level differences. A ship lock is a crucial structure for water navigation and plays a vital role in water transport. As the infrastructure that supports smooth waterway passage, its proper operation is crucial for smooth navigation. To ensure proper operation, the lock's operating status is typically monitored in real time. However, current lock monitoring methods only monitor external interference factors, such as whether the gates are opening and closing properly and whether there are any obstructions within the lock. Defects within the lock itself can also hinder its operation, but current monitoring methods are unable to directly and effectively detect these defects in real time, such as dents and cracks in the gates or the proper functioning of the valve plates connecting the two lock chambers. Summary of the Invention

[0003] Purpose of the invention: The purpose of the present invention is to provide an AI-based infrared sensor lock defect monitoring system and method; to solve the problem that the current lock monitoring method cannot effectively monitor the lock's own defects in real time.

[0004] Technical solution: An AI-based infrared sensor lock defect monitoring method, including: S1. Sensor deployment: Fixed infrared sensors are installed on opposite sides of the gates on both sides of the lock chamber, and infrared sensors are installed on the connecting valve plates of the gates; S2. Data collection and processing: Each infrared sensor installed on the opposite side of the gate is used as a monitoring point. The temperature monitored by each monitoring point is collected. At the same time, the distance between each monitoring point and the opposite gate is collected. The temperature and distance monitored by each monitoring point are then normalized to obtain the first risk value of the gate. S3. Data preprocessing: Based on the set opening of the gate's connecting valve plate, the valve plate's movement range is divided into multiple sub-opening zones. An infrared sensor installed on the connecting valve plate monitors the vertical distance between the connecting valve plate and the gate after the valve plate moves to each sub-opening zone. The vertical distances in each sub-opening zone are compared, and the comparison results are used as the second risk value. S4. Analysis and judgment: Obtain a ship lock risk contribution value based on the first risk value obtained in step S2 and the second risk value obtained in step S3, and determine whether the ship lock has defects based on the obtained ship lock risk contribution value; S5. Processing and control: After the lock is judged to have defects through the lock risk contribution value in step S4, the severity of the lock defect is judged, and the lock is processed and controlled according to the judged severity of the lock defect.

[0005] Preferably, in step S1, fixed infrared sensors are installed on opposite sides of the gates on both sides of the lock chamber in an alternating manner, the infrared sensor at the connecting valve plate is installed directly above the connecting valve plate and is distributed along the same central axis as the connecting valve plate, and the infrared sensor is a waterproof infrared sensor.

[0006] Preferably, in step S2, when obtaining the first risk value of the gate, the temperature collected at each monitoring point is first compared with the temperature threshold to obtain the temperature change of each monitoring point, and the distance collected at each monitoring point is compared with the distance threshold to obtain the distance change of each monitoring point; then the temperature change and the distance change of each monitoring point are normalized to obtain the first risk value of the gate, specifically:

[0007] in, S 1 is the first risk value of the gate, T act,i is the actual temperature of each monitoring point i, T y,i is the temperature threshold for each monitoring point i, α T is the temperature normalized weight coefficient, D act,i is the actual distance from each monitoring point i to the gate on the other side, D y,i is the distance threshold of each monitoring point i between the gates on both sides, α D is the distance normalization weight coefficient, ε i is the noise term for each monitoring point.

[0008] Preferably, in step S3, when obtaining the second risk value of the gate, the movement of the connecting valve plate is divided into three stages: acceleration, uniform speed and deceleration. In the three stages, the three stages are divided into multiple sub-opening areas according to the total time of the three stages; in the acceleration stage, the vertical amount of the connecting valve plate after reaching each sub-opening area is obtained, and the risk value of the acceleration stage is obtained according to the comparison result of the vertical amount of each sub-opening area. S 21 :

[0009] Similarly, the risk values of the uniform speed stage and the deceleration stage can be obtained S 22 、 S 23 :

[0010] in, L k is the vertical distance between each sub-opening area k and the gate; Then, the risk values of the three stages are combined and normalized to obtain the second risk value of the gate. S 2 :

[0011] in, α L is the normalized weight coefficient of the second risk value.

[0012] Preferably, in the acceleration, uniform speed and deceleration stages, 1s is used as the time node to divide the sub-opening area. At the same time, 1s is used as the monitoring node in the acceleration, uniform speed and deceleration stages to monitor the connecting valve plate entering the next opening area.

[0013] Preferably, in step S4, when judging whether the ship lock has defects, first, the ship lock risk contribution value S=S1+S2 is obtained according to the first risk value S1 and the second risk value S2, and then the temperature threshold, distance threshold and vertical change of the connecting valve plate at each stage of each monitoring point are obtained, and the above parameters are normalized to obtain the ship lock defect judgment threshold parameter M:

[0014] in, α 、 β 、 γ Normalization parameters of temperature threshold, distance threshold and vertical change, L k,A1 、 L k,A2 、 L k,A3 are the vertical values of each sub-opening area of the connecting valve plate in the acceleration, uniform speed and deceleration stages, A 1 、 A 2 、 A 3are the total amount of neutron opening area in the acceleration, uniform speed and deceleration stages respectively; if S>M, it is judged that there is a defect in the lock, otherwise, there is no defect.

[0015] Preferably, in step S5, when M<S<1.5M, it is judged that the severity of the lock defect is low, and the throughput of the lock is controlled at this time; when S≥1.5M, it is judged that the severity of the lock is high, and the lock is closed at this time, or the water level in the lock chamber is reduced while closing the lock.

[0016] An AI-based infrared sensor lock defect monitoring system, which is implemented based on the above method and includes a data acquisition module, a data preprocessing module, an analysis and judgment module, and a control module; Data acquisition module: used to collect the actual temperature of each monitoring point and the actual distance data between the monitoring point and the gate on the opposite side, and at the same time collect the vertical volume of the connecting valve plate after it reaches each sub-opening area during the movement; Data preprocessing module: obtains the temperature threshold, distance threshold, and vertical change of the connecting valve plate at each stage of each monitoring point, and normalizes the above parameters to obtain the threshold parameters for judging the ship lock defects; at the same time, the information parameters collected by the data acquisition module are processed to obtain the first risk value and the second risk value of the ship lock, and the ship lock risk contribution value is obtained based on the first risk value and the second risk value; Analysis and judgment module: Compares the lock risk contribution value obtained by the data preprocessing module with the lock defect judgment threshold parameter to determine whether the lock has defects. If a defect is found, the severity of the lock defect is further determined and the signal is fed back to the control module. Control module: used to receive the control signal fed back by the analysis and judgment module, and control the lock according to the control signal.

[0017] Beneficial effects: The present invention installs densely distributed infrared sensors on opposite sides of the gates on both sides of the ship lock, with each infrared sensor as a monitoring point, thereby obtaining a monitoring array, and collecting the temperature data monitored by the monitoring point and the distance data from the gate on the opposite side, and normalizing the temperature change and distance change of the monitoring array to obtain the first risk value of the gate. Since the first risk value is obtained by combining the temperature change and distance change of the monitoring array, the first risk value can accurately reflect whether there is a defect in the gate body. In addition, by dividing the movement of the connecting valve plate on the station door into three stages, and dividing each movement stage into multiple sub-opening areas, and according to the vertical change of the valve plate movement to each sub-opening area in each stage, the risk value of each stage is obtained, and the risk values of each stage are combined to obtain the second risk value. Since the second risk value is obtained by combining the vertical change of each sub-opening area in each stage of the connecting valve plate, the second risk value can accurately reflect whether there is abnormal movement of the connecting valve plate. The ship lock risk contribution value is obtained by combining the first risk value and the second risk value. Therefore, the risk contribution value can accurately reflect whether there are defects in the ship lock itself, thereby accurately monitoring the defects caused by the ship lock itself. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a system flow chart; Figure 2 This is a schematic diagram of the deployment of infrared sensors; Figure 3 It is a schematic diagram of the division of the movement stages of the connecting valve plate. DETAILED DESCRIPTION

[0019] In order to make the technical solution of the present invention clearer, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. Example

[0020] like Figure 1 As shown, first, waterproof infrared sensors are staggered and fixed on opposite sides of the gates on both sides of the lock chamber, and waterproof infrared sensors are installed on the connecting valve plate of the gate and the same central axis as the connecting valve plate, as shown in FIG. Figure 2As shown. Each infrared sensor is used as a monitoring point. Each monitoring point will monitor the temperature of the gate on the opposite side and the distance between the gate and the gate on the opposite side in real time, and collect the information data monitored by each monitoring point through data acquisition. The collected information data is then handed over to the data preprocessing module for processing, so that the first risk value of the gate can be obtained, specifically: the temperature collected at each monitoring point is compared with the temperature threshold to obtain the temperature change of each monitoring point, and the distance collected at each monitoring point is compared with the distance threshold to obtain the distance change of each monitoring point; then the temperature change and distance change of each monitoring point are normalized to obtain the first risk value of the gate. S 1 :

[0021] in, T act,i is the actual temperature of each monitoring point i, T y,i The temperature threshold of each monitoring point i is adjusted dynamically with the change of seasons. For example, the temperature threshold is lowered in winter and raised in summer. In this embodiment, summer is used as an example. α T is the temperature normalized weight coefficient. When a gate has defects such as dents, the distance temperature at the corresponding position of the gate will change. Therefore, temperature is one of the more important factors that can show gate defects. However, gate dents and other defects do not mean that the gate must be unusable. Therefore, its importance in showing gate defects is lower than that of the vertical change of the connecting valve plate. Therefore, α T Set to 0.3, D act,i is the actual distance from each monitoring point i to the gate on the other side, D y,i is the distance threshold of each monitoring point i between the gates on both sides, α D is the distance normalization weight coefficient. When a gate has defects such as depression, the distance at the corresponding position of the gate will change. Therefore, distance is one of the more important factors that can show gate defects. However, gate depression does not mean that the gate cannot be used. Therefore, its importance in showing gate defects is lower than that of the vertical change of the connecting valve plate. Therefore, α D Set to 0.3, ε iThe noise term for each monitoring point (environmental disturbances, sensor errors, etc.) has a value range of (0, 1). The specific value is dynamically adjusted based on factors such as external weather, water quality, and sensor failure. When the external weather is poor, the water quality is poor, or the sensor fails, the value is higher, and vice versa. In this embodiment, the value is 0.1, assuming good external weather and water quality, and no sensor failures. For example, four infrared sensors are staggered on opposite sides of the gates at both ends of the lock chamber, and each infrared sensor serves as a monitoring point. The data they monitor is as follows: At the first monitoring point, the temperature is 27°C and the distance is 131.1m. The temperature threshold is 28°C and the distance threshold is 130m.

[0022] Second monitoring point: The monitored temperature is 27.5°C and the monitored distance is 133.21m. The temperature threshold is 29°C and the distance threshold is 133m.

[0023] At the third monitoring point, the temperature was 26.3°C and the distance was 130.54m. The temperature threshold was 27°C and the distance threshold was 131m.

[0024] Second monitoring point: The monitored temperature is 28.2°C and the monitored distance is 132.62m. The temperature threshold is 29°C and the distance threshold is 132m.

[0025] Since the current ship locks generally use herringbone gates, the distribution of each monitoring point is different, so the monitored temperature and distance will vary. At this time, the first risk value of the gate can be obtained S 1 for: S1=(1+1.5+0.7+0.8)×0.3+(1.1+0.21+0.46+0.62)×0.3=1.917.

[0026] At the same time, according to the opening set by the connecting valve plate, the movement of the connecting valve plate is divided into three stages: acceleration, uniform speed and deceleration. In each stage, according to the running time of each stage, each node is divided into several sub-opening areas with 1s as the time node. At the same time, 1s is used as the monitoring node to monitor the movement of the connecting valve plate to the next sub-opening area, such as Figure 3As shown. Then, the infrared sensor is used to monitor the vertical distance between the connecting valve plate and the gate after it moves to each sub-opening area in each stage in real time, and the data is collected through the data acquisition module. Then, the results are compared based on the vertical distance of each sub-opening area, so that the risk value of each stage can be obtained. In the acceleration stage, the connecting valve plate is in an accelerated state. In this stage, the vertical change of each sub-opening area is constantly increasing. Therefore, the vertical distance between the connecting valve plate and each sub-opening area in the acceleration stage is compared, and the comparison result is used as the risk value of the acceleration stage. S 21 :

[0027] For example, in the acceleration phase, the acceleration time is 3s, so the acceleration phase is divided into three sub-opening areas. The vertical volume of the first sub-opening area is 0.1m, the vertical volume of the second sub-opening area is 0.3m, and the vertical volume of the third sub-opening area is 0.6m. Since 0.6-0.3>0.3-0.2, and 0.3-0.2>0.1-0. Therefore, the risk value S of the acceleration phase is 21 =0.

[0028] Similarly, in the uniform speed stage, the risk value S 22 :

[0029] The duration of the uniform speed is 4 seconds. Therefore, the uniform speed stage is divided into four sub-opening areas. The vertical volume of the first sub-opening area is 0.7m, the vertical volume of the second sub-opening area is 0.8m, the vertical volume of the third sub-opening area is 0.9m, and the vertical volume of the third sub-opening area is 1.0m. Since 1.0-0.9=0.9-0.8=0.8-0.7=0.7-0.6. Therefore, the risk value S of the uniform speed stage is 22 =0.

[0030] Similarly, in the deceleration phase, its risk value S 23 :

[0031] The deceleration time is 3s, so the deceleration stage is divided into three sub-opening areas. The vertical amount of the first sub-opening area is 1.1m, the vertical amount of the second sub-opening area is 1.15m, and the vertical amount of the third sub-opening area is 1.16m. Since 1.16-1.15<1.15-1.1<1.1-1.0. Therefore, the risk value S of the deceleration stage is 23 =0.

[0032] Therefore, the above parameters are then processed by the data preprocessing module to obtain the second risk value of the gateS 2 :

[0033] in, α L is the normalized weight coefficient for the second risk value. The movement of the connecting valve plate plays a significant role in the proper functioning of the ship lock. When the connecting valve plate fails to function properly, the connection between the two lock chambers becomes blocked or nonexistent, preventing the gate from opening and preventing ships from entering the next lock chamber. Therefore, compared to temperature and distance changes, the movement of the connecting valve plate has a greater impact on the degree of gate defects, so it is set to 0.5. Based on the above analysis, the second risk value of the gate at this time, S2, = (0 + 0 + 0) × 0.5 = 0. Adding the first and second risk values of the gate yields the ship lock risk contribution value, S = S1 + S2 = 1.917 + 0 = 1.917.

[0034] At the same time, the data preprocessing module obtains the temperature threshold, distance threshold, and vertical change of the connecting valve plate at each stage of each monitoring point, and normalizes the above parameters to obtain the lock defect judgment threshold parameter M:

[0035] in, α 、 β 、 γ Normalized parameters of temperature threshold, distance threshold and vertical variation. According to the above analysis, since temperature and distance have relatively little influence on the presentation of gate defects compared with the vertical variation of the connecting valve plate, α 、 β 、 γ The values are 0.03, 0.03, and 0.04 respectively. L k,A1 、 L k,A2 、 L k,A3 are the vertical values of each sub-opening area of the connecting valve plate in the acceleration, uniform speed and deceleration stages, A 1 、 A 2 、 A 3 are the total amount of neutron aperture area in the acceleration, uniform speed and deceleration stages respectively. Thus, we can get the ship lock defect judgment threshold M = 0.8475 + 3.945 + 0.014 = 4.8065.

[0036] Then, the lock risk contribution value S = 1.917 and the lock defect judgment threshold M = 4.8065 are input into the analysis and judgment. Since the lock risk contribution value S is less than the lock defect judgment threshold M, the lock is determined to have no defects. Otherwise, the lock is determined to have defects. When a lock defect is determined, the severity of the lock defect is further assessed and a signal is fed back to the control module. If M < S < 1.5M, the lock defect severity is determined to be low, and the control module controls the lock throughput. If S ≥ 1.5M, the lock severity is determined to be high, and the control module closes the lock or simultaneously reduces the water level in the lock chamber.

[0037] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. An AI-based infrared sensor lock defect monitoring method, characterized in that: include: S1. Sensor deployment: Fixed infrared sensors are installed on opposite sides of the gates on both sides of the lock chamber, and infrared sensors are installed on the connecting valve plates of the gates; S2. Data collection and processing: Each infrared sensor installed on the opposite side of the gate is used as a monitoring point. The temperature monitored by each monitoring point is collected. At the same time, the distance between each monitoring point and the opposite gate is collected. The temperature and distance monitored by each monitoring point are then normalized to obtain the first risk value of the gate. S3. Data preprocessing: Based on the set opening of the gate's connecting valve plate, the valve plate's movement range is divided into multiple sub-opening zones. An infrared sensor installed on the connecting valve plate monitors the vertical distance between the connecting valve plate and the gate after the valve plate moves to each sub-opening zone. The vertical distances in each sub-opening zone are compared, and the comparison results are used as the second risk value. S4. Analysis and judgment: Obtain a ship lock risk contribution value based on the first risk value obtained in step S2 and the second risk value obtained in step S3, and determine whether the ship lock has defects based on the obtained ship lock risk contribution value; S5. Processing and control: After the lock is judged to have defects through the lock risk contribution value in step S4, the severity of the lock defect is judged, and the lock is processed and controlled according to the judged severity of the lock defect.

2. The AI-based infrared sensor ship lock defect monitoring method according to claim 1 is characterized in that: In step S1, fixed infrared sensors are installed on opposite sides of the gates on both sides of the lock chamber in a staggered distribution. The infrared sensor at the connecting valve plate is installed directly above the connecting valve plate and is distributed along the same central axis as the connecting valve plate. The infrared sensor is a waterproof infrared sensor.

3. The AI-based infrared sensor ship lock defect monitoring method according to claim 1 is characterized in that: In step S2, when obtaining the first risk value of the gate, the temperature collected at each monitoring point is first compared with the temperature threshold to obtain the temperature change of each monitoring point. At the same time, the distance collected at each monitoring point is compared with the distance threshold to obtain the distance change of each monitoring point. The temperature change and the distance change of each monitoring point are then normalized to obtain the first risk value of the gate, specifically: in, S 1 is the first risk value of the gate, T act,i is the actual temperature of each monitoring point i, T y,i is the temperature threshold for each monitoring point i, α T is the temperature normalized weight coefficient, D act,i is the actual distance from each monitoring point i to the gate on the other side, D y,i is the distance threshold of each monitoring point i between the gates on both sides, α D is the distance normalization weight coefficient, ε i is the noise term for each monitoring point.

4. The AI-based infrared sensor ship lock defect monitoring method according to claim 3 is characterized in that: In step S3, when obtaining the second risk value of the gate, the movement of the connecting valve plate is divided into three stages: acceleration, uniform speed, and deceleration. In the three stages, the three stages are divided into multiple sub-opening areas according to the total time of the three stages; in the acceleration stage, the vertical amount of the connecting valve plate after reaching each sub-opening area is obtained, and the risk value of the acceleration stage is obtained according to the comparison result of the vertical amount of each sub-opening area. S 21 : Similarly, the risk values of the uniform speed stage and the deceleration stage can be obtained S 22 、 S 23 : in, L k is the vertical distance between each sub-opening area k and the gate; Then, the risk values of the three stages are combined and normalized to obtain the second risk value of the gate. S 2 : in, α L is the normalized weight coefficient of the second risk value.

5. The AI-based infrared sensor ship lock defect monitoring method according to claim 4 is characterized in that: In the acceleration, uniform speed and deceleration stages, 1s is used as the time node to divide the sub-opening area. At the same time, 1s is used as the monitoring node in the acceleration, uniform speed and deceleration stages to monitor the connecting valve plate entering the next opening area.

6. The AI-based infrared sensor ship lock defect monitoring method according to claim 4 is characterized in that: In step S4, when judging whether the ship lock has defects, first, the ship lock risk contribution value S = S1 + S2 is obtained based on the first risk value S1 and the second risk value S2. Then, the temperature threshold, distance threshold, and vertical change of the connecting valve plate at each stage of each monitoring point are obtained, and the above parameters are normalized to obtain the ship lock defect judgment threshold parameter M: in, α 、 β 、 γ Normalization parameters of temperature threshold, distance threshold and vertical change, L k,A1 、 L k,A2 、 L k,A3 are the vertical values of each sub-opening area of the connecting valve plate in the acceleration, uniform speed and deceleration stages, A 1 、 A 2 、 A 3 are the total amount of neutron opening area in the acceleration, uniform speed and deceleration stages respectively; if S>M, it is judged that there is a defect in the lock, otherwise, there is no defect.

7. The AI-based infrared sensor ship lock defect monitoring method according to claim 1 is characterized in that: In step S5, when M<S<1.5M, it is judged that the severity of the lock defect is low, and the throughput of the lock is controlled at this time; when S≥1.5M, it is judged that the severity of the lock is high, and the lock is closed at this time, or the water level in the lock chamber is reduced while closing the lock.

8. A system using the AI-based infrared sensor ship lock defect monitoring method according to any one of claims 1 to 7, characterized in that: It includes data acquisition module, data preprocessing module, analysis and judgment module and control module; Data acquisition module: used to collect the actual temperature of each monitoring point and the actual distance data between the monitoring point and the gate on the opposite side, and at the same time collect the vertical volume of the connecting valve plate after it reaches each sub-opening area during the movement; Data preprocessing module: obtains the temperature threshold, distance threshold, and vertical change of the connecting valve plate at each stage of each monitoring point, and normalizes the above parameters to obtain the threshold parameters for judging the ship lock defects; at the same time, the information parameters collected by the data acquisition module are processed to obtain the first risk value and the second risk value of the ship lock, and the ship lock risk contribution value is obtained based on the first risk value and the second risk value; Analysis and judgment module: compares the lock risk contribution value obtained by the data preprocessing module with the lock defect judgment threshold parameter to determine whether the lock has defects; When a defect is detected, the severity of the lock defect is further determined and the signal is fed back to the control module; Control module: used to receive the control signal fed back by the analysis and judgment module, and control the lock according to the control signal.

Images