An evaluation method for satellite remote sensing emergency mission planning algorithm based on AHP hierarchical structure

Abstract

The invention discloses a satellite remote sensing emergency task planning algorithm evaluation method based on the AHP hierarchical structure; firstly, an evaluation index system is established, which can reflect the functional requirements of the actual problem on the planning algorithm and the mutual relationship between the evaluation indicators at different levels; then, the relationship between the basic elements in the evaluation system is analyzed, and a hierarchical structure of the system is established, wherein the highest level (target level) represents the purpose of the system, that is, the goal to be achieved by AHP, the middle level (criteria level) represents the intermediate links involved in adopting certain measures or policies to achieve the predetermined goals, and the lowest level (measure level) represents the measures or policies to solve the problems. The method evaluates the results of multiple emergency plans, selects the better plan for execution, and can take into account the timeliness of emergency tasks and the maximization of all task benefits. It has the characteristics of clear and visible logic, intuitive evaluation index system, objective and easy-to-obtain evaluation parameters, high calculation efficiency, easy implementation and debugging, easy expansion, strong compatibility, etc.

Description

Technical Field

[0001] The invention proposes a satellite remote sensing emergency task planning algorithm evaluation method based on the AHP hierarchical structure, which is suitable for the problem of evaluating the emergency planning algorithm when a satellite remote sensing system causes an event. Background Art

[0002] The satellite remote sensing system consists of remote sensing satellites, ground operation and control systems, ground application systems, ground station network systems and their affiliated ground stations, such as Figure 1 As shown in the figure, the ground station is divided into a tracking and control station and a data receiving station. The remote sensing satellite receives the remote sensing instructions sent by the ground tracking and control station, images the ground remote sensing, and sends the imaging data to the data receiving station; the ground operation and control system receives the needs of each remote sensing user, and coordinates the remote sensing needs, remote sensing satellites, data receiving stations, tracking and control stations and other resources, generates satellite control instructions, and sends the satellite instructions to the station network center, which arranges the appropriate tracking and control station to send the instructions to the satellite for execution; the data receiving station receives the satellite remote sensing data and sends it to the ground application system, which corrects and processes the original data, generates advanced data products and sends them to remote sensing users. In addition to ground stations, relay satellites are often used to send instructions and receive data.

[0003] At present, the number of ground stations and relay satellites in my country is relatively small compared to the number of remote sensing satellites. The measurement and control circles and receiving circles allocated to each remote sensing satellite every day are very limited. Therefore, my country's remote sensing satellite system usually adopts a periodic and batch processing operation mode. Every day, the remote sensing needs of the next 1 to 2 days are planned in a coordinated manner, and instructions are generated and uploaded to the satellite at one time. The satellite stores the remote sensing data and waits for the right opportunity to transmit it to the ground station in a centralized manner. Figure 2 shown.

[0004] Due to the shortage of measurement and control resources and receiving resources, remote sensing missions are usually scheduled quite full. When an emergency occurs and requires emergency earth observation, it will conflict with the tasks already scheduled by the satellite and ground receiving station. It is necessary to cancel the existing tasks on the satellite and ground station to perform the observation and reception of emergency tasks. In order to ensure that the earth observation data is transmitted as quickly as possible, it is often necessary to select the satellite passing by the station with the shortest observation time to transmit data, such as Figure 3As shown, the ground station may have been scheduled to receive data from other satellites in advance, which will affect the data transmission of other satellites. That is, if satellite A passes the top of the emergency mission the fastest, then satellite A will be arranged to perform earth observation. After satellite A performs the observation mission, it will pass by station B immediately, so station B should be arranged to receive data from satellite A, but station B has been scheduled to receive data from satellite C in advance, which will affect the data reception of satellite C. It can be seen that an emergency mission often affects the original mission arrangements of multiple satellites. When multiple emergency missions are performed in a short period of time, a chain reaction will often occur. In order to take into account the timeliness of emergency missions and maximize the benefits, the impact on other satellite missions should be minimized when arranging emergency missions. Therefore, it is necessary to establish a set of evaluation index systems and evaluation algorithms to evaluate multiple emergency mission planning schemes, so as to select the optimal scheme for execution. Summary of the invention

[0005] The present invention proposes an evaluation algorithm for emergency mission planning schemes when a satellite remote sensing system performs emergency earth observation missions. The present invention has the characteristics of clear and visible logic, intuitive evaluation index system, objective and easy-to-obtain evaluation parameters, high calculation efficiency, easy implementation and debugging, easy expansion, and strong compatibility.

[0006] The technical problem to be solved by the present invention is achieved by the following technical solutions:

[0007] A satellite remote sensing emergency mission planning algorithm evaluation method based on AHP hierarchical structure includes the following steps:

[0008] (1) Construct an indicator evaluation system, including comprehensive indicators, derived indicators and basic indicators. Comprehensive indicators include multiple derived indicators, and each derived indicator includes multiple basic indicators.

[0009] (2) Calculate the values ​​of each basic indicator in step (1) according to the satellite planning task and the algorithm planning results;

[0010] (3) normalizing the calculation results of each basic index value in step (2);

[0011] (4) Comprehensively consider the factors affecting the evaluation algorithm of each basic indicator, determine the weight of each basic indicator and the weight of each derived indicator; and weight the normalized basic indicators to calculate each derived indicator, and then weight each derived indicator to calculate the comprehensive indicator.

[0012] Furthermore, step (1) is specifically as follows:

[0013] The hierarchical structure of the indicator evaluation system is constructed, which includes three levels; the first level is the comprehensive indicator level, which is an overall evaluation of a certain planning algorithm; the second level is the derived indicator level, which is an evaluation of a specific aspect of the planning algorithm's performance, including emergency task completion indicators, algorithm performance indicators and resource utilization indicators; the third level is the basic indicator level, and the basic indicators of the emergency task completion indicators include task completion rate, task completion benefit and data transmission benefit, the basic indicators of the algorithm performance indicators include time requirements and disturbances to the entire system, and the basic indicators of the resource utilization indicators include the number of satellites, total power-on time, number of ground stations and total receiving time of ground stations.

[0014] Furthermore, step (2) specifically includes the following process:

[0015] 1) The basic indicators of emergency task completion status are calculated as follows:

[0016] After the task planning is completed, the completion status of each observation emergency task is obtained from the planning results, and the task completion rate is calculated. The calculation formula is: Among them, Number represents the total number of emergency tasks to be observed, N c Indicates the total number of completed observation emergency tasks, Mission j represents the jth emergency task, Inv(x) is the emergency task completion function, if the emergency task x is photographed by the satellite, then Inv(x) = 1; otherwise, Inv(x) = 0;

[0017] The task completion benefit index is the weighted sum of the different weighted observation emergency tasks completed. After the task planning is completed, the total weight of the emergency task completion is calculated according to the observation emergency task level. The formula is: w j is the weight of the jth emergency task, w j Larger values ​​indicate greater importance;

[0018] The formula for calculating the data downlink benefit is: DT j is the deadline of emergency task j, π j is the end time of the ground station downlink time window corresponding to emergency task j;

[0019] 2) The basic indicators of the algorithm performance indicators are calculated as follows:

[0020] The time requirement Time is the running time of the emergency task planning algorithm itself, which is used as the time requirement indicator Time;

[0021] The disturbance index Disturb of the entire system is the disturbance of the original planning scheme Where i is the task in the original planning scheme Φ, and the change of the task state in Φ is called disturbance, and its measurement is expressed as Δdisturb i , when task i in Φ does not change, Δdisturb i Take 0, when the execution time window of task i changes, Δdisturb i Take 1, when task i is deleted from the planning list, Δdisturb i Take 2;

[0022] 3) The basic indicators of resource utilization indicators are calculated as follows:

[0023] Number of satellites NSatellite: According to the planning results, the number of satellites under observation is counted as the satellite number indicator;

[0024] Total power-on time TOpen: According to the planning results, the total power-on time of the satellites used for observation is counted as the total power-on time indicator;

[0025] Number of ground stations NStation: According to the planning results, the number of ground stations receiving data is counted as the number of ground stations indicator;

[0026] Total receiving time of ground stations DStation: According to the planning results, the total receiving time of the ground stations performing observations is counted as the total receiving time indicator.

[0027] Furthermore, step (3) specifically includes the following process:

[0028] Normalization of emergency task completion indicators:

[0029] Query all previous emergency task planning plans, obtain the maximum value MaxCompletion and minimum value MinCompletion of the task completion rate, the maximum value MaxWeight and MinWeight of the task completion benefit, and the maximum value MaxBenefit and minimum value MinBenefit of the data transmission benefit;

[0030] The normalized index of the completion rate of this emergency task is: The normalized index of the benefit Weight of this task completion is calculated as: The normalized benefit index of the data transmission benefit of this emergency task is calculated as follows:

[0031] Normalization of algorithm performance indicators:

[0032] Query all previous emergency task planning plans to obtain the maximum value MaxTime and the minimum value MinTime of the algorithm time consumption, the maximum value MaxDisturb and the minimum value MinDisturb of the disturbance;

[0033] The normalized index of the time requirement of this algorithm is: The normalized disturbance index of the entire system is calculated as:

[0034] Normalization of resource utilization indicators:

[0035] Query all previous emergency mission planning plans, obtain the maximum value MaxNSat and the minimum value MinNSat of the number of satellites, the maximum value MaxTOpen and the minimum value MinTOpen of the total power-on time, the maximum value MaxNSta and the minimum value MinNSta of the number of ground stations, and the maximum value MaxTSta and the minimum value MinTSta of the total data reception time;

[0036] The normalized index of the number of satellites NSat is: The normalized index of the total startup time Open is calculated as: The normalized index of the number of ground stations NSta is calculated as: The normalized indicator of the total receiving time TSta of the ground station is calculated as:

[0037] Furthermore, step (4) specifically includes the following steps:

[0038] (401) Based on the experience of multiple emergency missions, the weights of various basic indicators are specified, including the weights of mission completion rate, mission completion efficiency, data transmission benefits, time requirements, disturbance to the entire system, number of satellites, total power-on time, number of ground stations and total ground station reception time;

[0039] (402) Based on the experience of multiple emergency missions, the weights of various derived indicators are specified, including the weights of the task completion indicator, the algorithm performance indicator, and the resource utilization indicator;

[0040] (403) weighting each basic indicator value normalized in step (3) with the basic indicator weight specified in step (401) to obtain a task completion indicator, an algorithm performance indicator, and a resource utilization indicator;

[0041] (404) The task completion status index, algorithm performance index and resource utilization index are weighted with the weights of each derived index specified in step (402) to obtain a comprehensive index. The emergency planning plan with a large comprehensive index value is better.

[0042] The present invention has the following advantages:

[0043] 1. The basic indicators defined in the present invention can be obtained by querying or statistically calculating the satellite planning tasks and algorithm planning results, and the parameters are easy to obtain immediately.

[0044] 2. The present invention adopts a two-layer weighted calculation method, which has a small amount of code and high operating efficiency;

[0045] 3. The present invention has the characteristics of easy expansion and strong compatibility;

[0046] 4. The present invention is also easy to implement and debug. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a working principle diagram of the satellite remote sensing system.

[0048] Figure 2 Planning flow chart for routine satellite missions.

[0049] Figure 3 Flowchart for satellite emergency mission planning.

[0050] Figure 4 This is the indicator system structure diagram of the present invention. DETAILED DESCRIPTION

[0051] The present invention will be further described below in conjunction with the accompanying drawings.

[0052] The present invention comprehensively considers the completion of emergency tasks, the performance of algorithms and the utilization of resources, and proposes a satellite remote sensing emergency task planning algorithm evaluation method based on the AHP hierarchical structure. First, an evaluation index system is established, which can not only reflect the functional requirements of the planning algorithm for actual problems, but also reflect the relationship between evaluation indicators at different levels; then a hierarchical analysis structure model is established to analyze the relationship between the basic elements in the evaluation system and establish a hierarchical structure of the system, in which the highest level (target layer) represents the purpose of the system, that is, the goal to be achieved by AHP, the middle level (criterion layer) represents the intermediate links involved in adopting certain measures or policies to achieve the predetermined goals, and these links are usually criteria that need to be considered, and the lowest level (measure layer) represents the measures or policies to solve the problem; specifically, the following steps are included:

[0053] (1) Construct an indicator evaluation system, including comprehensive indicators, derived indicators and basic indicators. Comprehensive indicators include multiple derived indicators, and each derived indicator includes multiple basic indicators. The details are as follows:

[0054] The hierarchical structure of the indicator evaluation system is constructed, which includes three levels; the first level is the comprehensive indicator level, which is an overall evaluation of a certain planning algorithm; the second level is the derived indicator level, which is an evaluation of a specific aspect of the planning algorithm's performance, including emergency task completion indicators, algorithm performance indicators and resource utilization indicators; the third level is the basic indicator level, and the basic indicators of the emergency task completion indicators include task completion rate, task completion benefit and data transmission benefit, the basic indicators of the algorithm performance indicators include time requirements and disturbances to the entire system, and the basic indicators of the resource utilization indicators include the number of satellites, total power-on time, number of ground stations and total receiving time of ground stations.

[0055] (2) Calculate the values ​​of each basic indicator in step (1) according to the satellite planning task and the algorithm planning results; the details are as follows:

[0056] 1) The basic indicators of emergency task completion status are calculated as follows:

[0057] After the task planning is completed, the completion status of each observation emergency task is obtained from the planning results, and the task completion rate is calculated. The calculation formula is: Among them, Number represents the total number of emergency tasks to be observed, N c Indicates the total number of completed observation emergency tasks, Mission j represents the jth emergency task, Inv(x) is the emergency task completion function, if the emergency task x is photographed by the satellite, then Inv(x) = 1; otherwise, Inv(x) = 0;

[0058] The task completion benefit index is the weighted sum of the different weighted observation emergency tasks completed. After the task planning is completed, the total weight of the emergency task completion is calculated according to the observation emergency task level. The formula is: w j is the weight of the jth emergency task, w j Larger values ​​indicate greater importance;

[0059] The formula for calculating the data downlink benefit is: DT j is the deadline of emergency task j, π j is the end time of the ground station downlink time window corresponding to emergency task j;

[0060] 2) The basic indicators of the algorithm performance indicators are calculated as follows:

[0061] The time requirement Time is the running time of the emergency task planning algorithm itself, which is used as the time requirement indicator Time;

[0062] The disturbance index Disturb of the entire system is the disturbance of the original planning scheme Where i is the task in the original planning scheme Φ, and the change of the task state in Φ is called disturbance, and its measurement is expressed as Δdisturb i , when task i in Φ does not change, Δdisturb i Take 0, when the execution time window of task i changes, Δdisturb i Take 1, when task i is deleted from the planning list, Δdisturb i Take 2;

[0063] 3) The basic indicators of resource utilization indicators are calculated as follows:

[0064] Number of satellites NSatellite: According to the planning results, the number of satellites under observation is counted as the satellite number indicator;

[0065] Total power-on time TOpen: According to the planning results, the total power-on time of the satellites used for observation is counted as the total power-on time indicator;

[0066] Number of ground stations NStation: According to the planning results, the number of ground stations receiving data is counted as the number of ground stations indicator;

[0067] Total receiving time of ground stations DStation: According to the planning results, the total receiving time of the ground stations performing observations is counted as the total receiving time indicator.

[0068] (3) Normalize the calculation results of each basic index value in step (2); the details are as follows:

[0069] Normalization of emergency task completion indicators:

[0070] Query all previous emergency task planning plans, obtain the maximum value MaxCompletion and minimum value MinCompletion of the task completion rate, the maximum value MaxWeight and MinWeight of the task completion benefit, and the maximum value MaxBenefit and minimum value MinBenefit of the data transmission benefit;

[0071] The normalized index of the completion rate of this emergency task is: The normalized index of the benefit Weight of this task completion is calculated as: The normalized benefit index of the data transmission benefit of this emergency task is calculated as follows:

[0072] Normalization of algorithm performance indicators:

[0073] Query all previous emergency task planning plans to obtain the maximum value MaxTime and the minimum value MinTime of the algorithm time consumption, the maximum value MaxDisturb and the minimum value MinDisturb of the disturbance;

[0074] The normalized index of the time requirement of this algorithm is: The normalized disturbance index of the entire system is calculated as:

[0075] Normalization of resource utilization indicators:

[0076] Query all previous emergency mission planning plans, obtain the maximum value MaxNSat and the minimum value MinNSat of the number of satellites, the maximum value MaxTOpen and the minimum value MinTOpen of the total power-on time, the maximum value MaxNSta and the minimum value MinNSta of the number of ground stations, and the maximum value MaxTSta and the minimum value MinTSta of the total data reception time;

[0077] The normalized index of the number of satellites NSat is: The normalized index of the total startup time Open is calculated as: The normalized index of the number of ground stations NSta is calculated as: The normalized indicator of the total receiving time TSta of the ground station is calculated as:

[0078] (4) Comprehensively consider the factors affecting the evaluation algorithm of each basic indicator, determine the weight of each basic indicator and the weight of each derived indicator; and weight the normalized basic indicators to calculate each derived indicator, and then weight each derived indicator to calculate the comprehensive indicator. Figure 4 As shown, the details are as follows:

[0079] (401) Determination of basic indicator weights: Based on the experience of multiple emergency missions, the weights of basic indicators are as follows: mission completion rate 0.4, mission completion benefit 0.4, data transmission benefit 0.2, time requirement 0.5, disturbance 0.5, number of satellites 0.2, total power-on time 0.5, number of ground stations 0.2, total reception time 0.1;

[0080] (402) Determination of derived indicator weights: Based on the experience of multiple emergency missions, the weights of the basic indicators are as follows: task completion indicator 0.7, algorithm performance indicator 0.15, resource utilization indicator 0.15;

[0081] (403) Task completion indicator calculation: According to the basic indicator weight specified in step (401), the task completion indicator calculation formula is: Completion = 0.4 × BaseCompletion + 0.4 × BaseWeight + 0.2 × BaseBenefit; the algorithm performance indicator calculation formula is: Performance = 0.5 × BaseTime + 0.5 × BaseDisturb; the resource utilization indicator calculation formula is: Res = 0.2 × BaseNSat + 0.5 × BaseTOpen + 0.2 × BaseNSta + 0.1 × BaseTSta;

[0082] (404) According to the derived indicator weights specified in step (402), the calculation formula for the evaluation value of this emergency task planning algorithm is: Evaluate = 0.7×Completion+0.15×Performance+0.15×Res.

Claims

1. A satellite remote sensing emergency mission planning algorithm evaluation method based on AHP hierarchical structure, characterized in that: The following steps are involved: (1) Construct an indicator evaluation system, including comprehensive indicators, derived indicators and basic indicators. Comprehensive indicators include multiple derived indicators, and each derived indicator includes multiple basic indicators. (2) Calculate the values ​​of each basic indicator in step (1) according to the satellite planning task and the algorithm planning results; (3) normalizing the calculation results of each basic index value in step (2); (4) Comprehensively consider the factors affecting the evaluation algorithm of each basic indicator, determine the weight of each basic indicator and the weight of each derived indicator; and weight the normalized basic indicators to calculate each derived indicator, and then weight each derived indicator to calculate the comprehensive indicator; Wherein, step (1) is specifically as follows: Construct the hierarchical structure of the indicator evaluation system, which includes three levels; the first level is the comprehensive indicator level, which is the overall evaluation of a certain planning algorithm; the second level is the derived indicator level, which is the evaluation of a specific aspect of the planning algorithm, including emergency task completion indicators, algorithm performance indicators and resource utilization indicators; the third level is the basic indicator level, the basic indicators of the emergency task completion indicators include task completion rate, task completion benefits and data downlink benefits, the basic indicators of the algorithm performance indicators include time requirements and disturbances to the entire system, and the basic indicators of the resource utilization indicators include the number of satellites, total power-on time, number of ground stations and total ground station reception time; Wherein, step (2) specifically includes the following process: 1) The basic indicators of emergency task completion status are calculated as follows: After the task planning is completed, the completion status of each observation emergency task is obtained from the planning results, and the task completion rate is calculated. The calculation formula is: Among them, Number represents the total number of emergency tasks to be observed, N c Indicates the total number of completed observation emergency tasks, Mission j represents the jth emergency task, Inv(x) is the emergency task completion function, if the emergency task x is photographed by the satellite, then Inv(x) = 1; otherwise, Inv(x) = 0; The task completion benefit index is the weighted sum of the different weighted observation emergency tasks completed. After the task planning is completed, the total weight of the emergency task completion is calculated according to the observation emergency task level. The formula is: w j is the weight of the jth emergency task, w j Larger values ​​indicate greater importance; The calculation formula for data downlink benefit is: DT j is the deadline of emergency task j, π j is the end time of the ground station downlink time window corresponding to emergency task j; 2) The basic indicators of the algorithm performance indicators are calculated as follows: The time requirement Time is the running time of the emergency task planning algorithm itself, which is used as the time requirement indicator Time; The disturbance index Disturb of the entire system is the disturbance of the original planning scheme Where i is the task in the original planning scheme Φ, the change of the task state in Φ is called disturbance, and its measurement is expressed as disturbance i , when task i in Φ does not change, disturbance i Take 0, when the execution time window of task i changes, disturbance i Take 1, when task i is deleted from the planning list, disturbance i Take 2; 3) The basic indicators of resource utilization indicators are calculated as follows: Number of satellites NSatellite: According to the planning results, the number of satellites under observation is counted as the satellite number indicator; Total power-on time TOpen: According to the planning results, the total power-on time of the satellites used for observation is counted as the total power-on time indicator; Number of ground stations NStation: According to the planning results, the number of ground stations receiving data is counted as the number of ground stations indicator; Total receiving time of ground stations DStation: According to the planning results, the total receiving time of the ground stations performing observations is counted as the total receiving time indicator.

2. The evaluation method for satellite remote sensing emergency mission planning algorithm based on AHP hierarchical structure according to claim 1 is characterized in that: Step (3) specifically includes the following process: Normalization of emergency task completion indicators: Query all previous emergency task planning plans, obtain the maximum value MaxCompletion and minimum value MinCompletion of the task completion rate, the maximum value MaxWeight and MinWeight of the task completion benefit, and the maximum value MaxBenefit and minimum value MinBenefit of the data transmission benefit; The normalized index of the completion rate of this emergency task is: The normalized index of the benefit Weight of this task completion is calculated as: The normalized benefit index of the data transmission benefit of this emergency task is calculated as follows: Normalization of algorithm performance indicators: Query all previous emergency task planning plans to obtain the maximum value MaxTime and the minimum value MinTime of the algorithm time consumption, the maximum value MaxDisturb and the minimum value MinDisturb of the disturbance; The normalized index of the time requirement of this algorithm is: The normalized disturbance index of the entire system is calculated as: Normalization of resource utilization indicators: Query all previous emergency mission planning plans, obtain the maximum value MaxNSat and the minimum value MinNSat of the number of satellites, the maximum value MaxTOpen and the minimum value MinTOpen of the total power-on time, the maximum value MaxNSta and the minimum value MinNSta of the number of ground stations, and the maximum value MaxTSta and the minimum value MinTSta of the total data reception time; The normalized index of the number of satellites NSat is: The normalized index of the total startup time Open is calculated as: The normalized index of the number of ground stations NSta is calculated as: The normalized indicator of the total receiving time TSta of the ground station is calculated as:

3. The evaluation method for satellite remote sensing emergency mission planning algorithm based on AHP hierarchical structure according to claim 1 is characterized in that: Step (4) specifically includes the following steps: (401) Based on the experience of multiple emergency missions, the weights of various basic indicators are specified, including the weights of mission completion rate, mission completion efficiency, data transmission benefits, time requirements, disturbance to the entire system, number of satellites, total power-on time, number of ground stations and total ground station reception time; (402) Based on the experience of multiple emergency missions, the weights of various derived indicators are specified, including the weights of the task completion indicator, the algorithm performance indicator, and the resource utilization indicator; (403) weighting each basic indicator value normalized in step (3) with the basic indicator weight specified in step (401) to obtain a task completion indicator, an algorithm performance indicator, and a resource utilization indicator; (404) The task completion status index, algorithm performance index and resource utilization index are weighted with the weights of each derived index specified in step (402) to obtain a comprehensive index. The emergency planning plan with a large comprehensive index value is better.

Images