Real-time High-precision Correction Method for On-orbit Self-radiation of Reflector Antenna of Microwave Remote Sensing Instrument

Abstract

The present invention provides a real-time high-precision correction method for the on-orbit self-radiation of a reflector antenna of a microwave remote sensing instrument, which relates to the field of aerospace technology. This method deeply explores the physical origin of the self-radiation (RAntSelf) of the reflector antenna, and uses a neural network to store the dynamic law of the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument as a correction device. Based on artificial intelligence technology, taking the physical conditions determining the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument as the input and the on-orbit self-radiation value of the reflector antenna of the microwave remote sensing instrument as the output, the device is trained to obtain its parameters. Then, when it is necessary to correct the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument, the steady-state environmental temperature field of the reflector antenna of the microwave remote sensing instrument is input into the device, so as to timely obtain the accurate on-orbit self-radiation magnitude of the reflector antenna for target radiation measurement correction.

Description

Technical Field

[0001] The present invention relates to the field of aerospace technology, and particularly to a real-time high-precision correction method for on-orbit self-radiation of a reflecting antenna of a microwave remote sensing instrument. Background Art

[0002] The structure of a spaceborne microwave remote sensing instrument usually can be divided into a reflecting antenna, a feed, a receiver, a post-processing circuit, etc. Generally, a quantitative microwave remote sensing instrument needs to carry an on-board blackbody for radiation calibration, that is, when observing the blackbody, the corresponding output count value (Digital Number, DN) is obtained, and then the radiation calibration coefficient is obtained by fitting the output brightness temperature of the blackbody and the output count value. The calculation formula of the two-point radiation calibration coefficient is:

[0003] (1-1)

[0004] (1-2)

[0005] Wherein, and are respectively the radiance output by the blackbody and the radiance of the cold sky (which can also be replaced by the output brightness temperature of the blackbody and the brightness temperature of the cold sky), and are respectively the count values when observing the blackbody and the cold sky, and the calibration coefficients and are obtained.

[0006] However, for a microwave remote sensing instrument, since the reflecting surface area is relatively large, the blackbody capable of covering the reflecting surface often needs to have a large aperture and a heavy weight. Limited by the volume and weight of the satellite and the power required to heat the large blackbody, the positions of many on-board blackbodies are arranged in front of the feed or a certain reflecting surface, that is, when observing the blackbody / cold sky, only the part from the feed or a certain reflecting surface until the output count value participates in the observation. In this way, the blackbody can be made relatively small to meet the limitations of the satellite on the volume, weight and heating power of the blackbody. Thus, the above calibration coefficients and are the calibration coefficients from the first radiation link component (feed or the first reflecting surface capable of observing the blackbody) observing the blackbody to the output count value. However, when a microwave remote sensing instrument observes a target, the input energy of the calibration part (that is, the energy that can act on the output ) includes both the energy from the observed target and the self-radiation energy of the reflecting surface. The expression formula is:

[0007] (2)

[0008] Wherein, and respectively represent the radiation energy from a target and the self-radiation energy from the antenna when observing a certain target. Since both of these two parts of energy are before the calibratable part of the microwave remote sensing instrument, thus these two parts of energy together make the calibratable part of the microwave remote sensing instrument satisfy the right side of the above formula (2). is the output count value when observing this target. In order to accurately obtain the self-radiation of the antenna , usually the cold sky calibration method is used to obtain , because the microwave background radiation energy of the cosmic space (cold sky) is determined. During cold sky calibration, the microwave remote sensing instrument is pointed at the cold sky for observation. At this time, only in formula (2) is the unknown, and thus the self-radiation of the reflector antenna of the microwave remote sensing instrument can be obtained . Then when observing the target again, since the self-radiation of the reflector antenna is known, the radiation energy from the target can be accurately known in magnitude.

[0009] The cold sky calibration method can obtain the self-radiation of the reflector antenna , however, a microwave remote sensing instrument cannot perform cold sky calibration all the time. If at a certain moment, after measuring until the next measurement (to distinguish the two measurements, here represents the obtained from the next measurement, to show the difference from the obtained from the previous measurement), if and have very little difference, then the measurement and using it until the next measurement will generate a very small error , and the error introduced to calculate the radiation energy of the target is also very small. However, when has a large difference compared with (that is, is very large), obviously the error introduced to calculate the radiation energy of the target is also very large, thus the radiation measurement accuracy of this microwave remote sensing instrument cannot be guaranteed. SUMMARY OF THE INVENTION

[0010] In view of this, in order to solve the problems existing in the prior art, the object of the present invention is to provide a real-time high-precision correction method for the on-orbit self-radiation of the reflecting antenna of a microwave remote sensing instrument. The method uses a neural network to store the change law of the on-orbit self-radiation of the reflecting antenna of the microwave remote sensing instrument as a correction device. Based on artificial intelligence technology, the physical conditions that determine the self-radiation value of the reflecting antenna (including the support structure that supports the reflecting antenna) are used as the input, and the self-radiation value of the reflecting antenna is used as the output. The device is trained to obtain its parameters. Then, when the self-radiation value of the reflecting antenna is required, the steady-state environmental temperature field of the reflecting antenna is input into the device, so as to obtain the accurate self-radiation value of the reflecting antenna in time for target radiation measurement correction.

[0011] The present invention deeply explores the physical origin of, and analyzes the self-radiation of the reflecting antenna the physical origin of. According to Planck's law, any object above absolute zero will emit radiation outward. Therefore, temperature is the fundamental origin; at the same time, if the reflecting surface system includes multiple specific reflecting surfaces, then its structure is also the cause of the change. However, the reason for the change in the structure of the multiple reflecting surfaces of the reflecting surface system is thermal expansion and contraction. Therefore, the temperature of the reflecting surface system can not only characterize the origin of how much is, but also characterize the influence of the structural change of the reflecting surface system on .

[0012] To achieve the above object and based on the analysis of the physical origin of the self-radiation of the reflecting antenna the present invention is implemented by the following technical solutions:

[0013] The present invention provides a real-time high-precision correction method for the on-orbit self-radiation of the reflecting antenna of a microwave remote sensing instrument. The method includes the following steps:

[0014] Step S1: Based on the cold sky calibration method, measure the self-radiation value of the reflecting antenna at a certain moment, and record the temperature of all engineering temperature measurement points on the reflecting antenna at this moment.

[0015] Step S2: Repeat the initial measurement process of Step S1 to obtain multiple groups of self-radiation values of the reflecting antenna and the temperature data of all engineering temperature measurement points on the corresponding reflecting antenna.

[0016] Step S3: Using the physical conditions that affect the self-radiation of the reflecting surface as the input and the self-radiation value of the reflecting antenna as the label, train the RSNN device that describes the change law of the self-radiation of the reflecting antenna to obtain the neural network weights. Among them, the physical conditions that affect the self-radiation of the reflecting surface are the temperatures of all engineering temperature measurement points on the reflecting antenna.

[0017] Step S4: By measuring the steady-state ambient temperature field of the reflector antenna, obtain the physical conditions affecting the self-radiation of the reflector surface at the observation target moment, and input the obtained physical conditions into the RSNN device with trained neural network weights to obtain the self-radiation value of the reflector antenna at the target moment for high-precision correction.

[0018] As a further aspect of the present invention, when measuring the self-radiation value of the reflector antenna at a certain moment, based on the calculation formula of the input energy of the calibration part of the microwave remote sensing instrument when observing the target, that is, formula (2), measure the self-radiation of the reflector antenna at a certain moment through the cold sky calibration of the microwave remote sensing instrument of the reflector antenna , where is a label at a certain moment.

[0019] As a further aspect of the present invention, the calculation formula of the input energy of the calibration part of the microwave remote sensing instrument when observing the target is:

[0020]

[0021] where and respectively represent the radiation energy from the target and the self-radiation energy from the antenna when observing a certain target , the calibration coefficients and are the calibration coefficients from the first radiation link component observing the blackbody to the output count value, obtained through cold sky and blackbody calibrations respectively, represents the radiation energy from the target when observing a certain target , is the output count value when observing this target.

[0022] As a further aspect of the present invention, record the temperatures of all engineering temperature measurement points on the reflector antenna at this moment, denoted as , representing the temperatures of all engineering temperature measurement points on the reflector antenna at the moment.

[0023] As a further aspect of the present invention, in step S2, obtain groups of self-radiation of the reflector antenna ; the corresponding temperatures of all engineering temperature measurement points on the reflector antenna are , where the temperatures of all engineering temperature measurement points on the reflector antenna are the physical conditions affecting the self-radiation of the reflector antenna of the microwave remote sensing instrument.

[0024] As a further aspect of the present invention, the training steps of the RSNN device in step S3 are:

[0025] ​(1) Training is carried out with the physical conditions affecting the self-radiation of the reflecting surface as the input and the self-radiation value of the reflecting antenna as the label;

[0026] (2) During training, all inputs and labels are normalized;

[0027] (3) A fully connected neural network containing an input layer, at least one hidden layer, and an output layer is constructed to form an RSNN;

[0028] (4) The defined RSNN is trained with the normalized inputs and labels.

[0029] As a further solution of the present invention, before the real-time high-precision correction of the on-orbit self-radiation of the reflecting antenna of the microwave remote sensing instrument, the self-radiation of the reflecting antenna at any moment is obtained between two actual measurements of the self-radiation of the reflecting antenna. and

[0030] As a further solution of the present invention, in step S4, the physical conditions affecting the self-radiation of the reflecting surface at the target moment obtained by measuring the steady-state environmental temperature field of the reflecting antenna are the temperatures of all engineering temperature measurement points.

[0031] Compared with the prior art, a method for real-time high-precision correction of the on-orbit self-radiation of the reflecting antenna of a microwave remote sensing instrument provided by the present invention realizes the real-time high-precision correction of the on-orbit self-radiation of the reflecting antenna of the microwave remote sensing instrument, and has the following beneficial effects:

[0032] 1. High-precision correction is realized. The self-radiation value of the reflecting antenna is accurately calculated and corrected through the neural network model, effectively improving the accuracy of microwave remote sensing data; since the traditional method cannot perform cold sky calibration frequently, errors are generated in the self-radiation value over time, while the method for real-time high-precision correction of the on-orbit self-radiation of the reflecting antenna of the microwave remote sensing instrument of the present invention can be updated and corrected in real time to ensure high-precision data.

[0033] 2. Real-time and automated correction is realized. By adopting artificial intelligence technology, the self-radiation value of the reflecting antenna can be corrected in real time on orbit without manual intervention, greatly improving the automation level of operation. This real-time nature is especially important for remote sensing satellites in orbit, enabling timely response to environmental changes and maintaining high-quality observation data.

[0034] 3. The calibration frequency is reduced. Since the traditional calibration method needs to perform cold sky calibration frequently to maintain the measurement accuracy, while the method for real-time high-precision correction of the on-orbit self-radiation of the reflecting antenna of the microwave remote sensing instrument of the present invention reduces the dependence on frequent calibration through the continuous learning and adjustment of the neural network model, reducing the consumption of satellite resources and operating costs.

[0035] 4. Improved reliability. By using a neural network to model the self-radiation variation law of the reflector antenna, even in complex environments (such as temperature changes, component aging, etc.), high-precision radiation value calculations can be ensured, improving the reliability and stability of the remote sensing instrument. It is not only applicable to specific microwave remote sensing instruments but can also be extended to other similar remote sensing devices. By simply training the neural network accordingly, it can adapt to the self-radiation correction requirements of different devices, with broad adaptability and application prospects.

[0036] 5. Cost and resource savings. Compared with the traditional large-aperture blackbody calibration device, which is costly and takes up valuable satellite space and weight, the real-time high-precision correction method for the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument of the present invention achieves high-precision correction, avoids the use of the blackbody device, saves costs and resources, and optimizes the design and operation of the satellite. Through high-precision self-radiation correction, the quality of remote sensing data has been significantly improved, providing more reliable data support for fields such as scientific research, environmental monitoring, and meteorological forecasting, and promoting the development of related applications.

[0037] In summary, the real-time high-precision correction method for the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument of the present invention has significant beneficial effects in improving the accuracy, real-time performance, reliability, and adaptability of remote sensing data, and can greatly promote the development and application of microwave remote sensing technology.

[0038] These aspects or other aspects of the present invention will be more clearly understood in the following description of the embodiments. It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or in related technologies, the following briefly introduces the drawings required for describing the exemplary embodiments or related technologies. The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0040] Figure 1 is a flowchart of the real-time high-precision correction method for the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument according to the embodiment of the present invention.

[0041] Figure 2 is a flowchart of the training of the RSNN device in the real-time high-precision correction method for the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0043] In some processes described in the specification and claims of the present invention and the above-mentioned drawings, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0044] Next, the technical solutions in the exemplary embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings in the exemplary embodiments of the present invention. Obviously, the described exemplary embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0045] Glossary of English Abbreviations:

[0046] RSNN: Reflective Surface Neural Network; Chinese translation: Reflective Surface Neural Network.

[0047] Reflective Surface (RS): Reflective surface, which is a reflective component in microwave remote sensing instruments, such as antennas and other surfaces that can reflect electromagnetic waves. These reflective surfaces play a crucial role in the operation of remote sensing instruments.

[0048] Neural Network (NN): Neural network, which is widely used in the fields of machine learning and artificial intelligence for pattern recognition, decision-making and complex calculations. The neural network is trained to understand and correct the radiation of the reflective surface itself.

[0049] In the application of the present invention in remote sensing, RSNN aims to improve the accuracy of microwave remote sensing data in the following ways:

[0050] 1. Real-time correction: Continuously monitor and real-time correct the radiation value of the reflective surface itself to ensure high-precision data acquisition.

[0051] 2. Improve data quality: By reducing the error introduced by the self-radiation of the reflecting surface, the overall quality of the data is improved, which is crucial for applications such as environmental monitoring, weather forecasting, and scientific research.

[0052] 3. Reduce calibration frequency: Traditional methods require frequent calibration to correct self-radiation. RSNN reduces the need for such frequent calibration, saving time and resources.

[0053] In the application of remote sensing of the present invention, the advantages of the RSNN device include: high precision: providing accurate self-radiation correction, thus making the data more reliable; automation: operating automatically without manual intervention; cost-effectiveness: reducing dependence on expensive calibration equipment and procedures; versatility: being able to adapt to different types of remote sensing instruments with reflecting surfaces.

[0054] In summary, the RSNN (Reflecting Surface Neural Network) device significantly improves the accuracy and reliability of microwave remote sensing by correcting the influence of the self-radiation of the reflecting surface and can perform real-time correction.

[0055] The following further illustrates the technical solution of the present invention with specific embodiments:

[0056] Refer to Figure 1 As shown, the embodiment of the present invention provides a real-time high-precision correction method for the on-orbit self-radiation of a reflecting antenna of a microwave remote sensing instrument, including the following steps:

[0057] Step S1: Based on the cold sky calibration method, measure the self-radiation value of the reflecting antenna at a certain moment and record the temperature of all engineering temperature measurement points on the reflecting antenna at this moment.

[0058] In this step, the calculation formula for the energy input to the calibration part of the microwave remote sensing instrument when observing a target is:

[0059]

[0060] Wherein, and respectively represent the radiation energy from the target and the self-radiation energy from the antenna when observing a certain target . The calibration coefficients and are the calibration coefficients from the first radiation link component observing the blackbody to the output count value, and are obtained through cold sky and blackbody calibration respectively. represents the radiation energy from the target when observing a certain target , is the output count value when observing this target.

[0061] When measuring the self-radiation value of the reflecting antenna at a certain moment, based on the calculation formula of the energy input in the calibration part of the microwave remote sensing instrument when observing the target, measure at a certain moment through the cold sky calibration of the microwave remote sensing instrument of the self-radiation of the reflecting antenna , where is the label at a certain moment.

[0062] Among them, the temperature of all engineering temperature measurement points on the reflecting antenna at this moment is recorded as , indicating at the moment, all on the reflecting antenna temperatures of the engineering temperature measurement points.

[0063] Step S2: Repeat the initial measurement process of Step S1 to obtain multiple sets of self-radiation values of the reflecting antenna and the temperature data of all engineering temperature measurement points on the corresponding reflecting antenna.

[0064] In this step, obtain sets of self-radiation of the reflecting antenna ; the temperatures of all engineering temperature measurement points on the corresponding reflecting antenna are , among which, the temperatures of all engineering temperature measurement points on the reflecting antenna are the physical conditions affecting the self-radiation of the reflecting antenna of the microwave remote sensing instrument.

[0065] Step S3: Use the physical conditions affecting the self-radiation of the reflecting surface as the input and the self-radiation value of the reflecting antenna as the label to train the RSNN device that describes the change law of the self-radiation of the reflecting antenna, and obtain the neural network weights, where the physical conditions affecting the self-radiation of the reflecting surface are the temperatures of all engineering temperature measurement points on the reflecting antenna.

[0066] In this step, as shown in Figure 2 , the training steps of the RSNN device are:

[0067] Step S301: Use the physical conditions affecting the self-radiation of the reflecting surface as the input and the self-radiation value of the reflecting antenna as the label for training;

[0068] Step S302: During training, perform normalization processing on all inputs and labels;

[0069] Step S303: Construct a fully connected type neural network containing an input layer, at least one hidden layer and an output layer to form an RSNN;

[0070] Step S304: Use the normalized inputs and labels to train the defined RSNN.

[0071] In this step, use the physical conditions affecting the self-radiation of the reflecting antenna of the microwave remote sensing instrument as the input and the self-radiation value of the reflecting antenna The RSNN device describing the changing law of the self-radiation of the reflective antenna is trained as a label to obtain the neural network weights.

[0072] Step S4, by measuring the steady-state ambient temperature field of the reflecting antenna, the physical conditions that affect the self-radiation of the reflecting surface at the target time of observation are obtained, and the obtained physical conditions are input into the RSNN device with trained neural network weights to obtain the self-radiation value of the reflecting antenna at the target time for high-precision correction.

[0073] In this step, by measuring the steady-state ambient temperature field of the reflective antenna, the physical conditions affecting the self-radiation of the reflective surface at the target time are obtained as follows: The temperature of each engineering temperature measurement point.

[0074] By measuring the reflector antenna (including the supporting structure), we can know the physical conditions that affect the self-radiation of the reflector at a specific time (i.e. all The temperature of each engineering temperature measurement point) is then input into the RSNN device with known neural network weights to obtain the accurate self-radiation value of the reflector antenna at a specific moment, which is used in the subsequent self-radiation correction of the reflector antenna or other data processing. The temperature of each engineering temperature measurement point can be obtained continuously, so the self-radiation of the reflector antenna can be measured twice. and The self-radiation of the reflecting antenna at any moment can be obtained, and then the on-orbit self-radiation of the reflecting antenna of the microwave remote sensing instrument can be corrected in real time with high precision.

[0075] The embodiment of the present invention provides a real-time high-precision correction method for the on-orbit self-radiation of a microwave remote sensing instrument reflector antenna. A neural network is used to store the dynamic law of on-orbit self-radiation of a microwave remote sensing instrument reflector antenna as a correction device. Based on artificial intelligence technology, the physical conditions that determine the on-orbit self-radiation of the microwave remote sensing instrument reflector antenna are used as input, and the on-orbit self-radiation value of the microwave remote sensing instrument reflector antenna is used as output. The device is trained to obtain its parameters. Then, when it is necessary to correct the on-orbit self-radiation of the microwave remote sensing instrument reflector antenna, the steady-state ambient temperature field of the microwave remote sensing instrument reflector antenna is input into the device, thereby timely obtaining the precise on-orbit self-radiation size of the reflector antenna for target radiation measurement correction.

[0076] The real-time high-precision correction method for the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument of the present invention analyzes the self-radiation of the reflector antenna The physical origin of the temperature is that according to Planck's law, any object above absolute zero will emit radiation. At the same time, if the reflector system contains multiple specific reflectors, then its structure is also the cause. The reason for the change. However, the reason for the change in the multiple reflector structures of the reflector system is thermal expansion and contraction. Therefore, the temperature of the reflector system can characterize both how much is the origin of and can also characterize the influence of the structural change of the reflector system on

[0077] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, reference can be made to the description in the method embodiment section.

[0078] Those skilled in the art should also be able to further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0079] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A real-time high-precision correction method for on-orbit self-radiation of a reflecting antenna of a microwave remote sensing instrument, characterized in that The method comprises the following steps: Step S1: Based on the cold sky calibration method, measure the self-radiation value of the reflecting antenna at a certain moment, and record the temperatures of all engineering temperature measurement points on the reflecting antenna at this moment; Step S2: Repeat the initial measurement process of Step S1 to obtain multiple sets of self-radiation values of the reflecting antenna and the temperature data of all engineering temperature measurement points on the corresponding reflecting antenna; Step S3: Use the physical conditions affecting the self-radiation of the reflector as the input and the self-radiation value of the reflecting antenna as the label to train the RSNN device that describes the variation law of the self-radiation of the reflecting antenna to obtain the neural network weights. Among them, the physical conditions affecting the self-radiation of the reflector are the temperatures of all engineering temperature measurement points on the reflecting antenna; Step S4: By measuring the steady-state environmental temperature field of the reflecting antenna, obtain the physical conditions affecting the self-radiation of the reflector at the observed target moment, and input the obtained physical conditions into the RSNN device with the trained neural network weights to obtain the self-radiation value of the reflecting antenna at the target moment for high-precision correction.

2. The real-time high-precision correction method for the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument according to claim 1, characterized in that, When measuring the self-radiation value of the reflecting antenna at a certain moment, based on the calculation formula of the energy input in the calibration part when the microwave remote sensing instrument observes the target, the self-radiation of the reflecting antenna at a certain moment is measured through the cold sky calibration of the microwave remote sensing instrument of the reflecting antenna , where is the mark at a certain moment.

3. The real-time high-precision correction method for the on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument according to claim 2, wherein The calculation formula for the input energy of the calibration part of the microwave remote sensing instrument when observing the target is: ， Among them, and respectively represent the radiation energy from the target and the self-radiation energy from the antenna when observing a certain target . The calibration coefficients and are the calibration coefficients from the first radiation link component observing the blackbody to the output count value, and are obtained through cold sky and blackbody calibrations respectively represents the radiation energy from the target when observing a certain target , and is the output count value when observing this target 4. The real-time high-precision correction method for the on-orbit self-radiation of the reflecting antenna of the microwave remote sensing instrument according to claim 3, characterized in that, The temperature of all engineering temperature measurement points on the reflecting antenna at this recorded moment is denoted as , representing the temperature of all engineering temperature measurement points on the reflecting antenna at the moment.

5. The real-time high-precision correction method for on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument according to claim 4, characterized in that, In step S2, obtain self-radiation of the set of reflector antennas ; the temperatures of all engineering temperature measurement points on the corresponding reflector antennas are , where the temperatures of all engineering temperature measurement points on the reflector antennas are physical conditions affecting the self-radiation of the reflector antennas of microwave remote sensing instruments.

6. The real-time high-precision correction method for on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument according to claim 5, characterized in that, The training steps of the RSNN device in Step S3 are: Use the physical conditions affecting the self-radiation of the reflector as the input and the self-radiation value of the reflecting antenna as the label for training; During training, normalize all inputs and labels; Construct a fully connected neural network containing an input layer, at least one hidden layer, and an output layer to form an RSNN; Use the normalized inputs and labels to train the defined RSNN.

7. The real-time high-precision correction method for on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument according to claim 5, characterized in that, Before the real-time high-precision correction of the on-orbit self-radiation of the reflector antenna of a microwave remote sensing instrument, the self-radiation of the reflector antenna at any moment is obtained between two actual measurements of the self-radiation of the reflector antenna and ​ 8. The real-time high-precision correction method for on-orbit self-radiation of the reflector antenna of the microwave remote sensing instrument according to claim 4, characterized in that, In step S4, by measuring the steady-state ambient temperature field of the reflector antenna, the physical conditions affecting the self-radiation of the reflector surface at the target moment are all the temperatures of the engineering temperature measurement points.

Images