Method for realizing radar fault early warning based on massive time domain data and LSTM (Long Short Term Memory)

• Main Authors: XIANG CHENCHEN, ZHUAN JIAN, LI SHUAISHUAI, ZHANG YUE, WANG BENGE, JIN PENG, MA QING, YU HAO, MIZU TAKATOSHI, XU XIAOFENG, WANG XIAOYAN, GAO YONGGANG, DI CHENXI, SONG MIANMIAN, GUO LINHUI, YU ZHONGLIANG, SUN SHILONG, HU BIN
• Format: Patent
• Language: Chinese; English
• Published: 27.06.2023
• Subjects: ANALOGOUS ARRANGEMENTS USING OTHER WAVES; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; ELECTRIC DIGITAL DATA PROCESSING; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION ORRERADIATION OF RADIO WAVES; MEASURING; PHYSICS; RADIO DIRECTION-FINDING; RADIO NAVIGATION; TESTING

The invention discloses a radar fault early warning method based on massive time domain data and LSTM, and the method comprises the steps: obtaining the vibration sensing data of a radar turntable based on a vibration sensor, storing the original time domain data through a big data related technology, employing the historical time domain data in a normal state as a sample, calculating the T-square statistical magnitude of Hotelling, and carrying out the early warning of a radar fault. The method comprises the following steps: acquiring time domain data of a radar turntable, determining a threshold value of a Hotelling T-party statistical magnitude in a normal state of the radar turntable, training a long short-term memory network model, hereinafter referred to as an LSTM model, for historical time domain data, feeding the trained LSTM model with the time domain data acquired in real time, predicting time domain data of a next time unit, and calculating the Hotelling T-party statistical magnitude predicted by