A novel high precision weighted integral ratio algorithm for fluorescent optical fiber temperature sensor

In the field of medical thermotherapy, heating cancer cells to 42.5–43 °C by microwave can make cancer cells inactive. Beyond this temperature range, normal cells are also damaged, which makes the accuracy of temperature detection very important. The active temperature sensor is seriously disturbed by strong electromagnetic field, which is difficult to meet the temperature measurement requirements in the field of microwave hyperthermia. Therefore, the fluorescent optical fiber temperature sensor has been applied in this field due to its insulation advantages. In order to improve the accuracy of the fluorescent optical fiber temperature sensor, a new weighted integral ratio algorithm is implemented in this paper. The numerical simulation show that the algorithm is completely unaffected by the DC bias in the system. In addition, the influence of random noise is greatly suppressed by integral processing and weighting processing. In this paper, the high-precision fluorescent optical fiber temperature sensor is realized by establishing the optical path, circuit and data processing algorithm of the system. The system is tested at 16 to 100 °C. The test results show that compared with the two-point method and the integral ratio method, the standard deviation of the weighted integral ratio algorithm is reduced by 67% and 50%, respective. The standard deviation of this algorithm is about 0.1°C. In addition, the response time of the algorithm is reduced by about 4 times. In summary, this algorithm can solve the problems of low accuracy, poor stability and long response time caused by data processing algorithms in the system.