Strain monitoring is an integral component of routine structural inspection; however, traditional electrical sensors are not viable options in extreme environments due to their operating properties. Fiber Bragg Grating (FBG) sensors have gained significant popularity in aerospace, bridges, and engineering monitoring due to their high sensitivity to deformation, corrosion resistance, anti-electromagnetic interference, and ease of networking, amongst other advantages. However, the majority of fiber grating strain sensors developed to date have been designed to enhance sensitivity, thereby improving measurement accuracy but not allowing for the required sensitivity when sensors with different sensitivities are required for measurement adjustment.In addressing this gap in the literature, the authors have designed a fiber grating strain sensor with adjustable sensitivity and adjustable negative strain range, and built an experimental platform to verify the performance of the sensor.In order to fabricate FBG sensors with adjustable sensitivity, we prepared FBG sensors using polyimide fibre optic grating and metal substrate, and built a strain experiment platform to test the sensor performance. Initially, the principle of adjustable sensitivity was analysed, and the structure of the metal substrate was designed in combination with the principle of strain concentration. The corresponding elastic structure for the principle of strain concentration was then designed, and a numerical simulation was performed to determine the optimal size for this elastic structure. The final processing size was selected after analysing the simulation results (Fig.2). Subsequently, the principle of adjustable sensitivity was employed to design and process holes of varying lengths on the metal substrate, with the sensitivity being adjusted by utilising different connecting holes (Fig.1).The negative range adjustable process was described in detail, and the tensile limit of the polyimide fibre grating used was obtained (Tab.1).Normal temperature strain calibration of the fibre optic grating strain sensors was carried out using a universal laboratory machine (Fig.5). Temperature calibration experiments were then conducted on the fibre optic grating strain sensors using a high and low temperature chamber (Fig.7).The use of a universal laboratory machine and a high temperature furnace was then employed to construct a composite experimental platform, which could calibrate the strain temperature dual-parameter measurement of the sensor (Fig.9). Subsequently, the data processing was performed on experimental results in order to objectively evaluate the accuracy of the performance of the sensor. Thermocouples and the tensile experimental machine derived data were used as the temperature standard and strain standard to analyse the accuracy of the sensor measurement data.A metal substrate FBG strain sensor based on an elastic element was designed, which has the capacity for temperature compensation. The performance of the sensor was verified using a universal testing machine, a high and low temperature chamber and a fibre optic grating demodulator. The sensor demonstrated good linearity and repeatability, which renders it easy to install and suitable for engineering structural deformation monitoring applications that require different ranges.This study outlines the performance of an FBG strain sensor encapsulated in a metal substrate, which has been demonstrated to achieve adjustable sensitivity and adjustable negative range. For L/LFBG of 1, the strain sensitivity is measured at 0.594 pm/µε, the temperature sensitivity at 24.42 pm/°C, the repeatability error at 0.75%, the hysteresis error at 1.377%, and the change in value of L/LFBG at 0.000 1. BG. The error in comparison with the theoretical value is not more than ±5%, and the negative range adjustment limit is at 12 nm. The temperature strain experimental platform was built using a universal experimental machine and a high-temperature furnace. The two-parameter measurement data exhibited good linearity, and the temperature decoupling eliminated the influence of the temperature, thus ensuring the linearity of the strain data.