为了同时确定监测结构的位移大小和方向，提出了一种基于光纤光栅组合的大量程、结构简单的机械传感装置。该传感装置以U型不锈钢结构组合光纤光栅为核心传感单元，以杠杆杆件来传递位移，测点位移大小和方向可以通过杠杆杆件两侧的U型不锈钢结构上的光纤光栅波长漂移计算得出。开展了传感装置6组辅助结构（核心传感单元）的标定试验，结果表明所有传感单元均具有较高的线性度，测量范围为0~140 mm，灵敏度为4.362 pm/mm，迟滞性误差为3.25%，重复性误差为6.62%。将传感装置用于室内堆积体边坡模型试验中，研究它们在土体连续滑动变形过程中的监测性能，并与粒子图像测速技术测量的土体位移变化进行比较，两种技术数据上呈现出较好的一致性、吻合度较高，表明所研发的光纤光栅矢量位移测量传感装置能够准确捕获坡面矢量变形，在智能监控领域有广阔的应用前景。

In practical applications such as slope instability deformation, random cracking of concrete and collapse of wind turbines, position tracking always requires two-dimensional sensing. The fiber-optic displacement sensors have been widespread applied in civil engineering field due to their intrinsic advantages, including electromagnetic interference immunity, miniature size, electrically-passive operation, and multiplexing capability, however, they are not able to retrieve the displacement direction and amplitude simultaneously. In view of this reason, a vector displacement measurement sensing device based on Fiber Bragg Grating (FBG) with large range and simple structure is proposed to identify the displacement magnitude and direction of the monitored structure simultaneously. The sensing device is mainly constructed of four FBGs, a base, an upper free rotation rod, springs 1 and 2, self-made U-shaped structures 1 and 2. The pre-tensioned FBGs are respectively pasted on the inner and outer sides of center position of the U-shaped structure as the sensing unit. When the deformation occurs, the screw plays a connecting role, and the spring is not affected by the screw. The bottom spring of U-shaped structure 2 is connected with the monitored point, and the movement of the monitored point will cause axial tension of U-shaped structure 2. The force will be applied to the spring and the U-shaped structure. The internal and external sides of the U-shaped structure are subject to tension and compression, respectively. Additionally, the movement of U-shaped structure 2 causes the upper rod to rotate around the center point, which further makes the spring of U-shaped structure 1 elongate and produces tension on the U-shaped structure. FBGs are bonded to both the upper and lower surface of the stainless steel plate of the U-shaped structure for temperature compensation. Even if temperature change occurs in an FBG sensor unit, strain in the upper and lower two FBGs bonded to both surfaces of the stainless steel plate are equal, so thermal strain can be neglected. The sensing principle of determining the displacement direction and amplitude simultaneously is introduced, and its expression is also derived. Calibration experiments of six sets of auxiliary structures (i.e., core sensing elements) were conducted. The experimental results showed that the sensing element is characterized by a superb linearity, a measurement range of 0~140 mm, a sensitivity of 4.362 pm/mm, a hysteresis error of 3.25%, and a repeatability error of 6.62%, respectively. Additionally, an indoor accumulation slope model test was performed to verify the performance of the FBG displacement sensing device in monitoring the continuous sliding deformation process of soil. The displacement values calculated by FBG-based sensor is basically consistent with that measured by Particle Image Velocimetry (PIV) technology, with an average relative error of 5.63%. The maximum relative error of horizontal displacement is 10.83%, the minimum value is 0.11%, the maximum relative error of vertical displacement is 11.17%, the minimum value is 0.67%, which can meet the measurement requirement of the sensor in slope monitoring. The errors of some measuring points exceeding 10% may be due to the fact that the probe of FBG displacement sensing device was buried shallowly in the soil and not fully in accordance with the soil deformation. Meanwhile, the relative error of displacement azimuths calculated by two technologies is basically within 10%, with a maximum error of 10.31% and a minimum error of 0.09%, which is basically the same as the error analysis of above displacement. The error analysis of displacement azimuth calculated by two techniques also proves the reliability of the FBG displacement sensing device developed in this paper for monitoring vector deformation of soil slopes. This capability demonstrates its broad application prospects in the field of intelligent monitoring.