RESUMEN
An accurate calculation of the indicator diagram of a pumping unit is the key factor in analyzing the performance of an oilfield production and operation and in preparing and optimizing an oilfield development plan. Aiming at the problems of the poor stability of the conventional load-displacement sensor method and the wave equation method, owing to the influence of an alternating load on the force sensor and the difficulty in measuring the crank angle using the electrical parameter method, a new soft sensing method employing the input electrical parameters of the motor and the beam inclination has been proposed to obtain the indicator diagram. At first, this method is established based on the beam angle of the pumping unit, which is easily measured using the suspension point displacement mathematics calculation model and the torque factor. Subsequently, the electric motor input parameters, the parameters of the four-bar linkage, and the relationship between the polished rod load have been established. Finally, the motor and the beam angle of the measured electrical parameters have been substituted into the calculation of the suspension point displacement and load value and pull in accordance with the guidelines to eliminate the singularity mutation values. After processing the measured data through a Butterworth filter, the indicator diagram is obtained. The results of the engineering experiment and application show that the average relative error of the method is less than 3.95%, and the maximum relative error remains within 2% for 6 months, which verifies the stability of the soft sensing method.
RESUMEN
The text similarity calculation plays a crucial role as the core work of artificial intelligence commercial applications such as traditional Chinese medicine (TCM) auxiliary diagnosis, intelligent question and answer, and prescription recommendation. However, TCM texts have problems such as short sentence expression, inaccurate word segmentation, strong semantic relevance, high feature dimension, and sparseness. This study comprehensively considers the temporal information of sentence context and proposes a TCM text similarity calculation model based on the bidirectional temporal Siamese network (BTSN). We used the enhanced representation through knowledge integration (ERNIE) pretrained language model to train character vectors instead of word vectors and solved the problem of inaccurate word segmentation in TCM. In the Siamese network, the traditional fully connected neural network was replaced by a deep bidirectional long short-term memory (BLSTM) to capture the contextual semantics of the current word information. The improved similarity BLSTM was used to map the sentence that is to be tested into two sets of low-dimensional numerical vectors. Then, we performed similarity calculation training. Experiments on the two datasets of financial and TCM show that the performance of the BTSN model in this study was better than that of other similarity calculation models. When the number of layers of the BLSTM reached 6 layers, the accuracy of the model was the highest. This verifies that the text similarity calculation model proposed in this study has high engineering value.