#### Table of Content

Engineering—Special Topic on Artificial Intelligence Application
Review on application of artificial intelligence in power system restoration
Yutian LIU, Runjia SUN, Hongtao WANG, Xueping GU
Journal of Shandong University(Engineering Science). 2019, 49(5):  1-8.  doi:10.6040/j.issn.1672-3961.0.2019.122
Abstract ( 796 )   HTML ( 54 )   PDF (1128KB) ( 859 )
References | Related Articles | Metrics

The research results of expert system, fuzzy mathematics, evolutionary algorithm and machine learning in power system restoration were summarized. It was pointed out that the existing researches were mainly offline restoration method and the researches about online decision-making were in the initial stage. In addition, the application potential of artificial intelligence technology in system restoration was prospected.

Optimization method for parallel coordination restoration strategy of asynchronous power grid
Xueping GU, Chao YANG, Haiping LIANG, Yuanbo WANG, Shaoyan LI
Journal of Shandong University(Engineering Science). 2019, 49(5):  9-16.  doi:10.6040/j.issn.1672-3961.0.2019.096
Abstract ( 607 )   HTML ( 16 )   PDF (1429KB) ( 185 )
Figures and Tables | References | Related Articles | Metrics

Aiming at asynchronous power grid restoration after a blackout, an optimization method of parallel coordinated restoration strategy was proposed. This method adopted the thinking of serial and parallel restoration, used the high voltage direct current (HVDC) to coordinate sending end and receiving end power grids, and finally restored whole power system. Based on determining each AC subnet as a sending role or receiving role, the source characteristic of converter station in sending end gird and the load characteristic of converter station in receiving end gird were analysed. An optimization model for parallel coordinated restoration of asynchronous power grid was established, and the coordinated restoration process of HVDC, sending end and receiving end power grids were emphatically studied. The multi-population genetic algorithm (MPGA) was used to solve this model, then the optimal restoration scheme and the optimal starting time of HVDC were obtained. An asynchronous power grid example which was constructed based on the IEEE 39 bus system was given to verify the proposed method. The results showed that this optimization method was feasible, and it was suitable for the formulation of asynchronous power grid restoration scheme.

Transmission network reconfiguration strategy based on preference multiobjective optimization and genetic algorithm
Runjia SUN,Hainan ZHU,Yutian LIU
Journal of Shandong University(Engineering Science). 2019, 49(5):  17-23.  doi:10.6040/j.issn.1672-3961.0.2019.123
Abstract ( 537 )   HTML ( 4 )   PDF (4095KB) ( 319 )
Figures and Tables | References | Related Articles | Metrics

This paper proposed a transmission network reconfiguration strategy based on preference multiobjective optimization and genetic algorithm, which incorporated the preferences for different objectives to obtain network reconfiguration schemes. Considering the factors about generators, transmission lines and loads, three objectives were proposed to establish a preference multiobjective optimization model. Considering the preference and discreteness of the model, a preference-based nondominated sorting genetic algorithm Ⅱ was designed. To improve the solve efficiency, the preference-based dominance relation, population sacle control technique and repetitive individual filtration technique were proposed to obtain a preference Pareto solution set with a controllable number of solutions. The simulation results demonstrated that the strategy could reasonably leverage the tradeoff among different factors about network reconfiguration, and the proposed algorithm was highly efficient in solving network reconfiguration optimization problems.

Prediction method of wind power and PV ramp event based on deep learning
Zhixiang LIANG,Xiaoming LIU,Ying MU,Yutian LIU
Journal of Shandong University(Engineering Science). 2019, 49(5):  24-28.  doi:10.6040/j.issn.1672-3961.0.2019.132
Abstract ( 828 )   HTML ( 17 )   PDF (1178KB) ( 382 )
Figures and Tables | References | Related Articles | Metrics

With the gradual increase of the renewable energy penetration rate, the ramp event that caused the unbalanced active power occured sometimes, and even a large load loss. Due to the insufficient accuracy of wind power and photovoltaic prediction, there were many operational scenarios to be considered. The time domain simulation could not meet the online assessment requirements. A method based on deep learning was proposed in this paper. Considering the generation unit and tie line adjustment ability, the stacked denoising autoencoder was used to extract each layer feature to train support vector machine. The wind power, photovoltaic and load forecast data, and the power of the tie line at the previous moment were taken as inputs, and whether the ramp event occured as an output. The vector machine was used to quickly predict whether a ramp event occured. The simulation results of practical power grid showed that the proposed method was fast and accurate. It could effectively identify ramp events.

None-consumption users filtering algorithm based on decision tree and data-driven methods
Bo ZHANG,Feng LU,Hanyu DONG,Qingtai CHEN,Zhenzhi LIN,Hongtao WANG
Journal of Shandong University(Engineering Science). 2019, 49(5):  29-36.  doi:10.6040/j.issn.1672-3961.0.2019.116
Abstract ( 611 )   HTML ( 4 )   PDF (2084KB) ( 100 )
Figures and Tables | References | Related Articles | Metrics

With the increasing of power consumers and diversification of power consumption in power systems, the number of none-consumption users (NCUs) was also increasing rapidly. Thus, lots of manpower and material resources of power supply companies were arranged to perform troubleshooting on NCUs. Given this background, a data-driven method based on the electricity information of NCUs collected by electricity information acquisition system (EIAS) was proposed to determine the filtering results of normal NCUs and abnormal NCUs. The decision tree was utilized to analyze the electricity data of NCUs, and determine the types of NCUs. The key factors suitable for NCUs filtering were determined based on the original data to filter the NCUs that could not be screened by the decision tree, and the evaluation system for NCUs filtering was constructed. On this basis, CRITIC and radar chart methods were adopted to determine the weights of the key factors and to determine the filtering results of NCUs, respectively. The NCUs power-supplied by an actual power supply station in Zhejiang Province were served for demonstrating the proposed algorithm of NCUs filtering, and the simulation and on-site inspection results showed that the proposed data-driven method was effective for screening out the abnormal NCUs.

Bi-level optimal configuration of energy storage system in an active distribution network
Liyan WANG,Fei WANG,Yongji CAO,Tao ZHANG,Yaxin ZHANG,Yi LU,Zihan LIU
Journal of Shandong University(Engineering Science). 2019, 49(5):  37-43, 51.  doi:10.6040/j.issn.1672-3961.0.2019.082
Abstract ( 508 )   HTML ( 2 )   PDF (3422KB) ( 406 )
Figures and Tables | References | Related Articles | Metrics

In order to optimize the configuration of energy storage system in an active distribution network, a bi-level optimization method was proposed, considering the impact of operation strategy on planning scheme. In the short-scale inner optimization, the high-frequency components of integration power was extracted by the low-pass filtering algorithm. And a multi-objective optimization model was constructed to minimize the variation coefficient of extracted high-frequency component and the rate of the loss of renewable energy, which was simplified into a scalar optimization problem and solved by the particle swarm optimization algorithm. In the long-scale outer optimization, a multi-objective optimization model was established to minimize the investment cost and the rate of the loss of renewable energy, of which the Pareto optimal solutions were searched by the NSGA-Ⅱ. The location and capacity of energy storage system and the rate of the loss of renewable energy were taken as coupling variables, based on which the inner and outer models with different time scales were solved in a united optimization frame. The case study validated the effectiveness of the proposed model and corresponding solving methods, of which the results indicate that the optimal configuration of energy storage system in an active distribution network could enhance the accommodation ability of renewable energy.

Optimal complementary photovoltaic capacity configuration for grid-connected wind farms expansion
Dong YANG,Shiwen WANG,Yong WANG,Bo CHEN,Tianru ZHENG,Ning ZHOU,Tian XIAO,Yawen ZHAO
Journal of Shandong University(Engineering Science). 2019, 49(5):  44-51.  doi:10.6040/j.issn.1672-3961.0.2019.162
Abstract ( 570 )   HTML ( 4 )   PDF (3118KB) ( 141 )
Figures and Tables | References | Related Articles | Metrics

According to the complementarity of wind and solar energy sources, expanding photovoltaic panels in wind farms into wind-PV hybrid generation systems was helpful to smooth power fluctuation and improve operation economy. An approach to solve the optimization of the PV capacity for grid-connected wind farm was presented. Based on the complementarity of wind speed and solar radiation in time scales, modified meteorological probability models were established. A multi-objective optimization model was proposed with three objectives: maximizing the utilization of electrical equipment, minimizing the power fluctuation and the loss of renewable energy. The contradiction of three objectives and influence of the step-up transformer were incorporated. Meteorological data were simulated based on the Monte Carlo method. And the multi-objective particle swarm optimization was used to search the Pareto optimal solution set, from which an ultimate planning scheme was selected considering the engineering requirements and economic index. A numerical example was provided to validate the effectiveness of proposed approach.

Energy and Power Engineering—Special Topic on Refrigeration Technology
TRCC series system based on LNG cold energy and fuel cellwaste heat utilization
Yinglun GUO,Fuqiang XI,Ruizhi SU,Guoxiang LI,Zeting YU
Journal of Shandong University(Engineering Science). 2019, 49(5):  52-57.  doi:10.6040/j.issn.1672-3961.0.2019.154
Abstract ( 566 )   HTML ( 5 )   PDF (1261KB) ( 267 )
Figures and Tables | References | Related Articles | Metrics

A cogeneration system based on solid oxide fuel cell (SOFC for short) and transcritical carbon dioxide cycle (TRCC for short) was proposed. The transcritial carbon dioxide cycle was used to recover the exhaust heat of the SOFC while utilizing the LNG refrigeration capacity. The mathematical model of the system was established, and the influence of parameter changes on system performance was analyzed. The results showed that under the design conditions, the thermal efficiencies of SOFC, TRCC, and the whole system were 64.2%, 22.4%, and 74.1%, respectively. The system thermal efficiency increased with the inlet temperature of the fuel cell and decreased with the increase of the steam-carbon ratio. The thermal efficiency increased as the turbine inlet pressure to the TRCC increased.

Performance analysis for an absorption refrigeration system driven by parabolic trough solar collector
Tongtong WANG,Jianan SUN,Tao ZHANG,Zeting YU,Jiqiang YIN
Journal of Shandong University(Engineering Science). 2019, 49(5):  58-63, 71.  doi:10.6040/j.issn.1672-3961.0.2019.155
Abstract ( 521 )   HTML ( 10 )   PDF (1534KB) ( 156 )
Figures and Tables | References | Related Articles | Metrics

In order to make rational use of solar energy, enhance the seasonal adaptability of the refrigeration system, a medium temperature solar driven ammonia water absorption refrigeration system was proposed. Based on the parabolic trough solar collector (PTSC) driven ammonia single-effect absorption refrigeration system, according to the laws of thermodynamics and the energy balance equation, under the engineering equation solver (ESS), the solar collector model and the refrigeration system model were established respectively, and the key parameters of the system were calculated. The effects of system high pressure, system low pressure, evaporator outlet temperature and rectifier outlet mass fraction on the system were analyzed from three aspects: refrigeration capacity, rectification heat and coefficient of performance (COP). The results showed that the cooling capacity decreased with the increase of the system low pressure; the rectification heat and COP increased with the increasing of the system low pressure; when the outlet temperature of the evaporator rised, the cooling capacity and COP increased; when the rectifier mass fraction increased from 0.977 to 0.999, the COP showed a maximum when the ammonia mass fraction was 0.992. The results provided a theoretical basis for the feasibility of the single-stage absorption refrigeration cycle driven by solar energy.

Thermodynamic characteristic analysis of power and cooling system drived by SOFC
Hanbing WANG, Xiaohui LIU, Minli TIAN, Guihua WANG, Zeting YU, Shaobo JI
Journal of Shandong University(Engineering Science). 2019, 49(5):  64-71.  doi:10.6040/j.issn.1672-3961.0.2019.156
Abstract ( 554 )   HTML ( 4 )   PDF (2854KB) ( 195 )
Figures and Tables | References | Related Articles | Metrics

In order to improve the energy utilization efficiency of a solid oxide fuel cell, a combined power and cooling system which integrated a solid oxide fuel cell, a gas turbine and an absorption chiller was proposed. A mathematical model was developed, and the thermodynamic parameters of the system were given to study the system performance under steady-state conditions. Results showed that, under the given conditions, the SOFC electrical efficiency, the total electrical efficiency of combined system and the power-cooling efficiency were 46.81%, 54.53% and 72.24%, respectively. When the SOFC inlet temperature was 620 ℃, the maximum total electrical efficiency of combined system and the maximum power-cooling efficiency were obtained 54.66% and 72.42%, respectively. When the inlet temperature of fuel cell was 600 ℃, the cooling capacity of the combined supply system was the largest.

Heat transfer performance analysis of phase change energy storage heat exchanger based on entransy theory
Xinchen ZHOU,Xuelai ZHANG,Yue CHEN,Lu LIU
Journal of Shandong University(Engineering Science). 2019, 49(5):  72-84.  doi:10.6040/j.issn.1672-3961.0.2019.151
Abstract ( 514 )   HTML ( 9 )   PDF (4434KB) ( 310 )
Figures and Tables | References | Related Articles | Metrics

Entransy transfer efficiency, entransy dissipation number and thermal resistance of heat exchanger based on entransy dissipation were applied to heat transfer performance analysis of phase change energy storage heat exchanger on the basis of successful application of entransy theory on conventional heat exchanger. The generalized entransy dissipation rate was defined to derive (instantaneous) entransy transfer efficiency of phase change energy storage heat exchanger in heat storage, heat release and total process, and heat transfer rate was determined to calculate entransy dissipation number and thermal resistance of heat exchanger based on entransy dissipation. A kind of phase change energy storage heat exchanger was selected as the object, and the temperature variation of main parts were described by theoretical analyses. The temperature variation of outlet of silicon oil and water were further simplified to derive their expression, as the basis of calculation and analyses. The results showed that the application range of entransy transfer efficiency was the widest, which was used to calculate the (instantaneous) irreversible heat loss of phase change energy storage heat exchanger in heat storage, heat release and total process. The evaluation results of entransy transfer efficiency were consistent with heat transfer performance and its instantaneous values were increased first, then unchanged, finally increased, with increasing heat storage time, and were decreased first, then unchanged, finally decreased, with increasing heat release time. The evaluation results of entransy dissipation number in heat storage and total process were consistent with that of entransy transfer efficiency. With increasing heat storage time, its instantaneous values were decreased first, then unchanged, finally decreased, while its application was limited in heat release process. The application of thermal resistance of heat exchanger based on entransy dissipation was the most limited since parts of its evaluation results were inconsistent with actual state. In heat storage and total process, the entransy transfer efficiency, entransy dissipation number and thermal resistance of heat exchanger based on entransy dissipation were nearly unchangeable when the heat storage quantity, heat release quantity and stage time in the process of heat storage and release were synchronous. The entransy transfer efficiency was increased, while the entransy dissipation number and the thermal resistance of heat exchanger based on entransy dissipation were decreased when the heat transfer efficiency was improved. In the heat release process, entransy transfer efficiency was unchangeable since the heat transfer performance of system was not influenced by the change of parameters setted.

Influence of performance of direct-expansion solar-assisted heat pump system at different solar radiation intensity
Maoyuan ZHANG,Xiangqiang KONG,Yimeng YANG,Xiaodong LIU
Journal of Shandong University(Engineering Science). 2019, 49(5):  85-90,97.  doi:10.6040/j.issn.1672-3961.0.2019.145
Abstract ( 472 )   HTML ( 19 )   PDF (1737KB) ( 194 )
Figures and Tables | References | Related Articles | Metrics

To analyze the effect of solar radiation intensity on direct-expansion solar-assisted heat pump system (DX-SAHP), a direct-expansion solar-assisted heat pump water heater (DX-SAHPWH) system using variable frequency control and R134a was introduced and studied in this research. By carrying out the experiments under typical winter meteorologic conditions, the system performance and critical parameters were discussed. It revealed that the increasing solar radiation intensity made a higher superheat degree and coefficient of performance (COP), while the discharge temperature decreased with it. At similar ambient temperature, when the solar radiation intensity increased from 233 W·m-2 to 631 W·m-2, the system COP varied from 3.39 to 4.31, increased by 28.9%. Meanwhile, the solar radiation intensity had an opposite effect on the solar efficiency, which meant the higher solar radiation intensity was, the smaller collector efficiency was. Based on collector area of 2.10 m2, the solar collector efficiency decreased from 2.17 to 0.9 with the rising solar radiation intensity from 233 W·m-2 to 631 W·m-2, decreased by 58.5%.

Machine Learning & Data Mining
Pedestrian recognition based on singular value decomposition pedestrian alignment network
Ji ZHANG,Cui JIN,Hongyuan WANG,Shoubing CHEN
Journal of Shandong University(Engineering Science). 2019, 49(5):  91-97.  doi:10.6040/j.issn.1672-3961.0.2018.347
Abstract ( 589 )   HTML ( 8 )   PDF (1244KB) ( 203 )
Figures and Tables | References | Related Articles | Metrics

In order to solve the problem that the background of pedestrian image was too large and the part of pedestrian was missing in the training data set of pedestrian recognition, the spatial transformation network layer was used to process the image dislocation. In order to optimize the deep learning process of the whole network and improve the image retrieval ability, a feature layer was added to the network, and singular value vector decomposition was used to process it. By combining the pedestrian alignment network with the singular vector decomposition and constructing the singular value decomposition pedestrian alignment network, the image dislocation problem could be solved and the effect of similarity measurement of image features could be improved. Experiments were conducted on Market1501, CUHK03 and DukeMTMC-reID datasets, and compared with pedestrian alignment network and other pedestrian re-recognition methods of deep learning and non-deep learning. In the experimental results, the values of rank-1 and mAP mean average precision reached 80% and 65% on average, which indicated that singular value decomposition of pedestrian alignment network had certain benefits on pedestrian matching effect.

Object detection of 3D point clouds based on F-PointNet
Peng WAN
Journal of Shandong University(Engineering Science). 2019, 49(5):  98-104.  doi:10.6040/j.issn.1672-3961.0.2018.348
Abstract ( 1964 )   HTML ( 40 )   PDF (3128KB) ( 731 )
Figures and Tables | References | Related Articles | Metrics

Aiming at the problem of poor detection accuracy of the current 3D point cloud object detection model, the F-PointNet model, which directly processed point cloud data, was used to detect cars, pedestrians and cyclists, and the model was fine-tuned to further improve the object detection accuracy. The model was tested by different parameter initialization methods, $\ell$2 regularization and modifying convolution kernels. The experimental results showed that the Xavier parameter initialization method converged faster 0.09s than the truncated normal distribution method, and the vehicle detection accuracy and the cyclists detection accuracy was about 3% and 2% higher respectively. By adding $\ell$2 regularization, the detection accuracy of pedestrians and cyclists was increased by about 2% and 1% respectively. By reducing the number of convolution kernels in the first layer of T-Net (Transformer Networks) to 128, the detection accuracy of cars and cyclists was increased by about 1% and 2% respectively, which confirmed that the model could effectively improve object detection accuracy.

An improved algorithm of maximal information coefficient based on dynamic equipartition
Yanxia MENG,Yuchen GUO,Li WANG
Journal of Shandong University(Engineering Science). 2019, 49(5):  105-111.  doi:10.6040/j.issn.1672-3961.0.2018.209
Abstract ( 792 )   HTML ( 6 )   PDF (2337KB) ( 226 )
Figures and Tables | References | Related Articles | Metrics

In order to solve the problem of high computational time complexity of the maximal information coefficient algorithm, an improved algorithm of the maximal information coefficient(MIC) based on dynamic equipartition was presented. The scattered points shown in the grid were iterated and optimized by using dynamic mean division pairs of variables. The obtained mutual information entropy was regularized to obtain the optimal MIC value, and the multi-thread computation of the data set was carried out by using the POSIX parallel strategy, which made the computation more efficient in the computation of large data sets. Compared with the RapidMIC method on multiple data sets, the DE-MIC algorithm was faster and more efficient under the premise of preserving the generality and equitability of the maximal information coefficient algorithm.

Lightweight self-adaptive CSI-based positioning algorithm in underground mine
Junmei YUE,Dongmei ZHANG
Journal of Shandong University(Engineering Science). 2019, 49(5):  112-118.  doi:10.6040/j.issn.1672-3961.0.2018.356
Abstract ( 546 )   HTML ( 3 )   PDF (2758KB) ( 273 )
Figures and Tables | References | Related Articles | Metrics

To solve the problem of high cost and working hazard factor of traditional downhole positioning methods, a lightweight self-adaptive CSI-based positioning algorithm in underground mine was proposed. The fine-grained CSI was used to obtain higher positioning accuracy rather than coarse-grained RSSI, inverse fast Fourier transform was adopted to transform CSI data to channel impulse response so as to get the line-of-sight signal, an attenuation model of line-of-sight signal of CSI was built to implement accurate ranging, position features of existing point access points (APs) in wireless fidelity and characteristics of rock roadways was utilized to calculate orientation of target relative to AP, which finally completed location according to orientation and distance. LSA was adaptive to arbitrary deployment modes, and the corner recognition optimization algorithm was used to improve positioning accuracy. The experimental results showed that LSA method median error could reach 0.53 m and eliminate the need to deploy any positioning system in the well alone, the performance was superrior to CDPF and FILA.

A force-based method for robot hole-searching and assembly
Yujun WU,Wei WU,Yu GUO,Jian GUO
Journal of Shandong University(Engineering Science). 2019, 49(5):  119-126.  doi:10.6040/j.issn.1672-3961.0.2018.396
Abstract ( 711 )   HTML ( 12 )   PDF (7332KB) ( 309 )
Figures and Tables | References | Related Articles | Metrics

In the process of replacing arresters for live working robots, a method based on force control for robot to search hole on insulated cross-arms and assembling arresters was proposed for accurate assembly. In the control process, a force sensor was installed at the end of the robot for the special shape of the arrester's lower end and the scene of the pole cross arm, and the converted force was used as the controlled amount. A peg-in-hole assembly strategy was designed, which included four processes of touching, hunting, evolving and inserting. The robot could insert the arrester's lower end into the mounting hole on the cross arm accurately. The experimental results verified the effectiveness of the proposed method.