Research On Calculation Model, Condition Assessment, And Prediction For On-line Icing-monitoring System Of Overhead Transmission Lines

Icing and snowing on overhead transmission lines in extremely cold climatic events is anenormous threat to the stability and safety of power system. Some problems caused by icing,such as line galloping, insulator flashover and tower collapse, etc., occurred in many countries.Because of reliability and practicability of the weight method, it is widely used for an on-linemonitoring and pre-warning system for ice coating on overhead transmission lines, which is agood solution to guarantee the safe operation of power grid.Although an electric power enterprise specification have been initially formed focusedon the principles and the monitoring devices of the on-line monitoring system, some keyproblems have not been solved yet, for example, the mechanical calculation model is needdeep studying, and current researches are lack of condition assessment, and warning ofoverhead transmission line under icing and snowing condition. So the research on themechanical calculation model, condition assessment and forecast of icing is of importantacademic significance, and is the key technology that needs to be resolved in transmissionline on-line monitoring application.Based by the National Basic Research Program of China (973program), theoreticalanalysis and modeling, artificial intelligence techniques, true line tests in laboratory and fieldtests are combined to carry out the project. The main works in this paper are as follows:The reasonableness of statics mechanical-calculation model using weighing method isproved. Take the insulator string as rigid body, and its acceleration equation is derived basedon theoretical mechanics principle, to prove that the conductor for the wind swing conditionscan be processed as static. The paper researches the deformation characteristics of typicalconductors in an isolated span, by adding the external loads which include wind loads and iceloads to the conductors. For accurate calculation on overhead transmission line icing, a newmechanical calculation model is established based on suspended point pull of insulator stringat tower and obliquities of insulator string, considering the deviation angel of insulator stringfor wind. In this model, mechanics parameters of conductor are turned into the newcoordinates after deviation angel of insulator string for wind, and a statics force balanceequation is built in vertical direction of the new coordinates after deviation. This model is also applicable to the no-wind case. Furthermore, aiming at tower basic parameter errors cases, theparameter accuracy test and parameter correction methods are proposed. And a novelcomputed method for the ranges of icing cover, in which the vertical load of conductors isfirst calculated and to acquire the ranges of the length of conductors by fitting with Gaussiandistribution based on weather constraints, is proposed for the cases of lack of tower basicparameter. The on-line monitoring data and field ice data obtained by people are used toverify the effectiveness and feasibility of the above two ice computed methods.The simulation experiment of ice cover on overhead transmission line is carried out atthe National Engineering Laboratory for UHV Engineering Technology (Kunming). Thenatural ice is simulated by lumped loads hanging on the line, the equivalent ice thickness ofwhich is5mm. Linearity of the tension and angle sensor and the measurement errors are bothobtained. Uniformed and non-uniformed ice cases of the line are tested and analyzed torealize the calculation accuracy and the range of application of the methods proposed in thispaper.The paper researches the correlation between local meteorology parameters and theoverhead transmission line icing. The correlation is not comprehensive and systemic currently,and the main influence factor on icing cannot be confirmed. In this paper, result unconformitybetween line correlation and the qualitative analysis about the correlation is pointed out, andthe synthetic grey relational analysis, which combines general relational degree proposed byDeng with slope relational degree, is constructed to realize the relationships amongtransmission line icing, conductor temperature and local meteorology parameters, such asambient temperature, relative humidity, wind speed, etc.. Finally the order of the relationdegree is given.The paper presents two ways of condition assessment of overhead transmission line icing.Firstly, a method is proposed to improve the accuracy of condition assessment based onmultivariable fuzzy logic control, considering equivalent thickness of overhead transmissionline mainly, local meteorology parameters and ice cover duration. A hierarchical multitudesstructure of multivariable fuzzy logic controller and a set of fuzzy reasoning rules are set up.Second, a fuzzy expert system which is characteristic of fuzziness and indefiniteness isproposed, for analyzing the potential for damage to transmission line and tower as a result of icing events. The characteristic quantities for condition assessment are decided to establishtwo kinds of models: static mechanical load and external insulation assessment model. Andthe condition assessment knowledge base corresponding to each assessment model is builtbased on test results, field data and expert knowledge. Furthermore, field data from ChinaSouthern Power Grid on-line monitoring system are used to test the effectiveness and theaccuracy of the proposed methods.The paper also presents a short-term prediction method for icing based on time seriesanalysis. The theory of wavelet analysis is applied to eliminate the noise the real-time icequality of the unit length, and the time series analysis method, autoregressive integratedmoving average (ARIMA) model, is used to predict ice quality in12h. The numericalexample on the field data is showed to testify the proposed model and a comparison of theproposed model and the rime growth model is investigated.